diff --git a/demo/envpool_demo_ant.ipynb b/demo/envpool_demo_ant.ipynb
new file mode 100644
index 00000000..6210b7d7
--- /dev/null
+++ b/demo/envpool_demo_ant.ipynb
@@ -0,0 +1,2784 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "source": [
+ "[![](\"https://colab.research.google.com/assets/colab-badge.svg\")
](https://colab.research.google.com/github/sail-sg/envpool/blob/main/demo/envpool-demo-ant.ipynb)"
+ ],
+ "metadata": {
+ "id": "9jbtt6I1SyrM"
+ }
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "# OS: Ubuntu 18.04.4 LTS x86_64\n",
+ "# Kernel: 4.18.0-15-generic\n",
+ "# CPU: Intel(R) Core(TM) i9-10920X CPU (24) @ 3.50GHz\n",
+ "# GPU: NVIDIA GeForce RTX 2080 Ti\n",
+ "!sudo apt-get install -y \\\n",
+ " libgl1-mesa-dev \\\n",
+ " libgl1-mesa-glx \\\n",
+ " libglew-dev \\\n",
+ " libosmesa6-dev \\\n",
+ " software-properties-common\n",
+ "\n",
+ "!sudo apt-get install -y patchelf\n",
+ "!pip install git+https://github.com/Denys88/rl_games\n",
+ "!pip install envpool\n",
+ "!pip install gym\n",
+ "\n",
+ "!pip install free-mujoco-py\n",
+ "!sudo apt-get install -y xvfb python-opengl ffmpeg > /dev/null 2>&1\n",
+ "!pip install imageio==2.4.1\n",
+ "!pip install -U colabgymrender"
+ ],
+ "metadata": {
+ "id": "mmKGtYq6N7kz"
+ },
+ "execution_count": null,
+ "outputs": []
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "id": "q7w9yaqQ0Hb_"
+ },
+ "outputs": [],
+ "source": [
+ "from rl_games.torch_runner import Runner"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "id": "yAkZ1aqe5ur2"
+ },
+ "outputs": [],
+ "source": [
+ "## ant_envpool config:\n",
+ "ant_config = {'params': {'seed': 5, \n",
+ " 'algo': {'name': 'a2c_continuous'}, \n",
+ " 'model': {'name': 'continuous_a2c_logstd'}, \n",
+ " 'network': {'name': 'actor_critic', \n",
+ " 'separate': False, \n",
+ " 'space': {'continuous': {'mu_activation': 'None', \n",
+ " 'sigma_activation': 'None', \n",
+ " 'mu_init': {'name': 'default'}, \n",
+ " 'sigma_init': {'name': 'const_initializer', 'val': 0}, \n",
+ " 'fixed_sigma': True}}, \n",
+ " 'mlp': {'units': [256, 128, 64], 'activation': 'elu', 'initializer': {'name': 'default'}}}, \n",
+ " 'config': {'name': 'Ant-v4_envpool', \n",
+ " 'env_name': 'envpool', \n",
+ " 'score_to_win': 20000, \n",
+ " 'normalize_input': True, \n",
+ " 'normalize_value': True, \n",
+ " 'value_bootstrap': True, \n",
+ " 'normalize_advantage': True, \n",
+ " 'reward_shaper': {'scale_value': 0.1}, \n",
+ " 'gamma': 0.99, \n",
+ " 'tau': 0.95, \n",
+ " 'learning_rate': '3e-4', \n",
+ " 'lr_schedule': 'adaptive', \n",
+ " 'kl_threshold': 0.008, \n",
+ " 'grad_norm': 1.0, \n",
+ " 'entropy_coef': 0.0, \n",
+ " 'truncate_grads': True, \n",
+ " 'e_clip': 0.2, \n",
+ " 'clip_value': True, \n",
+ " 'use_smooth_clamp': True, \n",
+ " 'bound_loss_type': 'regularisation', \n",
+ " 'bounds_loss_coef': 0.0, \n",
+ " 'max_epochs': 2000, \n",
+ " 'num_actors': 64, \n",
+ " 'horizon_length': 64, \n",
+ " 'minibatch_size': 2048, \n",
+ " 'mini_epochs': 4, \n",
+ " 'critic_coef': 2, \n",
+ " 'env_config': {'env_name': 'Ant-v4', 'seed': 5}, \n",
+ " 'player': {'render': False}}}}"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "id": "dX0TgQa4DpYo",
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "outputId": "8dbd42eb-b9a7-4e19-93df-fdf314644518"
+ },
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "self.seed = 5\n",
+ "Started to train\n",
+ "current training device: cuda:0\n",
+ "build mlp: 27\n",
+ "RunningMeanStd: (1,)\n",
+ "RunningMeanStd: (27,)\n",
+ "fps step: 16584 fps step and policy inference: 9207 fps total: 7682 epoch: 1/2000\n",
+ "fps step: 17417 fps step and policy inference: 11287 fps total: 9875 epoch: 2/2000\n",
+ "fps step: 17829 fps step and policy inference: 11602 fps total: 10138 epoch: 3/2000\n",
+ "fps step: 17163 fps step and policy inference: 10351 fps total: 8915 epoch: 4/2000\n",
+ "fps step: 14484 fps step and policy inference: 7770 fps total: 6902 epoch: 5/2000\n",
+ "fps step: 16067 fps step and policy inference: 9708 fps total: 8599 epoch: 6/2000\n",
+ "fps step: 16852 fps step and policy inference: 10780 fps total: 9566 epoch: 7/2000\n",
+ "fps step: 17624 fps step and policy inference: 12090 fps total: 10535 epoch: 8/2000\n",
+ "fps step: 14597 fps step and policy inference: 8048 fps total: 7216 epoch: 9/2000\n",
+ "fps step: 17085 fps step and policy inference: 11054 fps total: 9701 epoch: 10/2000\n",
+ "fps step: 17730 fps step and policy inference: 12098 fps total: 10302 epoch: 11/2000\n",
+ "fps step: 17610 fps step and policy inference: 12318 fps total: 10725 epoch: 12/2000\n",
+ "fps step: 16700 fps step and policy inference: 10695 fps total: 9351 epoch: 13/2000\n",
+ "fps step: 17078 fps step and policy inference: 11524 fps total: 10097 epoch: 14/2000\n",
+ "fps step: 16603 fps step and policy inference: 10549 fps total: 9093 epoch: 15/2000\n",
+ "fps step: 15401 fps step and policy inference: 9074 fps total: 7888 epoch: 16/2000\n",
+ "fps step: 15758 fps step and policy inference: 9424 fps total: 8227 epoch: 17/2000\n",
+ "fps step: 16189 fps step and policy inference: 9887 fps total: 8947 epoch: 18/2000\n",
+ "fps step: 17430 fps step and policy inference: 11223 fps total: 9648 epoch: 19/2000\n",
+ "fps step: 15064 fps step and policy inference: 8632 fps total: 8155 epoch: 20/2000\n",
+ "fps step: 15134 fps step and policy inference: 8371 fps total: 7384 epoch: 21/2000\n",
+ "fps step: 15828 fps step and policy inference: 9021 fps total: 7958 epoch: 22/2000\n",
+ "fps step: 15467 fps step and policy inference: 8874 fps total: 7998 epoch: 23/2000\n",
+ "fps step: 16909 fps step and policy inference: 10925 fps total: 9530 epoch: 24/2000\n",
+ "fps step: 16617 fps step and policy inference: 10496 fps total: 9322 epoch: 25/2000\n",
+ "fps step: 16862 fps step and policy inference: 10738 fps total: 9499 epoch: 26/2000\n",
+ "fps step: 15662 fps step and policy inference: 9005 fps total: 7876 epoch: 27/2000\n",
+ "fps step: 16974 fps step and policy inference: 10731 fps total: 9119 epoch: 28/2000\n",
+ "fps step: 15722 fps step and policy inference: 8982 fps total: 7866 epoch: 29/2000\n",
+ "fps step: 17397 fps step and policy inference: 11343 fps total: 10011 epoch: 30/2000\n",
+ "fps step: 15175 fps step and policy inference: 8345 fps total: 7350 epoch: 31/2000\n",
+ "fps step: 16790 fps step and policy inference: 10384 fps total: 8927 epoch: 32/2000\n",
+ "fps step: 15939 fps step and policy inference: 9283 fps total: 8030 epoch: 33/2000\n",
+ "fps step: 18092 fps step and policy inference: 12783 fps total: 11015 epoch: 34/2000\n",
+ "fps step: 16394 fps step and policy inference: 9771 fps total: 8520 epoch: 35/2000\n",
+ "fps step: 17555 fps step and policy inference: 11446 fps total: 10063 epoch: 36/2000\n",
+ "fps step: 15260 fps step and policy inference: 8514 fps total: 7451 epoch: 37/2000\n",
+ "fps step: 15233 fps step and policy inference: 8490 fps total: 7600 epoch: 38/2000\n",
+ "fps step: 17160 fps step and policy inference: 11072 fps total: 9766 epoch: 39/2000\n",
+ "fps step: 15708 fps step and policy inference: 8908 fps total: 7772 epoch: 40/2000\n",
+ "fps step: 16046 fps step and policy inference: 9360 fps total: 8121 epoch: 41/2000\n",
+ "fps step: 17407 fps step and policy inference: 11394 fps total: 10001 epoch: 42/2000\n",
+ "fps step: 14376 fps step and policy inference: 7594 fps total: 6756 epoch: 43/2000\n",
+ "fps step: 15417 fps step and policy inference: 8578 fps total: 7505 epoch: 44/2000\n",
+ "fps step: 15353 fps step and policy inference: 8805 fps total: 7753 epoch: 45/2000\n",
+ "fps step: 14308 fps step and policy inference: 7837 fps total: 6838 epoch: 46/2000\n",
+ "fps step: 15123 fps step and policy inference: 8369 fps total: 7380 epoch: 47/2000\n",
+ "fps step: 15339 fps step and policy inference: 8539 fps total: 7513 epoch: 48/2000\n",
+ "fps step: 15655 fps step and policy inference: 9144 fps total: 8240 epoch: 49/2000\n",
+ "fps step: 17108 fps step and policy inference: 10876 fps total: 9518 epoch: 50/2000\n",
+ "fps step: 14887 fps step and policy inference: 8207 fps total: 7420 epoch: 51/2000\n",
+ "fps step: 14916 fps step and policy inference: 8053 fps total: 7121 epoch: 52/2000\n",
+ "fps step: 17207 fps step and policy inference: 10924 fps total: 9371 epoch: 53/2000\n",
+ "fps step: 16389 fps step and policy inference: 9715 fps total: 8385 epoch: 54/2000\n",
+ "fps step: 14662 fps step and policy inference: 7922 fps total: 7000 epoch: 55/2000\n",
+ "fps step: 18025 fps step and policy inference: 12784 fps total: 11017 epoch: 56/2000\n",
+ "fps step: 17546 fps step and policy inference: 11706 fps total: 10254 epoch: 57/2000\n",
+ "fps step: 17447 fps step and policy inference: 11163 fps total: 9562 epoch: 58/2000\n",
+ "fps step: 15199 fps step and policy inference: 8366 fps total: 7346 epoch: 59/2000\n",
+ "fps step: 16155 fps step and policy inference: 9516 fps total: 8331 epoch: 60/2000\n",
+ "fps step: 16061 fps step and policy inference: 9868 fps total: 8798 epoch: 61/2000\n",
+ "fps step: 14684 fps step and policy inference: 7821 fps total: 6901 epoch: 62/2000\n",
+ "fps step: 17228 fps step and policy inference: 11076 fps total: 9802 epoch: 63/2000\n",
+ "fps step: 17411 fps step and policy inference: 11213 fps total: 9768 epoch: 64/2000\n",
+ "fps step: 16035 fps step and policy inference: 9442 fps total: 8170 epoch: 65/2000\n",
+ "fps step: 16965 fps step and policy inference: 10890 fps total: 9539 epoch: 66/2000\n",
+ "fps step: 16666 fps step and policy inference: 10278 fps total: 9043 epoch: 67/2000\n",
+ "fps step: 16686 fps step and policy inference: 10395 fps total: 8991 epoch: 68/2000\n",
+ "fps step: 17736 fps step and policy inference: 12111 fps total: 10237 epoch: 69/2000\n",
+ "fps step: 16352 fps step and policy inference: 10203 fps total: 9016 epoch: 70/2000\n",
+ "fps step: 15508 fps step and policy inference: 9083 fps total: 8133 epoch: 71/2000\n",
+ "fps step: 15272 fps step and policy inference: 8776 fps total: 7645 epoch: 72/2000\n",
+ "fps step: 17251 fps step and policy inference: 11041 fps total: 9394 epoch: 73/2000\n",
+ "fps step: 15239 fps step and policy inference: 8757 fps total: 7678 epoch: 74/2000\n",
+ "fps step: 15422 fps step and policy inference: 8985 fps total: 7890 epoch: 75/2000\n",
+ "fps step: 15350 fps step and policy inference: 8815 fps total: 7707 epoch: 76/2000\n",
+ "fps step: 15782 fps step and policy inference: 9467 fps total: 8328 epoch: 77/2000\n",
+ "fps step: 17438 fps step and policy inference: 11647 fps total: 10733 epoch: 78/2000\n",
+ "fps step: 18431 fps step and policy inference: 12882 fps total: 10836 epoch: 79/2000\n",
+ "fps step: 15643 fps step and policy inference: 9590 fps total: 8561 epoch: 80/2000\n",
+ "fps step: 16583 fps step and policy inference: 10565 fps total: 9360 epoch: 81/2000\n",
+ "fps step: 17004 fps step and policy inference: 10811 fps total: 9157 epoch: 82/2000\n",
+ "fps step: 15616 fps step and policy inference: 9498 fps total: 8468 epoch: 83/2000\n",
+ "fps step: 15559 fps step and policy inference: 9228 fps total: 8050 epoch: 84/2000\n",
+ "fps step: 15292 fps step and policy inference: 8982 fps total: 7924 epoch: 85/2000\n",
+ "fps step: 15738 fps step and policy inference: 9655 fps total: 8526 epoch: 86/2000\n",
+ "fps step: 15721 fps step and policy inference: 9409 fps total: 8222 epoch: 87/2000\n",
+ "fps step: 16927 fps step and policy inference: 11023 fps total: 9681 epoch: 88/2000\n",
+ "fps step: 14596 fps step and policy inference: 8204 fps total: 7230 epoch: 89/2000\n",
+ "fps step: 15843 fps step and policy inference: 9833 fps total: 8752 epoch: 90/2000\n",
+ "fps step: 16603 fps step and policy inference: 10250 fps total: 8649 epoch: 91/2000\n",
+ "fps step: 16422 fps step and policy inference: 10120 fps total: 8770 epoch: 92/2000\n",
+ "fps step: 16041 fps step and policy inference: 9784 fps total: 8398 epoch: 93/2000\n",
+ "fps step: 17086 fps step and policy inference: 11566 fps total: 10109 epoch: 94/2000\n",
+ "fps step: 15476 fps step and policy inference: 9457 fps total: 8457 epoch: 95/2000\n",
+ "fps step: 14710 fps step and policy inference: 8619 fps total: 7702 epoch: 96/2000\n",
+ "fps step: 14177 fps step and policy inference: 7891 fps total: 6976 epoch: 97/2000\n",
+ "fps step: 17543 fps step and policy inference: 11764 fps total: 10261 epoch: 98/2000\n",
+ "fps step: 15372 fps step and policy inference: 9046 fps total: 7877 epoch: 99/2000\n",
+ "fps step: 16427 fps step and policy inference: 10642 fps total: 9342 epoch: 100/2000\n",
+ "saving next best rewards: [126.037155]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14944 fps step and policy inference: 8655 fps total: 7618 epoch: 101/2000\n",
+ "saving next best rewards: [128.81575]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17998 fps step and policy inference: 12778 fps total: 10960 epoch: 102/2000\n",
+ "fps step: 14714 fps step and policy inference: 8481 fps total: 7600 epoch: 103/2000\n",
+ "fps step: 15455 fps step and policy inference: 9109 fps total: 7974 epoch: 104/2000\n",
+ "saving next best rewards: [137.70943]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15011 fps step and policy inference: 8732 fps total: 7639 epoch: 105/2000\n",
+ "saving next best rewards: [159.41953]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16488 fps step and policy inference: 10629 fps total: 9430 epoch: 106/2000\n",
+ "saving next best rewards: [162.80412]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17527 fps step and policy inference: 12248 fps total: 10555 epoch: 107/2000\n",
+ "saving next best rewards: [165.10896]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15376 fps step and policy inference: 9199 fps total: 8053 epoch: 108/2000\n",
+ "saving next best rewards: [166.06831]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14502 fps step and policy inference: 8176 fps total: 7206 epoch: 109/2000\n",
+ "saving next best rewards: [170.75818]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15490 fps step and policy inference: 9236 fps total: 8040 epoch: 110/2000\n",
+ "saving next best rewards: [177.24358]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14919 fps step and policy inference: 8583 fps total: 7521 epoch: 111/2000\n",
+ "saving next best rewards: [184.79018]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16685 fps step and policy inference: 10767 fps total: 9437 epoch: 112/2000\n",
+ "saving next best rewards: [202.53056]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15225 fps step and policy inference: 9203 fps total: 8136 epoch: 113/2000\n",
+ "saving next best rewards: [205.42586]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16775 fps step and policy inference: 10954 fps total: 9613 epoch: 114/2000\n",
+ "saving next best rewards: [218.95233]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15013 fps step and policy inference: 8739 fps total: 7652 epoch: 115/2000\n",
+ "saving next best rewards: [225.18484]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17788 fps step and policy inference: 12405 fps total: 10828 epoch: 116/2000\n",
+ "fps step: 14824 fps step and policy inference: 8448 fps total: 7421 epoch: 117/2000\n",
+ "saving next best rewards: [237.20009]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16780 fps step and policy inference: 11126 fps total: 9782 epoch: 118/2000\n",
+ "saving next best rewards: [249.55316]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14975 fps step and policy inference: 8708 fps total: 7773 epoch: 119/2000\n",
+ "saving next best rewards: [254.78987]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15352 fps step and policy inference: 9300 fps total: 8338 epoch: 120/2000\n",
+ "saving next best rewards: [258.34442]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14323 fps step and policy inference: 7983 fps total: 7043 epoch: 121/2000\n",
+ "saving next best rewards: [283.79758]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15252 fps step and policy inference: 8970 fps total: 7867 epoch: 122/2000\n",
+ "saving next best rewards: [288.6324]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15348 fps step and policy inference: 8955 fps total: 7817 epoch: 123/2000\n",
+ "saving next best rewards: [291.47186]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17416 fps step and policy inference: 11628 fps total: 9973 epoch: 124/2000\n",
+ "saving next best rewards: [307.29645]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15772 fps step and policy inference: 9677 fps total: 8630 epoch: 125/2000\n",
+ "saving next best rewards: [328.10175]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14903 fps step and policy inference: 8672 fps total: 7611 epoch: 126/2000\n",
+ "saving next best rewards: [340.87946]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14666 fps step and policy inference: 8427 fps total: 7552 epoch: 127/2000\n",
+ "saving next best rewards: [347.90576]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17910 fps step and policy inference: 12515 fps total: 10495 epoch: 128/2000\n",
+ "saving next best rewards: [355.48187]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16874 fps step and policy inference: 10799 fps total: 9236 epoch: 129/2000\n",
+ "saving next best rewards: [368.41205]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16936 fps step and policy inference: 11113 fps total: 9675 epoch: 130/2000\n",
+ "saving next best rewards: [369.96533]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14646 fps step and policy inference: 8350 fps total: 7350 epoch: 131/2000\n",
+ "saving next best rewards: [395.28085]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16836 fps step and policy inference: 11104 fps total: 9767 epoch: 132/2000\n",
+ "saving next best rewards: [406.48682]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15141 fps step and policy inference: 8969 fps total: 8037 epoch: 133/2000\n",
+ "saving next best rewards: [426.57855]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16642 fps step and policy inference: 10746 fps total: 9891 epoch: 134/2000\n",
+ "saving next best rewards: [451.64807]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15814 fps step and policy inference: 9655 fps total: 8467 epoch: 135/2000\n",
+ "saving next best rewards: [465.22354]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14774 fps step and policy inference: 8481 fps total: 7520 epoch: 136/2000\n",
+ "saving next best rewards: [484.68707]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15223 fps step and policy inference: 8753 fps total: 7701 epoch: 137/2000\n",
+ "saving next best rewards: [488.5973]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17007 fps step and policy inference: 10744 fps total: 9148 epoch: 138/2000\n",
+ "fps step: 14414 fps step and policy inference: 8074 fps total: 7092 epoch: 139/2000\n",
+ "fps step: 14782 fps step and policy inference: 8339 fps total: 7335 epoch: 140/2000\n",
+ "saving next best rewards: [501.09222]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17553 fps step and policy inference: 11829 fps total: 10296 epoch: 141/2000\n",
+ "saving next best rewards: [519.5827]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16128 fps step and policy inference: 10064 fps total: 8642 epoch: 142/2000\n",
+ "saving next best rewards: [530.04767]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16621 fps step and policy inference: 10555 fps total: 9030 epoch: 143/2000\n",
+ "saving next best rewards: [548.363]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14747 fps step and policy inference: 8413 fps total: 7431 epoch: 144/2000\n",
+ "saving next best rewards: [563.355]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14799 fps step and policy inference: 8303 fps total: 7305 epoch: 145/2000\n",
+ "saving next best rewards: [583.3787]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17122 fps step and policy inference: 11337 fps total: 9874 epoch: 146/2000\n",
+ "saving next best rewards: [601.0418]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 18081 fps step and policy inference: 12729 fps total: 11058 epoch: 147/2000\n",
+ "saving next best rewards: [626.8574]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16588 fps step and policy inference: 10573 fps total: 9139 epoch: 148/2000\n",
+ "saving next best rewards: [656.5153]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17361 fps step and policy inference: 11683 fps total: 10033 epoch: 149/2000\n",
+ "saving next best rewards: [672.0848]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17074 fps step and policy inference: 11443 fps total: 9985 epoch: 150/2000\n",
+ "fps step: 14869 fps step and policy inference: 8477 fps total: 7346 epoch: 151/2000\n",
+ "fps step: 14908 fps step and policy inference: 8832 fps total: 7969 epoch: 152/2000\n",
+ "fps step: 15479 fps step and policy inference: 9445 fps total: 8364 epoch: 153/2000\n",
+ "fps step: 14990 fps step and policy inference: 8665 fps total: 7588 epoch: 154/2000\n",
+ "saving next best rewards: [697.0124]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15641 fps step and policy inference: 9384 fps total: 8164 epoch: 155/2000\n",
+ "saving next best rewards: [718.1951]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15648 fps step and policy inference: 9476 fps total: 8467 epoch: 156/2000\n",
+ "saving next best rewards: [744.3185]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15368 fps step and policy inference: 9098 fps total: 7949 epoch: 157/2000\n",
+ "saving next best rewards: [746.1306]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15995 fps step and policy inference: 9666 fps total: 8330 epoch: 158/2000\n",
+ "saving next best rewards: [770.0716]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17330 fps step and policy inference: 11514 fps total: 10128 epoch: 159/2000\n",
+ "saving next best rewards: [796.5363]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15157 fps step and policy inference: 8587 fps total: 7529 epoch: 160/2000\n",
+ "saving next best rewards: [813.1471]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15749 fps step and policy inference: 9477 fps total: 8068 epoch: 161/2000\n",
+ "fps step: 16112 fps step and policy inference: 9797 fps total: 8690 epoch: 162/2000\n",
+ "fps step: 16404 fps step and policy inference: 10071 fps total: 8680 epoch: 163/2000\n",
+ "saving next best rewards: [822.22485]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17464 fps step and policy inference: 12154 fps total: 10499 epoch: 164/2000\n",
+ "saving next best rewards: [830.0704]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15978 fps step and policy inference: 9523 fps total: 8241 epoch: 165/2000\n",
+ "saving next best rewards: [837.138]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15681 fps step and policy inference: 9224 fps total: 7985 epoch: 166/2000\n",
+ "saving next best rewards: [873.1602]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16980 fps step and policy inference: 11165 fps total: 9762 epoch: 167/2000\n",
+ "saving next best rewards: [889.7333]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16798 fps step and policy inference: 10680 fps total: 9122 epoch: 168/2000\n",
+ "saving next best rewards: [895.9468]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15019 fps step and policy inference: 8592 fps total: 7581 epoch: 169/2000\n",
+ "saving next best rewards: [957.5649]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16030 fps step and policy inference: 9953 fps total: 8905 epoch: 170/2000\n",
+ "saving next best rewards: [958.708]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15199 fps step and policy inference: 8821 fps total: 7822 epoch: 171/2000\n",
+ "saving next best rewards: [972.2806]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15781 fps step and policy inference: 9387 fps total: 8367 epoch: 172/2000\n",
+ "saving next best rewards: [1006.54315]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15853 fps step and policy inference: 9545 fps total: 8082 epoch: 173/2000\n",
+ "fps step: 15764 fps step and policy inference: 9321 fps total: 8029 epoch: 174/2000\n",
+ "fps step: 15208 fps step and policy inference: 8925 fps total: 8019 epoch: 175/2000\n",
+ "fps step: 17610 fps step and policy inference: 12062 fps total: 10537 epoch: 176/2000\n",
+ "saving next best rewards: [1024.7625]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17643 fps step and policy inference: 11642 fps total: 10191 epoch: 177/2000\n",
+ "saving next best rewards: [1026.8611]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15564 fps step and policy inference: 9250 fps total: 8261 epoch: 178/2000\n",
+ "saving next best rewards: [1076.731]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14736 fps step and policy inference: 8422 fps total: 7483 epoch: 179/2000\n",
+ "saving next best rewards: [1127.3539]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16743 fps step and policy inference: 10823 fps total: 9505 epoch: 180/2000\n",
+ "fps step: 16456 fps step and policy inference: 10313 fps total: 9135 epoch: 181/2000\n",
+ "saving next best rewards: [1145.4104]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16517 fps step and policy inference: 10307 fps total: 8842 epoch: 182/2000\n",
+ "saving next best rewards: [1184.2633]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16406 fps step and policy inference: 10094 fps total: 8627 epoch: 183/2000\n",
+ "fps step: 16158 fps step and policy inference: 9717 fps total: 8662 epoch: 184/2000\n",
+ "fps step: 17719 fps step and policy inference: 12133 fps total: 10553 epoch: 185/2000\n",
+ "fps step: 17398 fps step and policy inference: 11823 fps total: 10322 epoch: 186/2000\n",
+ "fps step: 16266 fps step and policy inference: 10026 fps total: 8873 epoch: 187/2000\n",
+ "fps step: 18564 fps step and policy inference: 13319 fps total: 11430 epoch: 188/2000\n",
+ "fps step: 16510 fps step and policy inference: 10556 fps total: 9371 epoch: 189/2000\n",
+ "fps step: 14917 fps step and policy inference: 8461 fps total: 7538 epoch: 190/2000\n",
+ "fps step: 15711 fps step and policy inference: 9401 fps total: 8420 epoch: 191/2000\n",
+ "fps step: 15080 fps step and policy inference: 8620 fps total: 7456 epoch: 192/2000\n",
+ "fps step: 15307 fps step and policy inference: 8904 fps total: 7780 epoch: 193/2000\n",
+ "saving next best rewards: [1203.1482]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 18209 fps step and policy inference: 12810 fps total: 11058 epoch: 194/2000\n",
+ "saving next best rewards: [1209.2064]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17655 fps step and policy inference: 11863 fps total: 10234 epoch: 195/2000\n",
+ "saving next best rewards: [1213.4927]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15421 fps step and policy inference: 8931 fps total: 7836 epoch: 196/2000\n",
+ "fps step: 15103 fps step and policy inference: 8645 fps total: 7619 epoch: 197/2000\n",
+ "fps step: 15496 fps step and policy inference: 9416 fps total: 8439 epoch: 198/2000\n",
+ "fps step: 15065 fps step and policy inference: 8626 fps total: 7568 epoch: 199/2000\n",
+ "fps step: 16449 fps step and policy inference: 10003 fps total: 8787 epoch: 200/2000\n",
+ "fps step: 15086 fps step and policy inference: 8677 fps total: 7598 epoch: 201/2000\n",
+ "fps step: 16415 fps step and policy inference: 10288 fps total: 8985 epoch: 202/2000\n",
+ "fps step: 15634 fps step and policy inference: 9237 fps total: 8068 epoch: 203/2000\n",
+ "fps step: 15700 fps step and policy inference: 9513 fps total: 8454 epoch: 204/2000\n",
+ "fps step: 14996 fps step and policy inference: 8706 fps total: 7742 epoch: 205/2000\n",
+ "fps step: 16329 fps step and policy inference: 9976 fps total: 8565 epoch: 206/2000\n",
+ "fps step: 17756 fps step and policy inference: 11860 fps total: 10330 epoch: 207/2000\n",
+ "fps step: 16325 fps step and policy inference: 9824 fps total: 8476 epoch: 208/2000\n",
+ "fps step: 18485 fps step and policy inference: 13050 fps total: 11284 epoch: 209/2000\n",
+ "saving next best rewards: [1220.0215]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17971 fps step and policy inference: 12383 fps total: 10775 epoch: 210/2000\n",
+ "saving next best rewards: [1229.0999]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14848 fps step and policy inference: 8533 fps total: 7528 epoch: 211/2000\n",
+ "saving next best rewards: [1248.6155]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16979 fps step and policy inference: 10838 fps total: 9290 epoch: 212/2000\n",
+ "saving next best rewards: [1276.8545]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15853 fps step and policy inference: 9455 fps total: 8315 epoch: 213/2000\n",
+ "saving next best rewards: [1288.7316]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16340 fps step and policy inference: 10144 fps total: 8727 epoch: 214/2000\n",
+ "saving next best rewards: [1312.2596]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15422 fps step and policy inference: 8914 fps total: 7835 epoch: 215/2000\n",
+ "saving next best rewards: [1390.1454]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15584 fps step and policy inference: 9153 fps total: 7891 epoch: 216/2000\n",
+ "saving next best rewards: [1413.1088]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15572 fps step and policy inference: 9496 fps total: 8512 epoch: 217/2000\n",
+ "saving next best rewards: [1426.7623]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 18165 fps step and policy inference: 12419 fps total: 10469 epoch: 218/2000\n",
+ "saving next best rewards: [1463.4314]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15665 fps step and policy inference: 9376 fps total: 8158 epoch: 219/2000\n",
+ "saving next best rewards: [1479.604]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17159 fps step and policy inference: 11126 fps total: 9566 epoch: 220/2000\n",
+ "fps step: 14728 fps step and policy inference: 8520 fps total: 7657 epoch: 221/2000\n",
+ "saving next best rewards: [1506.7329]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17678 fps step and policy inference: 12012 fps total: 10330 epoch: 222/2000\n",
+ "fps step: 15335 fps step and policy inference: 9068 fps total: 8200 epoch: 223/2000\n",
+ "saving next best rewards: [1515.0823]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15691 fps step and policy inference: 9451 fps total: 8416 epoch: 224/2000\n",
+ "fps step: 15993 fps step and policy inference: 9970 fps total: 8919 epoch: 225/2000\n",
+ "fps step: 15640 fps step and policy inference: 9461 fps total: 8101 epoch: 226/2000\n",
+ "saving next best rewards: [1525.7914]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15035 fps step and policy inference: 8493 fps total: 7484 epoch: 227/2000\n",
+ "saving next best rewards: [1568.9957]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 18110 fps step and policy inference: 12569 fps total: 10967 epoch: 228/2000\n",
+ "saving next best rewards: [1623.5573]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17569 fps step and policy inference: 11324 fps total: 9629 epoch: 229/2000\n",
+ "fps step: 16089 fps step and policy inference: 9970 fps total: 8836 epoch: 230/2000\n",
+ "fps step: 16784 fps step and policy inference: 10577 fps total: 9299 epoch: 231/2000\n",
+ "saving next best rewards: [1628.4644]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16859 fps step and policy inference: 10672 fps total: 9424 epoch: 232/2000\n",
+ "saving next best rewards: [1654.514]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17133 fps step and policy inference: 11496 fps total: 10127 epoch: 233/2000\n",
+ "saving next best rewards: [1721.9598]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15421 fps step and policy inference: 8947 fps total: 7864 epoch: 234/2000\n",
+ "fps step: 18656 fps step and policy inference: 13617 fps total: 11722 epoch: 235/2000\n",
+ "fps step: 16888 fps step and policy inference: 10665 fps total: 9161 epoch: 236/2000\n",
+ "fps step: 15468 fps step and policy inference: 8902 fps total: 7678 epoch: 237/2000\n",
+ "fps step: 15293 fps step and policy inference: 8849 fps total: 7918 epoch: 238/2000\n",
+ "fps step: 15320 fps step and policy inference: 8941 fps total: 8044 epoch: 239/2000\n",
+ "fps step: 17150 fps step and policy inference: 11024 fps total: 9379 epoch: 240/2000\n",
+ "fps step: 15237 fps step and policy inference: 8702 fps total: 7638 epoch: 241/2000\n",
+ "fps step: 15490 fps step and policy inference: 9180 fps total: 8128 epoch: 242/2000\n",
+ "fps step: 14733 fps step and policy inference: 8308 fps total: 7368 epoch: 243/2000\n",
+ "fps step: 16137 fps step and policy inference: 9893 fps total: 8729 epoch: 244/2000\n",
+ "fps step: 17254 fps step and policy inference: 11135 fps total: 9643 epoch: 245/2000\n",
+ "fps step: 14966 fps step and policy inference: 8664 fps total: 7793 epoch: 246/2000\n",
+ "fps step: 14446 fps step and policy inference: 8102 fps total: 7139 epoch: 247/2000\n",
+ "fps step: 15604 fps step and policy inference: 9092 fps total: 7891 epoch: 248/2000\n",
+ "fps step: 16292 fps step and policy inference: 9970 fps total: 8629 epoch: 249/2000\n",
+ "fps step: 15687 fps step and policy inference: 9307 fps total: 8078 epoch: 250/2000\n",
+ "fps step: 16109 fps step and policy inference: 10066 fps total: 8915 epoch: 251/2000\n",
+ "fps step: 15230 fps step and policy inference: 8928 fps total: 7990 epoch: 252/2000\n",
+ "fps step: 16507 fps step and policy inference: 10253 fps total: 8889 epoch: 253/2000\n",
+ "saving next best rewards: [1738.7327]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16351 fps step and policy inference: 10275 fps total: 8928 epoch: 254/2000\n",
+ "saving next best rewards: [1759.3364]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14433 fps step and policy inference: 8120 fps total: 7101 epoch: 255/2000\n",
+ "saving next best rewards: [1776.6987]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16807 fps step and policy inference: 10813 fps total: 9590 epoch: 256/2000\n",
+ "saving next best rewards: [1808.6443]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15431 fps step and policy inference: 9181 fps total: 8190 epoch: 257/2000\n",
+ "saving next best rewards: [1832.2311]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16461 fps step and policy inference: 10196 fps total: 8737 epoch: 258/2000\n",
+ "saving next best rewards: [1882.7411]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16424 fps step and policy inference: 10083 fps total: 8652 epoch: 259/2000\n",
+ "saving next best rewards: [1892.3798]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16068 fps step and policy inference: 9705 fps total: 8979 epoch: 260/2000\n",
+ "fps step: 15575 fps step and policy inference: 9102 fps total: 8041 epoch: 261/2000\n",
+ "fps step: 16630 fps step and policy inference: 10290 fps total: 8869 epoch: 262/2000\n",
+ "fps step: 16943 fps step and policy inference: 11023 fps total: 9662 epoch: 263/2000\n",
+ "fps step: 16167 fps step and policy inference: 9757 fps total: 8456 epoch: 264/2000\n",
+ "saving next best rewards: [1911.8248]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16033 fps step and policy inference: 10002 fps total: 8938 epoch: 265/2000\n",
+ "saving next best rewards: [1937.9648]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15552 fps step and policy inference: 9325 fps total: 8343 epoch: 266/2000\n",
+ "saving next best rewards: [1945.3624]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16134 fps step and policy inference: 9641 fps total: 8310 epoch: 267/2000\n",
+ "fps step: 17076 fps step and policy inference: 10933 fps total: 9311 epoch: 268/2000\n",
+ "saving next best rewards: [1965.4565]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14958 fps step and policy inference: 8688 fps total: 7575 epoch: 269/2000\n",
+ "saving next best rewards: [1971.577]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14739 fps step and policy inference: 8434 fps total: 7430 epoch: 270/2000\n",
+ "saving next best rewards: [2010.0405]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15816 fps step and policy inference: 9572 fps total: 8482 epoch: 271/2000\n",
+ "saving next best rewards: [2037.0898]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17295 fps step and policy inference: 11890 fps total: 10348 epoch: 272/2000\n",
+ "saving next best rewards: [2070.2766]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15057 fps step and policy inference: 8699 fps total: 7789 epoch: 273/2000\n",
+ "fps step: 14878 fps step and policy inference: 8668 fps total: 7720 epoch: 274/2000\n",
+ "saving next best rewards: [2106.2983]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15148 fps step and policy inference: 9008 fps total: 8121 epoch: 275/2000\n",
+ "saving next best rewards: [2122.076]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16802 fps step and policy inference: 10791 fps total: 9529 epoch: 276/2000\n",
+ "fps step: 15786 fps step and policy inference: 9386 fps total: 8177 epoch: 277/2000\n",
+ "saving next best rewards: [2172.5286]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15678 fps step and policy inference: 9384 fps total: 8377 epoch: 278/2000\n",
+ "fps step: 16703 fps step and policy inference: 10693 fps total: 9458 epoch: 279/2000\n",
+ "saving next best rewards: [2197.3655]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15974 fps step and policy inference: 9485 fps total: 8243 epoch: 280/2000\n",
+ "saving next best rewards: [2226.7915]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15222 fps step and policy inference: 8815 fps total: 7737 epoch: 281/2000\n",
+ "fps step: 15537 fps step and policy inference: 9143 fps total: 8160 epoch: 282/2000\n",
+ "saving next best rewards: [2239.9841]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15120 fps step and policy inference: 8619 fps total: 7657 epoch: 283/2000\n",
+ "fps step: 14812 fps step and policy inference: 8394 fps total: 7422 epoch: 284/2000\n",
+ "fps step: 15744 fps step and policy inference: 9508 fps total: 8491 epoch: 285/2000\n",
+ "fps step: 15804 fps step and policy inference: 9720 fps total: 8418 epoch: 286/2000\n",
+ "fps step: 17646 fps step and policy inference: 11968 fps total: 10503 epoch: 287/2000\n",
+ "fps step: 17819 fps step and policy inference: 11935 fps total: 10478 epoch: 288/2000\n",
+ "fps step: 18669 fps step and policy inference: 13642 fps total: 11734 epoch: 289/2000\n",
+ "saving next best rewards: [2241.748]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16691 fps step and policy inference: 10648 fps total: 9303 epoch: 290/2000\n",
+ "saving next best rewards: [2278.5708]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17065 fps step and policy inference: 11245 fps total: 9892 epoch: 291/2000\n",
+ "saving next best rewards: [2323.8113]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16667 fps step and policy inference: 10673 fps total: 9418 epoch: 292/2000\n",
+ "saving next best rewards: [2356.768]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15931 fps step and policy inference: 9387 fps total: 8159 epoch: 293/2000\n",
+ "fps step: 14962 fps step and policy inference: 8475 fps total: 7408 epoch: 294/2000\n",
+ "fps step: 16721 fps step and policy inference: 10444 fps total: 8934 epoch: 295/2000\n",
+ "saving next best rewards: [2387.5625]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15853 fps step and policy inference: 9376 fps total: 8170 epoch: 296/2000\n",
+ "saving next best rewards: [2422.5986]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17123 fps step and policy inference: 11224 fps total: 9644 epoch: 297/2000\n",
+ "saving next best rewards: [2449.5688]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16648 fps step and policy inference: 10608 fps total: 9348 epoch: 298/2000\n",
+ "fps step: 14730 fps step and policy inference: 8457 fps total: 7419 epoch: 299/2000\n",
+ "saving next best rewards: [2465.5952]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15537 fps step and policy inference: 9135 fps total: 8544 epoch: 300/2000\n",
+ "saving next best rewards: [2514.0618]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15508 fps step and policy inference: 9266 fps total: 8277 epoch: 301/2000\n",
+ "fps step: 15638 fps step and policy inference: 9332 fps total: 8100 epoch: 302/2000\n",
+ "fps step: 16581 fps step and policy inference: 10464 fps total: 9042 epoch: 303/2000\n",
+ "fps step: 16964 fps step and policy inference: 10988 fps total: 9720 epoch: 304/2000\n",
+ "fps step: 14881 fps step and policy inference: 8466 fps total: 7497 epoch: 305/2000\n",
+ "saving next best rewards: [2549.4163]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17955 fps step and policy inference: 12727 fps total: 11013 epoch: 306/2000\n",
+ "fps step: 16496 fps step and policy inference: 10078 fps total: 8646 epoch: 307/2000\n",
+ "fps step: 16863 fps step and policy inference: 11171 fps total: 9802 epoch: 308/2000\n",
+ "saving next best rewards: [2558.5708]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17906 fps step and policy inference: 11833 fps total: 9895 epoch: 309/2000\n",
+ "saving next best rewards: [2604.417]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16091 fps step and policy inference: 9832 fps total: 8526 epoch: 310/2000\n",
+ "fps step: 16745 fps step and policy inference: 10558 fps total: 9036 epoch: 311/2000\n",
+ "fps step: 15919 fps step and policy inference: 9602 fps total: 8345 epoch: 312/2000\n",
+ "fps step: 16209 fps step and policy inference: 9900 fps total: 8387 epoch: 313/2000\n",
+ "fps step: 14794 fps step and policy inference: 8376 fps total: 7399 epoch: 314/2000\n",
+ "fps step: 16670 fps step and policy inference: 10629 fps total: 9365 epoch: 315/2000\n",
+ "fps step: 14620 fps step and policy inference: 8172 fps total: 7131 epoch: 316/2000\n",
+ "fps step: 15534 fps step and policy inference: 9414 fps total: 8386 epoch: 317/2000\n",
+ "fps step: 15672 fps step and policy inference: 9520 fps total: 8528 epoch: 318/2000\n",
+ "fps step: 17836 fps step and policy inference: 12347 fps total: 10737 epoch: 319/2000\n",
+ "fps step: 14842 fps step and policy inference: 8320 fps total: 7321 epoch: 320/2000\n",
+ "fps step: 15713 fps step and policy inference: 9451 fps total: 8434 epoch: 321/2000\n",
+ "fps step: 17419 fps step and policy inference: 11434 fps total: 9718 epoch: 322/2000\n",
+ "saving next best rewards: [2619.7734]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15219 fps step and policy inference: 8833 fps total: 7770 epoch: 323/2000\n",
+ "saving next best rewards: [2647.863]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16627 fps step and policy inference: 10602 fps total: 9345 epoch: 324/2000\n",
+ "saving next best rewards: [2725.2847]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16994 fps step and policy inference: 11188 fps total: 9816 epoch: 325/2000\n",
+ "saving next best rewards: [2775.505]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 18357 fps step and policy inference: 12654 fps total: 10646 epoch: 326/2000\n",
+ "fps step: 14964 fps step and policy inference: 8428 fps total: 7361 epoch: 327/2000\n",
+ "saving next best rewards: [2789.134]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16506 fps step and policy inference: 10302 fps total: 8939 epoch: 328/2000\n",
+ "saving next best rewards: [2796.324]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16672 fps step and policy inference: 10640 fps total: 9420 epoch: 329/2000\n",
+ "saving next best rewards: [2801.1006]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17606 fps step and policy inference: 11975 fps total: 10442 epoch: 330/2000\n",
+ "fps step: 16174 fps step and policy inference: 9468 fps total: 8183 epoch: 331/2000\n",
+ "saving next best rewards: [2818.5413]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15763 fps step and policy inference: 9611 fps total: 8528 epoch: 332/2000\n",
+ "fps step: 15962 fps step and policy inference: 9575 fps total: 8340 epoch: 333/2000\n",
+ "fps step: 15544 fps step and policy inference: 9166 fps total: 8005 epoch: 334/2000\n",
+ "fps step: 14764 fps step and policy inference: 8370 fps total: 7460 epoch: 335/2000\n",
+ "fps step: 17449 fps step and policy inference: 11220 fps total: 9547 epoch: 336/2000\n",
+ "fps step: 16806 fps step and policy inference: 10759 fps total: 9380 epoch: 337/2000\n",
+ "saving next best rewards: [2856.5283]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15241 fps step and policy inference: 8901 fps total: 7824 epoch: 338/2000\n",
+ "saving next best rewards: [2872.7837]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16768 fps step and policy inference: 10771 fps total: 9473 epoch: 339/2000\n",
+ "saving next best rewards: [2898.023]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15768 fps step and policy inference: 9390 fps total: 8142 epoch: 340/2000\n",
+ "saving next best rewards: [2959.8855]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16518 fps step and policy inference: 10495 fps total: 9091 epoch: 341/2000\n",
+ "saving next best rewards: [2983.4312]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15808 fps step and policy inference: 9575 fps total: 8390 epoch: 342/2000\n",
+ "saving next best rewards: [2999.978]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16200 fps step and policy inference: 10105 fps total: 8807 epoch: 343/2000\n",
+ "fps step: 15097 fps step and policy inference: 8665 fps total: 7593 epoch: 344/2000\n",
+ "saving next best rewards: [3042.1553]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15843 fps step and policy inference: 9595 fps total: 8478 epoch: 345/2000\n",
+ "saving next best rewards: [3070.0293]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15349 fps step and policy inference: 9044 fps total: 7880 epoch: 346/2000\n",
+ "saving next best rewards: [3113.909]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14815 fps step and policy inference: 8432 fps total: 7448 epoch: 347/2000\n",
+ "saving next best rewards: [3171.2205]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14880 fps step and policy inference: 8507 fps total: 7398 epoch: 348/2000\n",
+ "saving next best rewards: [3185.0881]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 18196 fps step and policy inference: 12259 fps total: 10669 epoch: 349/2000\n",
+ "saving next best rewards: [3189.8655]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15765 fps step and policy inference: 9343 fps total: 8086 epoch: 350/2000\n",
+ "fps step: 15504 fps step and policy inference: 9105 fps total: 8156 epoch: 351/2000\n",
+ "fps step: 15008 fps step and policy inference: 8475 fps total: 7428 epoch: 352/2000\n",
+ "saving next best rewards: [3202.8604]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16081 fps step and policy inference: 9641 fps total: 8314 epoch: 353/2000\n",
+ "saving next best rewards: [3253.558]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15988 fps step and policy inference: 9854 fps total: 8750 epoch: 354/2000\n",
+ "fps step: 15315 fps step and policy inference: 8907 fps total: 7766 epoch: 355/2000\n",
+ "fps step: 17162 fps step and policy inference: 11101 fps total: 9662 epoch: 356/2000\n",
+ "saving next best rewards: [3272.5715]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15541 fps step and policy inference: 9394 fps total: 8367 epoch: 357/2000\n",
+ "fps step: 16968 fps step and policy inference: 10905 fps total: 9395 epoch: 358/2000\n",
+ "saving next best rewards: [3273.7588]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14913 fps step and policy inference: 8572 fps total: 7591 epoch: 359/2000\n",
+ "saving next best rewards: [3277.4644]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15925 fps step and policy inference: 9568 fps total: 8346 epoch: 360/2000\n",
+ "saving next best rewards: [3283.871]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15278 fps step and policy inference: 8750 fps total: 7682 epoch: 361/2000\n",
+ "saving next best rewards: [3296.0264]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15701 fps step and policy inference: 9591 fps total: 8437 epoch: 362/2000\n",
+ "saving next best rewards: [3306.606]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16678 fps step and policy inference: 10662 fps total: 9442 epoch: 363/2000\n",
+ "fps step: 17196 fps step and policy inference: 10997 fps total: 9428 epoch: 364/2000\n",
+ "saving next best rewards: [3336.366]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15910 fps step and policy inference: 9566 fps total: 8343 epoch: 365/2000\n",
+ "fps step: 16072 fps step and policy inference: 9979 fps total: 8913 epoch: 366/2000\n",
+ "fps step: 15150 fps step and policy inference: 8685 fps total: 7604 epoch: 367/2000\n",
+ "fps step: 17148 fps step and policy inference: 11048 fps total: 9458 epoch: 368/2000\n",
+ "fps step: 14865 fps step and policy inference: 8406 fps total: 7392 epoch: 369/2000\n",
+ "saving next best rewards: [3360.7388]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15643 fps step and policy inference: 9412 fps total: 8073 epoch: 370/2000\n",
+ "saving next best rewards: [3386.4177]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16710 fps step and policy inference: 10836 fps total: 9550 epoch: 371/2000\n",
+ "saving next best rewards: [3423.9202]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17398 fps step and policy inference: 11877 fps total: 10357 epoch: 372/2000\n",
+ "saving next best rewards: [3445.1562]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15439 fps step and policy inference: 9215 fps total: 8109 epoch: 373/2000\n",
+ "saving next best rewards: [3472.1743]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15026 fps step and policy inference: 8654 fps total: 7606 epoch: 374/2000\n",
+ "saving next best rewards: [3481.9478]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15722 fps step and policy inference: 9451 fps total: 8326 epoch: 375/2000\n",
+ "fps step: 15919 fps step and policy inference: 10065 fps total: 9012 epoch: 376/2000\n",
+ "saving next best rewards: [3501.1401]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15524 fps step and policy inference: 9016 fps total: 7824 epoch: 377/2000\n",
+ "saving next best rewards: [3528.9702]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16882 fps step and policy inference: 11087 fps total: 9737 epoch: 378/2000\n",
+ "fps step: 16457 fps step and policy inference: 10054 fps total: 8655 epoch: 379/2000\n",
+ "fps step: 14889 fps step and policy inference: 8588 fps total: 7572 epoch: 380/2000\n",
+ "fps step: 16822 fps step and policy inference: 11013 fps total: 9699 epoch: 381/2000\n",
+ "fps step: 15596 fps step and policy inference: 9221 fps total: 8111 epoch: 382/2000\n",
+ "fps step: 15363 fps step and policy inference: 8995 fps total: 7676 epoch: 383/2000\n",
+ "fps step: 18241 fps step and policy inference: 12720 fps total: 10611 epoch: 384/2000\n",
+ "fps step: 15678 fps step and policy inference: 9112 fps total: 8006 epoch: 385/2000\n",
+ "fps step: 17183 fps step and policy inference: 11318 fps total: 10003 epoch: 386/2000\n",
+ "fps step: 16410 fps step and policy inference: 10130 fps total: 9270 epoch: 387/2000\n",
+ "fps step: 17155 fps step and policy inference: 11406 fps total: 10018 epoch: 388/2000\n",
+ "fps step: 17216 fps step and policy inference: 11189 fps total: 9578 epoch: 389/2000\n",
+ "fps step: 18010 fps step and policy inference: 12038 fps total: 10177 epoch: 390/2000\n",
+ "fps step: 16154 fps step and policy inference: 10127 fps total: 8942 epoch: 391/2000\n",
+ "fps step: 15185 fps step and policy inference: 8744 fps total: 7641 epoch: 392/2000\n",
+ "saving next best rewards: [3548.6042]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17248 fps step and policy inference: 11546 fps total: 10061 epoch: 393/2000\n",
+ "fps step: 15443 fps step and policy inference: 9163 fps total: 8096 epoch: 394/2000\n",
+ "fps step: 15740 fps step and policy inference: 9231 fps total: 8002 epoch: 395/2000\n",
+ "fps step: 14694 fps step and policy inference: 8129 fps total: 7166 epoch: 396/2000\n",
+ "fps step: 16679 fps step and policy inference: 10864 fps total: 9526 epoch: 397/2000\n",
+ "fps step: 14793 fps step and policy inference: 8372 fps total: 7522 epoch: 398/2000\n",
+ "fps step: 15102 fps step and policy inference: 8700 fps total: 7677 epoch: 399/2000\n",
+ "fps step: 15175 fps step and policy inference: 8779 fps total: 7747 epoch: 400/2000\n",
+ "fps step: 16790 fps step and policy inference: 10770 fps total: 9224 epoch: 401/2000\n",
+ "fps step: 16257 fps step and policy inference: 9819 fps total: 8424 epoch: 402/2000\n",
+ "saving next best rewards: [3557.403]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15093 fps step and policy inference: 8888 fps total: 7992 epoch: 403/2000\n",
+ "saving next best rewards: [3576.7336]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15156 fps step and policy inference: 8839 fps total: 7693 epoch: 404/2000\n",
+ "fps step: 16871 fps step and policy inference: 10858 fps total: 9606 epoch: 405/2000\n",
+ "fps step: 15828 fps step and policy inference: 9624 fps total: 8596 epoch: 406/2000\n",
+ "fps step: 15698 fps step and policy inference: 9312 fps total: 8102 epoch: 407/2000\n",
+ "saving next best rewards: [3582.7524]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15960 fps step and policy inference: 9893 fps total: 8803 epoch: 408/2000\n",
+ "saving next best rewards: [3624.8218]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14678 fps step and policy inference: 8323 fps total: 7362 epoch: 409/2000\n",
+ "fps step: 15720 fps step and policy inference: 9268 fps total: 8044 epoch: 410/2000\n",
+ "fps step: 15660 fps step and policy inference: 9312 fps total: 8123 epoch: 411/2000\n",
+ "fps step: 17190 fps step and policy inference: 11480 fps total: 10115 epoch: 412/2000\n",
+ "fps step: 16172 fps step and policy inference: 9894 fps total: 8407 epoch: 413/2000\n",
+ "fps step: 15823 fps step and policy inference: 9775 fps total: 8686 epoch: 414/2000\n",
+ "fps step: 14886 fps step and policy inference: 8653 fps total: 7713 epoch: 415/2000\n",
+ "fps step: 15147 fps step and policy inference: 8769 fps total: 7831 epoch: 416/2000\n",
+ "fps step: 16171 fps step and policy inference: 9827 fps total: 8454 epoch: 417/2000\n",
+ "fps step: 16255 fps step and policy inference: 9750 fps total: 8193 epoch: 418/2000\n",
+ "fps step: 15880 fps step and policy inference: 9938 fps total: 8852 epoch: 419/2000\n",
+ "fps step: 17377 fps step and policy inference: 11780 fps total: 10108 epoch: 420/2000\n",
+ "fps step: 16610 fps step and policy inference: 10741 fps total: 9218 epoch: 421/2000\n",
+ "fps step: 16739 fps step and policy inference: 10611 fps total: 9030 epoch: 422/2000\n",
+ "fps step: 15803 fps step and policy inference: 9488 fps total: 8257 epoch: 423/2000\n",
+ "fps step: 14597 fps step and policy inference: 8300 fps total: 7309 epoch: 424/2000\n",
+ "fps step: 14665 fps step and policy inference: 8524 fps total: 7646 epoch: 425/2000\n",
+ "fps step: 14999 fps step and policy inference: 8642 fps total: 7624 epoch: 426/2000\n",
+ "saving next best rewards: [3653.9224]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15130 fps step and policy inference: 8830 fps total: 7863 epoch: 427/2000\n",
+ "saving next best rewards: [3659.1033]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16090 fps step and policy inference: 9841 fps total: 8556 epoch: 428/2000\n",
+ "saving next best rewards: [3725.174]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15620 fps step and policy inference: 9335 fps total: 8235 epoch: 429/2000\n",
+ "fps step: 15798 fps step and policy inference: 9450 fps total: 8277 epoch: 430/2000\n",
+ "saving next best rewards: [3760.7568]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15026 fps step and policy inference: 9058 fps total: 8105 epoch: 431/2000\n",
+ "saving next best rewards: [3769.1665]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15212 fps step and policy inference: 8862 fps total: 7670 epoch: 432/2000\n",
+ "fps step: 14988 fps step and policy inference: 8709 fps total: 7858 epoch: 433/2000\n",
+ "fps step: 15154 fps step and policy inference: 8577 fps total: 7535 epoch: 434/2000\n",
+ "fps step: 17061 fps step and policy inference: 11408 fps total: 9955 epoch: 435/2000\n",
+ "fps step: 15736 fps step and policy inference: 9254 fps total: 7920 epoch: 436/2000\n",
+ "saving next best rewards: [3771.255]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15398 fps step and policy inference: 9060 fps total: 7939 epoch: 437/2000\n",
+ "saving next best rewards: [3807.3992]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15440 fps step and policy inference: 9217 fps total: 8286 epoch: 438/2000\n",
+ "saving next best rewards: [3852.153]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15356 fps step and policy inference: 9081 fps total: 8013 epoch: 439/2000\n",
+ "saving next best rewards: [3903.0828]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15954 fps step and policy inference: 9503 fps total: 8270 epoch: 440/2000\n",
+ "fps step: 15192 fps step and policy inference: 8651 fps total: 7558 epoch: 441/2000\n",
+ "fps step: 16240 fps step and policy inference: 10237 fps total: 9091 epoch: 442/2000\n",
+ "fps step: 16945 fps step and policy inference: 10747 fps total: 9299 epoch: 443/2000\n",
+ "saving next best rewards: [3941.4536]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15266 fps step and policy inference: 8885 fps total: 7788 epoch: 444/2000\n",
+ "saving next best rewards: [3978.6003]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14946 fps step and policy inference: 8632 fps total: 7677 epoch: 445/2000\n",
+ "saving next best rewards: [4026.7305]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15405 fps step and policy inference: 9354 fps total: 8345 epoch: 446/2000\n",
+ "saving next best rewards: [4033.899]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15664 fps step and policy inference: 9335 fps total: 8105 epoch: 447/2000\n",
+ "fps step: 15292 fps step and policy inference: 8926 fps total: 7835 epoch: 448/2000\n",
+ "fps step: 16327 fps step and policy inference: 10360 fps total: 9184 epoch: 449/2000\n",
+ "fps step: 15859 fps step and policy inference: 9540 fps total: 8285 epoch: 450/2000\n",
+ "fps step: 15538 fps step and policy inference: 9114 fps total: 7890 epoch: 451/2000\n",
+ "fps step: 16708 fps step and policy inference: 10623 fps total: 9323 epoch: 452/2000\n",
+ "fps step: 16546 fps step and policy inference: 10456 fps total: 9063 epoch: 453/2000\n",
+ "fps step: 15604 fps step and policy inference: 9563 fps total: 8597 epoch: 454/2000\n",
+ "fps step: 16529 fps step and policy inference: 10305 fps total: 8890 epoch: 455/2000\n",
+ "fps step: 15827 fps step and policy inference: 9475 fps total: 8350 epoch: 456/2000\n",
+ "fps step: 14747 fps step and policy inference: 8296 fps total: 7695 epoch: 457/2000\n",
+ "fps step: 15373 fps step and policy inference: 8916 fps total: 7809 epoch: 458/2000\n",
+ "fps step: 15550 fps step and policy inference: 9441 fps total: 8428 epoch: 459/2000\n",
+ "fps step: 17002 fps step and policy inference: 11216 fps total: 9905 epoch: 460/2000\n",
+ "saving next best rewards: [4074.3237]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16176 fps step and policy inference: 9857 fps total: 8478 epoch: 461/2000\n",
+ "fps step: 15631 fps step and policy inference: 9311 fps total: 8701 epoch: 462/2000\n",
+ "fps step: 15072 fps step and policy inference: 8764 fps total: 7798 epoch: 463/2000\n",
+ "fps step: 15382 fps step and policy inference: 9149 fps total: 8085 epoch: 464/2000\n",
+ "fps step: 16015 fps step and policy inference: 9812 fps total: 8702 epoch: 465/2000\n",
+ "fps step: 16708 fps step and policy inference: 10743 fps total: 9880 epoch: 466/2000\n",
+ "fps step: 15665 fps step and policy inference: 9285 fps total: 8160 epoch: 467/2000\n",
+ "fps step: 15802 fps step and policy inference: 9312 fps total: 8142 epoch: 468/2000\n",
+ "fps step: 17211 fps step and policy inference: 11453 fps total: 10002 epoch: 469/2000\n",
+ "fps step: 17941 fps step and policy inference: 11991 fps total: 10125 epoch: 470/2000\n",
+ "fps step: 15932 fps step and policy inference: 9804 fps total: 8762 epoch: 471/2000\n",
+ "fps step: 16905 fps step and policy inference: 10892 fps total: 9150 epoch: 472/2000\n",
+ "fps step: 16228 fps step and policy inference: 10241 fps total: 9141 epoch: 473/2000\n",
+ "fps step: 15867 fps step and policy inference: 9494 fps total: 8212 epoch: 474/2000\n",
+ "fps step: 15692 fps step and policy inference: 9489 fps total: 8511 epoch: 475/2000\n",
+ "fps step: 15593 fps step and policy inference: 9377 fps total: 8228 epoch: 476/2000\n",
+ "fps step: 16015 fps step and policy inference: 10063 fps total: 8977 epoch: 477/2000\n",
+ "fps step: 17843 fps step and policy inference: 12400 fps total: 10834 epoch: 478/2000\n",
+ "fps step: 16510 fps step and policy inference: 10261 fps total: 8862 epoch: 479/2000\n",
+ "fps step: 15345 fps step and policy inference: 9178 fps total: 8516 epoch: 480/2000\n",
+ "fps step: 14662 fps step and policy inference: 8303 fps total: 7352 epoch: 481/2000\n",
+ "fps step: 16390 fps step and policy inference: 10528 fps total: 9365 epoch: 482/2000\n",
+ "saving next best rewards: [4080.1497]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15747 fps step and policy inference: 9508 fps total: 8291 epoch: 483/2000\n",
+ "saving next best rewards: [4100.5947]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15474 fps step and policy inference: 9104 fps total: 7827 epoch: 484/2000\n",
+ "saving next best rewards: [4118.569]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15916 fps step and policy inference: 10092 fps total: 8998 epoch: 485/2000\n",
+ "saving next best rewards: [4157.9824]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15209 fps step and policy inference: 8886 fps total: 7828 epoch: 486/2000\n",
+ "saving next best rewards: [4198.8525]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17343 fps step and policy inference: 11499 fps total: 10458 epoch: 487/2000\n",
+ "fps step: 17895 fps step and policy inference: 12405 fps total: 10727 epoch: 488/2000\n",
+ "saving next best rewards: [4199.1455]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16571 fps step and policy inference: 10723 fps total: 9476 epoch: 489/2000\n",
+ "fps step: 16920 fps step and policy inference: 11112 fps total: 9755 epoch: 490/2000\n",
+ "saving next best rewards: [4232.6445]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17691 fps step and policy inference: 11933 fps total: 10405 epoch: 491/2000\n",
+ "saving next best rewards: [4267.0063]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15492 fps step and policy inference: 9099 fps total: 7960 epoch: 492/2000\n",
+ "saving next best rewards: [4285.294]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17067 fps step and policy inference: 11385 fps total: 10004 epoch: 493/2000\n",
+ "saving next best rewards: [4298.355]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15162 fps step and policy inference: 8747 fps total: 7780 epoch: 494/2000\n",
+ "fps step: 14792 fps step and policy inference: 8297 fps total: 7294 epoch: 495/2000\n",
+ "fps step: 17232 fps step and policy inference: 11166 fps total: 9573 epoch: 496/2000\n",
+ "saving next best rewards: [4303.9116]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16013 fps step and policy inference: 9684 fps total: 8361 epoch: 497/2000\n",
+ "saving next best rewards: [4319.456]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14685 fps step and policy inference: 8296 fps total: 7206 epoch: 498/2000\n",
+ "fps step: 15656 fps step and policy inference: 9374 fps total: 8352 epoch: 499/2000\n",
+ "fps step: 15112 fps step and policy inference: 8768 fps total: 7705 epoch: 500/2000\n",
+ "fps step: 17574 fps step and policy inference: 12007 fps total: 10402 epoch: 501/2000\n",
+ "fps step: 17481 fps step and policy inference: 11847 fps total: 10392 epoch: 502/2000\n",
+ "fps step: 15463 fps step and policy inference: 9107 fps total: 7761 epoch: 503/2000\n",
+ "saving next best rewards: [4341.419]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14909 fps step and policy inference: 8796 fps total: 7835 epoch: 504/2000\n",
+ "saving next best rewards: [4355.2188]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15382 fps step and policy inference: 9023 fps total: 7717 epoch: 505/2000\n",
+ "saving next best rewards: [4384.0864]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15637 fps step and policy inference: 9495 fps total: 8260 epoch: 506/2000\n",
+ "saving next best rewards: [4410.0493]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17030 fps step and policy inference: 11217 fps total: 9917 epoch: 507/2000\n",
+ "fps step: 15333 fps step and policy inference: 8970 fps total: 7893 epoch: 508/2000\n",
+ "saving next best rewards: [4415.491]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15536 fps step and policy inference: 9280 fps total: 8249 epoch: 509/2000\n",
+ "saving next best rewards: [4431.917]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15210 fps step and policy inference: 8807 fps total: 7589 epoch: 510/2000\n",
+ "saving next best rewards: [4442.373]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17977 fps step and policy inference: 12337 fps total: 10603 epoch: 511/2000\n",
+ "saving next best rewards: [4467.1357]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15709 fps step and policy inference: 9449 fps total: 8420 epoch: 512/2000\n",
+ "fps step: 15447 fps step and policy inference: 9245 fps total: 8209 epoch: 513/2000\n",
+ "fps step: 15058 fps step and policy inference: 8851 fps total: 7912 epoch: 514/2000\n",
+ "fps step: 15191 fps step and policy inference: 8836 fps total: 7897 epoch: 515/2000\n",
+ "fps step: 15197 fps step and policy inference: 8790 fps total: 7726 epoch: 516/2000\n",
+ "fps step: 16473 fps step and policy inference: 10166 fps total: 8542 epoch: 517/2000\n",
+ "saving next best rewards: [4471.218]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17518 fps step and policy inference: 11829 fps total: 10193 epoch: 518/2000\n",
+ "saving next best rewards: [4500.652]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17118 fps step and policy inference: 10866 fps total: 9309 epoch: 519/2000\n",
+ "saving next best rewards: [4514.2017]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15187 fps step and policy inference: 8790 fps total: 7967 epoch: 520/2000\n",
+ "saving next best rewards: [4551.3545]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15773 fps step and policy inference: 9553 fps total: 8502 epoch: 521/2000\n",
+ "saving next best rewards: [4568.0815]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16515 fps step and policy inference: 10202 fps total: 8691 epoch: 522/2000\n",
+ "saving next best rewards: [4573.204]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15925 fps step and policy inference: 9451 fps total: 8304 epoch: 523/2000\n",
+ "fps step: 15006 fps step and policy inference: 8636 fps total: 7544 epoch: 524/2000\n",
+ "fps step: 15505 fps step and policy inference: 9006 fps total: 7745 epoch: 525/2000\n",
+ "fps step: 16606 fps step and policy inference: 10583 fps total: 9379 epoch: 526/2000\n",
+ "fps step: 15113 fps step and policy inference: 8674 fps total: 7627 epoch: 527/2000\n",
+ "fps step: 15545 fps step and policy inference: 9117 fps total: 7967 epoch: 528/2000\n",
+ "fps step: 17155 fps step and policy inference: 11002 fps total: 9341 epoch: 529/2000\n",
+ "fps step: 17691 fps step and policy inference: 11968 fps total: 10074 epoch: 530/2000\n",
+ "fps step: 15013 fps step and policy inference: 8496 fps total: 7509 epoch: 531/2000\n",
+ "fps step: 16788 fps step and policy inference: 10954 fps total: 9655 epoch: 532/2000\n",
+ "fps step: 14822 fps step and policy inference: 8282 fps total: 7310 epoch: 533/2000\n",
+ "fps step: 16736 fps step and policy inference: 10800 fps total: 9129 epoch: 534/2000\n",
+ "fps step: 16328 fps step and policy inference: 10086 fps total: 8737 epoch: 535/2000\n",
+ "fps step: 16842 fps step and policy inference: 10622 fps total: 8915 epoch: 536/2000\n",
+ "fps step: 15489 fps step and policy inference: 9328 fps total: 8353 epoch: 537/2000\n",
+ "fps step: 16583 fps step and policy inference: 10429 fps total: 9103 epoch: 538/2000\n",
+ "fps step: 15844 fps step and policy inference: 9692 fps total: 8523 epoch: 539/2000\n",
+ "fps step: 15036 fps step and policy inference: 8646 fps total: 7599 epoch: 540/2000\n",
+ "fps step: 14783 fps step and policy inference: 8345 fps total: 7325 epoch: 541/2000\n",
+ "fps step: 15073 fps step and policy inference: 8876 fps total: 8005 epoch: 542/2000\n",
+ "fps step: 16777 fps step and policy inference: 10918 fps total: 9591 epoch: 543/2000\n",
+ "fps step: 15407 fps step and policy inference: 9075 fps total: 7861 epoch: 544/2000\n",
+ "fps step: 14845 fps step and policy inference: 8493 fps total: 7306 epoch: 545/2000\n",
+ "fps step: 16858 fps step and policy inference: 11072 fps total: 9759 epoch: 546/2000\n",
+ "fps step: 15515 fps step and policy inference: 9396 fps total: 8423 epoch: 547/2000\n",
+ "fps step: 17312 fps step and policy inference: 11591 fps total: 10124 epoch: 548/2000\n",
+ "fps step: 14977 fps step and policy inference: 8462 fps total: 7426 epoch: 549/2000\n",
+ "fps step: 14809 fps step and policy inference: 8403 fps total: 7406 epoch: 550/2000\n",
+ "fps step: 14529 fps step and policy inference: 8239 fps total: 7347 epoch: 551/2000\n",
+ "fps step: 15815 fps step and policy inference: 9399 fps total: 8156 epoch: 552/2000\n",
+ "fps step: 15938 fps step and policy inference: 9612 fps total: 8353 epoch: 553/2000\n",
+ "fps step: 15705 fps step and policy inference: 9509 fps total: 8381 epoch: 554/2000\n",
+ "fps step: 16200 fps step and policy inference: 10106 fps total: 8940 epoch: 555/2000\n",
+ "fps step: 15652 fps step and policy inference: 9135 fps total: 7993 epoch: 556/2000\n",
+ "fps step: 16039 fps step and policy inference: 9681 fps total: 8344 epoch: 557/2000\n",
+ "fps step: 16732 fps step and policy inference: 10522 fps total: 9027 epoch: 558/2000\n",
+ "fps step: 16198 fps step and policy inference: 9969 fps total: 8594 epoch: 559/2000\n",
+ "fps step: 14687 fps step and policy inference: 8201 fps total: 7227 epoch: 560/2000\n",
+ "fps step: 15375 fps step and policy inference: 8986 fps total: 8063 epoch: 561/2000\n",
+ "fps step: 17474 fps step and policy inference: 11549 fps total: 10035 epoch: 562/2000\n",
+ "fps step: 15962 fps step and policy inference: 9617 fps total: 8396 epoch: 563/2000\n",
+ "fps step: 16043 fps step and policy inference: 10002 fps total: 8899 epoch: 564/2000\n",
+ "fps step: 15921 fps step and policy inference: 9304 fps total: 8071 epoch: 565/2000\n",
+ "fps step: 15603 fps step and policy inference: 9167 fps total: 7981 epoch: 566/2000\n",
+ "fps step: 14899 fps step and policy inference: 8650 fps total: 7799 epoch: 567/2000\n",
+ "fps step: 16729 fps step and policy inference: 10868 fps total: 9324 epoch: 568/2000\n",
+ "fps step: 16288 fps step and policy inference: 9983 fps total: 9206 epoch: 569/2000\n",
+ "fps step: 16502 fps step and policy inference: 10642 fps total: 9440 epoch: 570/2000\n",
+ "fps step: 15634 fps step and policy inference: 9258 fps total: 7984 epoch: 571/2000\n",
+ "fps step: 17364 fps step and policy inference: 11616 fps total: 10081 epoch: 572/2000\n",
+ "fps step: 15280 fps step and policy inference: 9006 fps total: 8055 epoch: 573/2000\n",
+ "saving next best rewards: [4580.79]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15661 fps step and policy inference: 9191 fps total: 8314 epoch: 574/2000\n",
+ "saving next best rewards: [4596.8584]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15874 fps step and policy inference: 9360 fps total: 8167 epoch: 575/2000\n",
+ "saving next best rewards: [4654.837]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16689 fps step and policy inference: 10490 fps total: 8957 epoch: 576/2000\n",
+ "fps step: 16544 fps step and policy inference: 10539 fps total: 9305 epoch: 577/2000\n",
+ "fps step: 15147 fps step and policy inference: 8694 fps total: 7630 epoch: 578/2000\n",
+ "saving next best rewards: [4656.8433]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14657 fps step and policy inference: 8364 fps total: 7458 epoch: 579/2000\n",
+ "fps step: 15098 fps step and policy inference: 8530 fps total: 7453 epoch: 580/2000\n",
+ "saving next best rewards: [4702.2095]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 18214 fps step and policy inference: 13061 fps total: 11321 epoch: 581/2000\n",
+ "fps step: 17035 fps step and policy inference: 10862 fps total: 9384 epoch: 582/2000\n",
+ "saving next best rewards: [4718.683]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15176 fps step and policy inference: 8916 fps total: 8027 epoch: 583/2000\n",
+ "fps step: 14596 fps step and policy inference: 8195 fps total: 7089 epoch: 584/2000\n",
+ "fps step: 15627 fps step and policy inference: 9254 fps total: 8075 epoch: 585/2000\n",
+ "fps step: 16034 fps step and policy inference: 9619 fps total: 8369 epoch: 586/2000\n",
+ "fps step: 16128 fps step and policy inference: 9850 fps total: 8613 epoch: 587/2000\n",
+ "fps step: 15916 fps step and policy inference: 9727 fps total: 8663 epoch: 588/2000\n",
+ "fps step: 16084 fps step and policy inference: 9932 fps total: 8702 epoch: 589/2000\n",
+ "fps step: 15294 fps step and policy inference: 8765 fps total: 7731 epoch: 590/2000\n",
+ "fps step: 15867 fps step and policy inference: 9554 fps total: 8222 epoch: 591/2000\n",
+ "fps step: 17473 fps step and policy inference: 11773 fps total: 9932 epoch: 592/2000\n",
+ "fps step: 15993 fps step and policy inference: 9417 fps total: 8198 epoch: 593/2000\n",
+ "fps step: 17181 fps step and policy inference: 11239 fps total: 9861 epoch: 594/2000\n",
+ "fps step: 17886 fps step and policy inference: 12078 fps total: 10341 epoch: 595/2000\n",
+ "fps step: 17873 fps step and policy inference: 11979 fps total: 10556 epoch: 596/2000\n",
+ "fps step: 15715 fps step and policy inference: 9454 fps total: 8468 epoch: 597/2000\n",
+ "fps step: 15038 fps step and policy inference: 8723 fps total: 7854 epoch: 598/2000\n",
+ "fps step: 15961 fps step and policy inference: 9407 fps total: 8401 epoch: 599/2000\n",
+ "fps step: 16262 fps step and policy inference: 10235 fps total: 9115 epoch: 600/2000\n",
+ "fps step: 15590 fps step and policy inference: 9365 fps total: 8323 epoch: 601/2000\n",
+ "fps step: 15791 fps step and policy inference: 9344 fps total: 8162 epoch: 602/2000\n",
+ "fps step: 14809 fps step and policy inference: 8497 fps total: 7671 epoch: 603/2000\n",
+ "fps step: 15620 fps step and policy inference: 9127 fps total: 7909 epoch: 604/2000\n",
+ "fps step: 16126 fps step and policy inference: 9910 fps total: 8860 epoch: 605/2000\n",
+ "fps step: 16935 fps step and policy inference: 10815 fps total: 9148 epoch: 606/2000\n",
+ "fps step: 17095 fps step and policy inference: 11197 fps total: 9722 epoch: 607/2000\n",
+ "fps step: 15158 fps step and policy inference: 8559 fps total: 7559 epoch: 608/2000\n",
+ "saving next best rewards: [4743.4717]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16223 fps step and policy inference: 9978 fps total: 8643 epoch: 609/2000\n",
+ "saving next best rewards: [4766.2305]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15236 fps step and policy inference: 8838 fps total: 7702 epoch: 610/2000\n",
+ "fps step: 17115 fps step and policy inference: 11471 fps total: 10109 epoch: 611/2000\n",
+ "fps step: 16188 fps step and policy inference: 10220 fps total: 9077 epoch: 612/2000\n",
+ "saving next best rewards: [4769.135]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14987 fps step and policy inference: 8514 fps total: 7441 epoch: 613/2000\n",
+ "saving next best rewards: [4810.786]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15577 fps step and policy inference: 9349 fps total: 8384 epoch: 614/2000\n",
+ "fps step: 16320 fps step and policy inference: 10008 fps total: 8600 epoch: 615/2000\n",
+ "fps step: 16976 fps step and policy inference: 10909 fps total: 9288 epoch: 616/2000\n",
+ "fps step: 15079 fps step and policy inference: 8932 fps total: 8435 epoch: 617/2000\n",
+ "fps step: 15480 fps step and policy inference: 9212 fps total: 8215 epoch: 618/2000\n",
+ "fps step: 15303 fps step and policy inference: 8982 fps total: 7888 epoch: 619/2000\n",
+ "fps step: 16441 fps step and policy inference: 10502 fps total: 9333 epoch: 620/2000\n",
+ "fps step: 17017 fps step and policy inference: 11251 fps total: 9811 epoch: 621/2000\n",
+ "fps step: 14806 fps step and policy inference: 8371 fps total: 7396 epoch: 622/2000\n",
+ "fps step: 15730 fps step and policy inference: 9540 fps total: 8467 epoch: 623/2000\n",
+ "fps step: 15061 fps step and policy inference: 8749 fps total: 7643 epoch: 624/2000\n",
+ "fps step: 18344 fps step and policy inference: 13221 fps total: 11448 epoch: 625/2000\n",
+ "fps step: 16271 fps step and policy inference: 10429 fps total: 9287 epoch: 626/2000\n",
+ "fps step: 14878 fps step and policy inference: 8399 fps total: 7441 epoch: 627/2000\n",
+ "fps step: 15809 fps step and policy inference: 9676 fps total: 8633 epoch: 628/2000\n",
+ "fps step: 15691 fps step and policy inference: 9508 fps total: 8483 epoch: 629/2000\n",
+ "fps step: 15465 fps step and policy inference: 9020 fps total: 7895 epoch: 630/2000\n",
+ "saving next best rewards: [4815.254]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14919 fps step and policy inference: 8478 fps total: 7452 epoch: 631/2000\n",
+ "saving next best rewards: [4838.0874]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17102 fps step and policy inference: 10961 fps total: 9385 epoch: 632/2000\n",
+ "fps step: 15358 fps step and policy inference: 9091 fps total: 8180 epoch: 633/2000\n",
+ "fps step: 14930 fps step and policy inference: 8385 fps total: 7394 epoch: 634/2000\n",
+ "fps step: 16534 fps step and policy inference: 10442 fps total: 9069 epoch: 635/2000\n",
+ "fps step: 17408 fps step and policy inference: 11311 fps total: 9652 epoch: 636/2000\n",
+ "fps step: 15413 fps step and policy inference: 9121 fps total: 8024 epoch: 637/2000\n",
+ "saving next best rewards: [4842.34]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17132 fps step and policy inference: 11072 fps total: 9538 epoch: 638/2000\n",
+ "saving next best rewards: [4867.749]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16527 fps step and policy inference: 10178 fps total: 8687 epoch: 639/2000\n",
+ "fps step: 14917 fps step and policy inference: 8438 fps total: 7483 epoch: 640/2000\n",
+ "saving next best rewards: [4876.5176]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15232 fps step and policy inference: 8719 fps total: 7640 epoch: 641/2000\n",
+ "saving next best rewards: [4877.2056]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16157 fps step and policy inference: 9841 fps total: 8597 epoch: 642/2000\n",
+ "saving next best rewards: [4899.936]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14515 fps step and policy inference: 8027 fps total: 7071 epoch: 643/2000\n",
+ "fps step: 15820 fps step and policy inference: 9577 fps total: 8574 epoch: 644/2000\n",
+ "fps step: 15650 fps step and policy inference: 9243 fps total: 8081 epoch: 645/2000\n",
+ "fps step: 16473 fps step and policy inference: 9992 fps total: 8607 epoch: 646/2000\n",
+ "fps step: 16486 fps step and policy inference: 10049 fps total: 8645 epoch: 647/2000\n",
+ "fps step: 16312 fps step and policy inference: 9838 fps total: 8543 epoch: 648/2000\n",
+ "fps step: 18100 fps step and policy inference: 12170 fps total: 10284 epoch: 649/2000\n",
+ "saving next best rewards: [4919.885]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15876 fps step and policy inference: 9730 fps total: 8696 epoch: 650/2000\n",
+ "fps step: 15932 fps step and policy inference: 9631 fps total: 8573 epoch: 651/2000\n",
+ "fps step: 18215 fps step and policy inference: 12428 fps total: 10533 epoch: 652/2000\n",
+ "fps step: 17706 fps step and policy inference: 11468 fps total: 9682 epoch: 653/2000\n",
+ "saving next best rewards: [4931.9404]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16137 fps step and policy inference: 9704 fps total: 8398 epoch: 654/2000\n",
+ "saving next best rewards: [4952.803]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15535 fps step and policy inference: 9148 fps total: 8108 epoch: 655/2000\n",
+ "saving next best rewards: [4990.472]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15641 fps step and policy inference: 9534 fps total: 8499 epoch: 656/2000\n",
+ "saving next best rewards: [4998.367]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17432 fps step and policy inference: 11903 fps total: 10364 epoch: 657/2000\n",
+ "saving next best rewards: [5010.534]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15622 fps step and policy inference: 9254 fps total: 8215 epoch: 658/2000\n",
+ "fps step: 15387 fps step and policy inference: 8910 fps total: 7812 epoch: 659/2000\n",
+ "fps step: 17526 fps step and policy inference: 11734 fps total: 10320 epoch: 660/2000\n",
+ "fps step: 15723 fps step and policy inference: 9508 fps total: 8440 epoch: 661/2000\n",
+ "fps step: 15934 fps step and policy inference: 9716 fps total: 8661 epoch: 662/2000\n",
+ "fps step: 17168 fps step and policy inference: 10864 fps total: 9293 epoch: 663/2000\n",
+ "fps step: 16404 fps step and policy inference: 9907 fps total: 8610 epoch: 664/2000\n",
+ "fps step: 14832 fps step and policy inference: 8386 fps total: 7350 epoch: 665/2000\n",
+ "fps step: 16815 fps step and policy inference: 10958 fps total: 9692 epoch: 666/2000\n",
+ "fps step: 17084 fps step and policy inference: 11218 fps total: 9734 epoch: 667/2000\n",
+ "fps step: 16210 fps step and policy inference: 9801 fps total: 8394 epoch: 668/2000\n",
+ "fps step: 16928 fps step and policy inference: 10691 fps total: 8920 epoch: 669/2000\n",
+ "fps step: 15605 fps step and policy inference: 9171 fps total: 7954 epoch: 670/2000\n",
+ "fps step: 16496 fps step and policy inference: 10288 fps total: 9049 epoch: 671/2000\n",
+ "fps step: 17609 fps step and policy inference: 11707 fps total: 9991 epoch: 672/2000\n",
+ "fps step: 14932 fps step and policy inference: 8448 fps total: 7453 epoch: 673/2000\n",
+ "fps step: 15005 fps step and policy inference: 8528 fps total: 7591 epoch: 674/2000\n",
+ "fps step: 16780 fps step and policy inference: 10507 fps total: 9273 epoch: 675/2000\n",
+ "fps step: 15867 fps step and policy inference: 9675 fps total: 8654 epoch: 676/2000\n",
+ "fps step: 16236 fps step and policy inference: 10096 fps total: 8951 epoch: 677/2000\n",
+ "fps step: 16247 fps step and policy inference: 10007 fps total: 8780 epoch: 678/2000\n",
+ "fps step: 16116 fps step and policy inference: 9839 fps total: 9104 epoch: 679/2000\n",
+ "fps step: 15690 fps step and policy inference: 9147 fps total: 7920 epoch: 680/2000\n",
+ "fps step: 16398 fps step and policy inference: 9966 fps total: 8624 epoch: 681/2000\n",
+ "fps step: 17927 fps step and policy inference: 12021 fps total: 10055 epoch: 682/2000\n",
+ "fps step: 15106 fps step and policy inference: 8738 fps total: 7770 epoch: 683/2000\n",
+ "fps step: 16131 fps step and policy inference: 9998 fps total: 8893 epoch: 684/2000\n",
+ "fps step: 16353 fps step and policy inference: 9970 fps total: 8557 epoch: 685/2000\n",
+ "fps step: 14862 fps step and policy inference: 8424 fps total: 7428 epoch: 686/2000\n",
+ "fps step: 17085 fps step and policy inference: 10683 fps total: 9122 epoch: 687/2000\n",
+ "fps step: 15318 fps step and policy inference: 9051 fps total: 8401 epoch: 688/2000\n",
+ "fps step: 15408 fps step and policy inference: 8833 fps total: 7729 epoch: 689/2000\n",
+ "fps step: 15584 fps step and policy inference: 9006 fps total: 7931 epoch: 690/2000\n",
+ "fps step: 15603 fps step and policy inference: 9091 fps total: 7894 epoch: 691/2000\n",
+ "fps step: 15523 fps step and policy inference: 8980 fps total: 7893 epoch: 692/2000\n",
+ "fps step: 17630 fps step and policy inference: 11404 fps total: 9822 epoch: 693/2000\n",
+ "fps step: 15044 fps step and policy inference: 8609 fps total: 7605 epoch: 694/2000\n",
+ "fps step: 17428 fps step and policy inference: 11288 fps total: 9630 epoch: 695/2000\n",
+ "fps step: 16673 fps step and policy inference: 10448 fps total: 9230 epoch: 696/2000\n",
+ "fps step: 16451 fps step and policy inference: 10104 fps total: 8936 epoch: 697/2000\n",
+ "fps step: 17631 fps step and policy inference: 11455 fps total: 9686 epoch: 698/2000\n",
+ "fps step: 16360 fps step and policy inference: 10376 fps total: 9134 epoch: 699/2000\n",
+ "fps step: 15138 fps step and policy inference: 8619 fps total: 7567 epoch: 700/2000\n",
+ "fps step: 15378 fps step and policy inference: 8901 fps total: 7741 epoch: 701/2000\n",
+ "fps step: 16578 fps step and policy inference: 10408 fps total: 9134 epoch: 702/2000\n",
+ "fps step: 16024 fps step and policy inference: 9522 fps total: 8259 epoch: 703/2000\n",
+ "fps step: 16240 fps step and policy inference: 9752 fps total: 8418 epoch: 704/2000\n",
+ "fps step: 15953 fps step and policy inference: 9722 fps total: 8605 epoch: 705/2000\n",
+ "fps step: 15929 fps step and policy inference: 9590 fps total: 8270 epoch: 706/2000\n",
+ "fps step: 15482 fps step and policy inference: 9188 fps total: 8032 epoch: 707/2000\n",
+ "fps step: 15616 fps step and policy inference: 9321 fps total: 8081 epoch: 708/2000\n",
+ "fps step: 16307 fps step and policy inference: 10179 fps total: 9025 epoch: 709/2000\n",
+ "fps step: 17287 fps step and policy inference: 11435 fps total: 10066 epoch: 710/2000\n",
+ "fps step: 15193 fps step and policy inference: 8681 fps total: 7718 epoch: 711/2000\n",
+ "fps step: 16899 fps step and policy inference: 10777 fps total: 9482 epoch: 712/2000\n",
+ "fps step: 14679 fps step and policy inference: 8274 fps total: 7407 epoch: 713/2000\n",
+ "fps step: 17804 fps step and policy inference: 12404 fps total: 10828 epoch: 714/2000\n",
+ "fps step: 16076 fps step and policy inference: 9749 fps total: 8629 epoch: 715/2000\n",
+ "fps step: 15597 fps step and policy inference: 9337 fps total: 8369 epoch: 716/2000\n",
+ "fps step: 16305 fps step and policy inference: 10114 fps total: 8906 epoch: 717/2000\n",
+ "fps step: 14898 fps step and policy inference: 8400 fps total: 7401 epoch: 718/2000\n",
+ "fps step: 14753 fps step and policy inference: 8239 fps total: 7320 epoch: 719/2000\n",
+ "fps step: 17394 fps step and policy inference: 11528 fps total: 10101 epoch: 720/2000\n",
+ "fps step: 17230 fps step and policy inference: 11546 fps total: 10150 epoch: 721/2000\n",
+ "fps step: 18449 fps step and policy inference: 13468 fps total: 11542 epoch: 722/2000\n",
+ "fps step: 15152 fps step and policy inference: 8819 fps total: 7722 epoch: 723/2000\n",
+ "fps step: 15619 fps step and policy inference: 9394 fps total: 8446 epoch: 724/2000\n",
+ "fps step: 16980 fps step and policy inference: 10688 fps total: 9150 epoch: 725/2000\n",
+ "fps step: 16176 fps step and policy inference: 9962 fps total: 8849 epoch: 726/2000\n",
+ "fps step: 15861 fps step and policy inference: 9459 fps total: 8236 epoch: 727/2000\n",
+ "fps step: 14963 fps step and policy inference: 8530 fps total: 7618 epoch: 728/2000\n",
+ "saving next best rewards: [5024.9204]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14904 fps step and policy inference: 8473 fps total: 7504 epoch: 729/2000\n",
+ "saving next best rewards: [5031.1436]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15172 fps step and policy inference: 8813 fps total: 7819 epoch: 730/2000\n",
+ "saving next best rewards: [5043.6094]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16254 fps step and policy inference: 10144 fps total: 9005 epoch: 731/2000\n",
+ "saving next best rewards: [5075.0566]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17663 fps step and policy inference: 11800 fps total: 10337 epoch: 732/2000\n",
+ "fps step: 15911 fps step and policy inference: 9446 fps total: 8182 epoch: 733/2000\n",
+ "fps step: 16054 fps step and policy inference: 9823 fps total: 8720 epoch: 734/2000\n",
+ "fps step: 16607 fps step and policy inference: 10664 fps total: 9403 epoch: 735/2000\n",
+ "fps step: 15179 fps step and policy inference: 8875 fps total: 7964 epoch: 736/2000\n",
+ "fps step: 17783 fps step and policy inference: 12085 fps total: 10552 epoch: 737/2000\n",
+ "fps step: 16379 fps step and policy inference: 9965 fps total: 8599 epoch: 738/2000\n",
+ "fps step: 15897 fps step and policy inference: 9498 fps total: 8106 epoch: 739/2000\n",
+ "fps step: 15127 fps step and policy inference: 8699 fps total: 7777 epoch: 740/2000\n",
+ "fps step: 16322 fps step and policy inference: 10316 fps total: 9159 epoch: 741/2000\n",
+ "fps step: 15382 fps step and policy inference: 8975 fps total: 7866 epoch: 742/2000\n",
+ "fps step: 14675 fps step and policy inference: 8150 fps total: 7188 epoch: 743/2000\n",
+ "fps step: 15433 fps step and policy inference: 8900 fps total: 7912 epoch: 744/2000\n",
+ "fps step: 14865 fps step and policy inference: 8340 fps total: 7371 epoch: 745/2000\n",
+ "fps step: 15923 fps step and policy inference: 9455 fps total: 8224 epoch: 746/2000\n",
+ "fps step: 16098 fps step and policy inference: 9910 fps total: 9298 epoch: 747/2000\n",
+ "fps step: 17338 fps step and policy inference: 11145 fps total: 9464 epoch: 748/2000\n",
+ "fps step: 16609 fps step and policy inference: 10137 fps total: 8819 epoch: 749/2000\n",
+ "fps step: 16271 fps step and policy inference: 10178 fps total: 9027 epoch: 750/2000\n",
+ "fps step: 16439 fps step and policy inference: 10557 fps total: 9354 epoch: 751/2000\n",
+ "fps step: 15456 fps step and policy inference: 8986 fps total: 8011 epoch: 752/2000\n",
+ "fps step: 16634 fps step and policy inference: 10354 fps total: 9022 epoch: 753/2000\n",
+ "fps step: 15351 fps step and policy inference: 8914 fps total: 7802 epoch: 754/2000\n",
+ "fps step: 17049 fps step and policy inference: 11069 fps total: 9688 epoch: 755/2000\n",
+ "fps step: 17755 fps step and policy inference: 12280 fps total: 10641 epoch: 756/2000\n",
+ "fps step: 17902 fps step and policy inference: 11971 fps total: 10107 epoch: 757/2000\n",
+ "fps step: 15809 fps step and policy inference: 9440 fps total: 8393 epoch: 758/2000\n",
+ "fps step: 15200 fps step and policy inference: 8632 fps total: 7513 epoch: 759/2000\n",
+ "fps step: 16616 fps step and policy inference: 10441 fps total: 8943 epoch: 760/2000\n",
+ "fps step: 15276 fps step and policy inference: 8779 fps total: 7679 epoch: 761/2000\n",
+ "fps step: 16158 fps step and policy inference: 9906 fps total: 8556 epoch: 762/2000\n",
+ "fps step: 15894 fps step and policy inference: 9561 fps total: 8272 epoch: 763/2000\n",
+ "fps step: 17172 fps step and policy inference: 11033 fps total: 9392 epoch: 764/2000\n",
+ "fps step: 14887 fps step and policy inference: 8383 fps total: 7369 epoch: 765/2000\n",
+ "fps step: 15153 fps step and policy inference: 8708 fps total: 7560 epoch: 766/2000\n",
+ "fps step: 15655 fps step and policy inference: 9158 fps total: 7873 epoch: 767/2000\n",
+ "fps step: 15997 fps step and policy inference: 9936 fps total: 8853 epoch: 768/2000\n",
+ "fps step: 14853 fps step and policy inference: 8390 fps total: 7380 epoch: 769/2000\n",
+ "fps step: 15643 fps step and policy inference: 9199 fps total: 7966 epoch: 770/2000\n",
+ "fps step: 16792 fps step and policy inference: 10994 fps total: 9758 epoch: 771/2000\n",
+ "fps step: 15908 fps step and policy inference: 9695 fps total: 8474 epoch: 772/2000\n",
+ "fps step: 15997 fps step and policy inference: 9767 fps total: 9170 epoch: 773/2000\n",
+ "fps step: 18827 fps step and policy inference: 13543 fps total: 11745 epoch: 774/2000\n",
+ "fps step: 16067 fps step and policy inference: 9812 fps total: 8709 epoch: 775/2000\n",
+ "fps step: 15105 fps step and policy inference: 8782 fps total: 7867 epoch: 776/2000\n",
+ "fps step: 15735 fps step and policy inference: 9299 fps total: 8117 epoch: 777/2000\n",
+ "fps step: 15977 fps step and policy inference: 9619 fps total: 8249 epoch: 778/2000\n",
+ "fps step: 15509 fps step and policy inference: 9219 fps total: 8024 epoch: 779/2000\n",
+ "fps step: 15626 fps step and policy inference: 9155 fps total: 7928 epoch: 780/2000\n",
+ "fps step: 16704 fps step and policy inference: 10561 fps total: 9380 epoch: 781/2000\n",
+ "fps step: 15841 fps step and policy inference: 9642 fps total: 8496 epoch: 782/2000\n",
+ "fps step: 16782 fps step and policy inference: 10626 fps total: 9140 epoch: 783/2000\n",
+ "fps step: 15746 fps step and policy inference: 9255 fps total: 8080 epoch: 784/2000\n",
+ "fps step: 14587 fps step and policy inference: 8148 fps total: 7271 epoch: 785/2000\n",
+ "fps step: 16452 fps step and policy inference: 10328 fps total: 9233 epoch: 786/2000\n",
+ "fps step: 17181 fps step and policy inference: 10892 fps total: 9489 epoch: 787/2000\n",
+ "fps step: 15912 fps step and policy inference: 9528 fps total: 8127 epoch: 788/2000\n",
+ "fps step: 17285 fps step and policy inference: 11283 fps total: 9897 epoch: 789/2000\n",
+ "fps step: 17674 fps step and policy inference: 11480 fps total: 9753 epoch: 790/2000\n",
+ "saving next best rewards: [5076.8877]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16891 fps step and policy inference: 10618 fps total: 9026 epoch: 791/2000\n",
+ "fps step: 16203 fps step and policy inference: 9907 fps total: 8666 epoch: 792/2000\n",
+ "saving next best rewards: [5095.516]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15784 fps step and policy inference: 9240 fps total: 8020 epoch: 793/2000\n",
+ "saving next best rewards: [5111.2656]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14792 fps step and policy inference: 8340 fps total: 7396 epoch: 794/2000\n",
+ "saving next best rewards: [5122.7754]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15176 fps step and policy inference: 8713 fps total: 7828 epoch: 795/2000\n",
+ "fps step: 17975 fps step and policy inference: 12405 fps total: 10449 epoch: 796/2000\n",
+ "fps step: 16421 fps step and policy inference: 9953 fps total: 8631 epoch: 797/2000\n",
+ "fps step: 15219 fps step and policy inference: 8795 fps total: 7869 epoch: 798/2000\n",
+ "fps step: 16988 fps step and policy inference: 10912 fps total: 9656 epoch: 799/2000\n",
+ "saving next best rewards: [5124.113]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16073 fps step and policy inference: 9673 fps total: 8521 epoch: 800/2000\n",
+ "saving next best rewards: [5155.3237]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15472 fps step and policy inference: 9054 fps total: 8008 epoch: 801/2000\n",
+ "saving next best rewards: [5160.4844]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15478 fps step and policy inference: 8936 fps total: 7861 epoch: 802/2000\n",
+ "saving next best rewards: [5176.1636]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15192 fps step and policy inference: 8602 fps total: 7552 epoch: 803/2000\n",
+ "saving next best rewards: [5189.043]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16318 fps step and policy inference: 10025 fps total: 8806 epoch: 804/2000\n",
+ "saving next best rewards: [5208.1953]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16948 fps step and policy inference: 11180 fps total: 9888 epoch: 805/2000\n",
+ "saving next best rewards: [5222.5503]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17096 fps step and policy inference: 11152 fps total: 9837 epoch: 806/2000\n",
+ "saving next best rewards: [5244.8604]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15824 fps step and policy inference: 9709 fps total: 8693 epoch: 807/2000\n",
+ "saving next best rewards: [5248.9]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17829 fps step and policy inference: 12206 fps total: 11276 epoch: 808/2000\n",
+ "saving next best rewards: [5261.3813]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15059 fps step and policy inference: 8621 fps total: 7563 epoch: 809/2000\n",
+ "saving next best rewards: [5277.6787]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16190 fps step and policy inference: 10175 fps total: 9005 epoch: 810/2000\n",
+ "saving next best rewards: [5290.856]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16748 fps step and policy inference: 10685 fps total: 9231 epoch: 811/2000\n",
+ "fps step: 15581 fps step and policy inference: 9157 fps total: 8067 epoch: 812/2000\n",
+ "fps step: 15117 fps step and policy inference: 8547 fps total: 7483 epoch: 813/2000\n",
+ "fps step: 16726 fps step and policy inference: 10212 fps total: 8549 epoch: 814/2000\n",
+ "fps step: 15665 fps step and policy inference: 9242 fps total: 8007 epoch: 815/2000\n",
+ "fps step: 15402 fps step and policy inference: 8904 fps total: 7757 epoch: 816/2000\n",
+ "fps step: 17111 fps step and policy inference: 10890 fps total: 9365 epoch: 817/2000\n",
+ "fps step: 15114 fps step and policy inference: 8513 fps total: 7448 epoch: 818/2000\n",
+ "fps step: 17503 fps step and policy inference: 11310 fps total: 9562 epoch: 819/2000\n",
+ "fps step: 17104 fps step and policy inference: 10915 fps total: 9267 epoch: 820/2000\n",
+ "fps step: 15737 fps step and policy inference: 9252 fps total: 8252 epoch: 821/2000\n",
+ "fps step: 16779 fps step and policy inference: 10387 fps total: 8886 epoch: 822/2000\n",
+ "fps step: 15033 fps step and policy inference: 8658 fps total: 7535 epoch: 823/2000\n",
+ "fps step: 15621 fps step and policy inference: 9341 fps total: 8310 epoch: 824/2000\n",
+ "fps step: 16877 fps step and policy inference: 10687 fps total: 9445 epoch: 825/2000\n",
+ "fps step: 15443 fps step and policy inference: 8826 fps total: 7711 epoch: 826/2000\n",
+ "fps step: 14871 fps step and policy inference: 8303 fps total: 7364 epoch: 827/2000\n",
+ "fps step: 16634 fps step and policy inference: 10586 fps total: 9405 epoch: 828/2000\n",
+ "fps step: 17614 fps step and policy inference: 11920 fps total: 10377 epoch: 829/2000\n",
+ "fps step: 18774 fps step and policy inference: 13110 fps total: 11308 epoch: 830/2000\n",
+ "fps step: 16162 fps step and policy inference: 9796 fps total: 8642 epoch: 831/2000\n",
+ "fps step: 17210 fps step and policy inference: 11211 fps total: 9533 epoch: 832/2000\n",
+ "fps step: 17145 fps step and policy inference: 11097 fps total: 9497 epoch: 833/2000\n",
+ "fps step: 15049 fps step and policy inference: 8758 fps total: 7681 epoch: 834/2000\n",
+ "fps step: 17070 fps step and policy inference: 11372 fps total: 9878 epoch: 835/2000\n",
+ "fps step: 15148 fps step and policy inference: 8905 fps total: 7796 epoch: 836/2000\n",
+ "fps step: 15572 fps step and policy inference: 9101 fps total: 7888 epoch: 837/2000\n",
+ "fps step: 17229 fps step and policy inference: 11040 fps total: 9444 epoch: 838/2000\n",
+ "fps step: 15210 fps step and policy inference: 8892 fps total: 7965 epoch: 839/2000\n",
+ "fps step: 16359 fps step and policy inference: 10130 fps total: 8989 epoch: 840/2000\n",
+ "fps step: 14911 fps step and policy inference: 8491 fps total: 7455 epoch: 841/2000\n",
+ "fps step: 17587 fps step and policy inference: 11936 fps total: 10445 epoch: 842/2000\n",
+ "fps step: 16369 fps step and policy inference: 10202 fps total: 8917 epoch: 843/2000\n",
+ "fps step: 15583 fps step and policy inference: 9166 fps total: 7958 epoch: 844/2000\n",
+ "fps step: 15157 fps step and policy inference: 8649 fps total: 7605 epoch: 845/2000\n",
+ "fps step: 15230 fps step and policy inference: 8854 fps total: 7882 epoch: 846/2000\n",
+ "fps step: 15793 fps step and policy inference: 9686 fps total: 8423 epoch: 847/2000\n",
+ "fps step: 17618 fps step and policy inference: 11710 fps total: 9995 epoch: 848/2000\n",
+ "fps step: 15592 fps step and policy inference: 9259 fps total: 8013 epoch: 849/2000\n",
+ "fps step: 15830 fps step and policy inference: 9637 fps total: 8581 epoch: 850/2000\n",
+ "fps step: 14576 fps step and policy inference: 8234 fps total: 7174 epoch: 851/2000\n",
+ "fps step: 14655 fps step and policy inference: 8415 fps total: 7375 epoch: 852/2000\n",
+ "fps step: 16071 fps step and policy inference: 9991 fps total: 8761 epoch: 853/2000\n",
+ "fps step: 16330 fps step and policy inference: 10268 fps total: 9092 epoch: 854/2000\n",
+ "fps step: 16251 fps step and policy inference: 9873 fps total: 8493 epoch: 855/2000\n",
+ "fps step: 15285 fps step and policy inference: 8847 fps total: 7847 epoch: 856/2000\n",
+ "fps step: 16658 fps step and policy inference: 10559 fps total: 9116 epoch: 857/2000\n",
+ "fps step: 14509 fps step and policy inference: 8084 fps total: 7135 epoch: 858/2000\n",
+ "fps step: 17145 fps step and policy inference: 11404 fps total: 9962 epoch: 859/2000\n",
+ "fps step: 15128 fps step and policy inference: 8732 fps total: 7641 epoch: 860/2000\n",
+ "fps step: 15605 fps step and policy inference: 9125 fps total: 7929 epoch: 861/2000\n",
+ "fps step: 16748 fps step and policy inference: 10641 fps total: 9396 epoch: 862/2000\n",
+ "fps step: 17646 fps step and policy inference: 12104 fps total: 10600 epoch: 863/2000\n",
+ "fps step: 16288 fps step and policy inference: 9890 fps total: 8485 epoch: 864/2000\n",
+ "fps step: 15102 fps step and policy inference: 8566 fps total: 7532 epoch: 865/2000\n",
+ "fps step: 15785 fps step and policy inference: 9254 fps total: 8071 epoch: 866/2000\n",
+ "fps step: 15416 fps step and policy inference: 9084 fps total: 8124 epoch: 867/2000\n",
+ "fps step: 16660 fps step and policy inference: 10460 fps total: 8927 epoch: 868/2000\n",
+ "fps step: 16005 fps step and policy inference: 9461 fps total: 8197 epoch: 869/2000\n",
+ "fps step: 15470 fps step and policy inference: 8854 fps total: 7733 epoch: 870/2000\n",
+ "fps step: 16246 fps step and policy inference: 9910 fps total: 8556 epoch: 871/2000\n",
+ "fps step: 15992 fps step and policy inference: 9822 fps total: 8755 epoch: 872/2000\n",
+ "fps step: 16678 fps step and policy inference: 10811 fps total: 9524 epoch: 873/2000\n",
+ "fps step: 17073 fps step and policy inference: 11217 fps total: 9838 epoch: 874/2000\n",
+ "fps step: 15484 fps step and policy inference: 9155 fps total: 7941 epoch: 875/2000\n",
+ "fps step: 15121 fps step and policy inference: 8668 fps total: 7548 epoch: 876/2000\n",
+ "fps step: 16602 fps step and policy inference: 10582 fps total: 9124 epoch: 877/2000\n",
+ "fps step: 16548 fps step and policy inference: 10406 fps total: 9199 epoch: 878/2000\n",
+ "fps step: 17634 fps step and policy inference: 11662 fps total: 10065 epoch: 879/2000\n",
+ "fps step: 15070 fps step and policy inference: 8504 fps total: 7458 epoch: 880/2000\n",
+ "fps step: 17178 fps step and policy inference: 11315 fps total: 9968 epoch: 881/2000\n",
+ "fps step: 16502 fps step and policy inference: 10176 fps total: 8792 epoch: 882/2000\n",
+ "fps step: 14947 fps step and policy inference: 8377 fps total: 7380 epoch: 883/2000\n",
+ "fps step: 16352 fps step and policy inference: 9794 fps total: 8612 epoch: 884/2000\n",
+ "fps step: 14915 fps step and policy inference: 8529 fps total: 7566 epoch: 885/2000\n",
+ "fps step: 16935 fps step and policy inference: 10709 fps total: 9039 epoch: 886/2000\n",
+ "fps step: 17009 fps step and policy inference: 11072 fps total: 9488 epoch: 887/2000\n",
+ "fps step: 15125 fps step and policy inference: 8694 fps total: 7597 epoch: 888/2000\n",
+ "fps step: 16483 fps step and policy inference: 10680 fps total: 9392 epoch: 889/2000\n",
+ "fps step: 16367 fps step and policy inference: 10516 fps total: 9452 epoch: 890/2000\n",
+ "fps step: 15342 fps step and policy inference: 8859 fps total: 7780 epoch: 891/2000\n",
+ "fps step: 15056 fps step and policy inference: 8762 fps total: 7813 epoch: 892/2000\n",
+ "fps step: 15633 fps step and policy inference: 9514 fps total: 8479 epoch: 893/2000\n",
+ "fps step: 15027 fps step and policy inference: 8650 fps total: 7484 epoch: 894/2000\n",
+ "fps step: 16470 fps step and policy inference: 10579 fps total: 8919 epoch: 895/2000\n",
+ "saving next best rewards: [5297.269]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15543 fps step and policy inference: 9061 fps total: 7932 epoch: 896/2000\n",
+ "fps step: 17071 fps step and policy inference: 10697 fps total: 9212 epoch: 897/2000\n",
+ "fps step: 17442 fps step and policy inference: 11559 fps total: 10061 epoch: 898/2000\n",
+ "fps step: 15378 fps step and policy inference: 8861 fps total: 7752 epoch: 899/2000\n",
+ "saving next best rewards: [5323.837]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16871 fps step and policy inference: 10971 fps total: 9699 epoch: 900/2000\n",
+ "saving next best rewards: [5334.956]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17778 fps step and policy inference: 11773 fps total: 10105 epoch: 901/2000\n",
+ "fps step: 16166 fps step and policy inference: 10036 fps total: 8971 epoch: 902/2000\n",
+ "fps step: 17389 fps step and policy inference: 11455 fps total: 10028 epoch: 903/2000\n",
+ "fps step: 18025 fps step and policy inference: 12243 fps total: 10661 epoch: 904/2000\n",
+ "fps step: 15927 fps step and policy inference: 9332 fps total: 8128 epoch: 905/2000\n",
+ "fps step: 17753 fps step and policy inference: 12129 fps total: 10367 epoch: 906/2000\n",
+ "fps step: 17984 fps step and policy inference: 11754 fps total: 9903 epoch: 907/2000\n",
+ "fps step: 14777 fps step and policy inference: 8317 fps total: 7346 epoch: 908/2000\n",
+ "fps step: 15194 fps step and policy inference: 8732 fps total: 7723 epoch: 909/2000\n",
+ "saving next best rewards: [5345.693]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17132 fps step and policy inference: 10933 fps total: 9259 epoch: 910/2000\n",
+ "saving next best rewards: [5349.248]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16208 fps step and policy inference: 9959 fps total: 8684 epoch: 911/2000\n",
+ "saving next best rewards: [5361.198]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16124 fps step and policy inference: 10075 fps total: 8980 epoch: 912/2000\n",
+ "saving next best rewards: [5366.943]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 18139 fps step and policy inference: 12639 fps total: 10935 epoch: 913/2000\n",
+ "fps step: 15207 fps step and policy inference: 8595 fps total: 7558 epoch: 914/2000\n",
+ "fps step: 16921 fps step and policy inference: 10678 fps total: 9214 epoch: 915/2000\n",
+ "fps step: 16806 fps step and policy inference: 10734 fps total: 9444 epoch: 916/2000\n",
+ "fps step: 15535 fps step and policy inference: 9036 fps total: 7969 epoch: 917/2000\n",
+ "fps step: 16103 fps step and policy inference: 9504 fps total: 8255 epoch: 918/2000\n",
+ "fps step: 15695 fps step and policy inference: 9566 fps total: 8579 epoch: 919/2000\n",
+ "fps step: 15652 fps step and policy inference: 9115 fps total: 7976 epoch: 920/2000\n",
+ "fps step: 15430 fps step and policy inference: 8758 fps total: 7682 epoch: 921/2000\n",
+ "fps step: 18275 fps step and policy inference: 12582 fps total: 10456 epoch: 922/2000\n",
+ "fps step: 19059 fps step and policy inference: 12779 fps total: 11222 epoch: 923/2000\n",
+ "fps step: 15127 fps step and policy inference: 8680 fps total: 7671 epoch: 924/2000\n",
+ "fps step: 17537 fps step and policy inference: 11589 fps total: 9892 epoch: 925/2000\n",
+ "fps step: 14647 fps step and policy inference: 8166 fps total: 7191 epoch: 926/2000\n",
+ "fps step: 15624 fps step and policy inference: 9342 fps total: 8358 epoch: 927/2000\n",
+ "fps step: 15818 fps step and policy inference: 9505 fps total: 8488 epoch: 928/2000\n",
+ "fps step: 16562 fps step and policy inference: 10238 fps total: 8841 epoch: 929/2000\n",
+ "fps step: 15699 fps step and policy inference: 9147 fps total: 7776 epoch: 930/2000\n",
+ "fps step: 15871 fps step and policy inference: 9625 fps total: 8529 epoch: 931/2000\n",
+ "fps step: 16932 fps step and policy inference: 10752 fps total: 9292 epoch: 932/2000\n",
+ "fps step: 15449 fps step and policy inference: 9085 fps total: 8029 epoch: 933/2000\n",
+ "fps step: 16806 fps step and policy inference: 10690 fps total: 9455 epoch: 934/2000\n",
+ "fps step: 15873 fps step and policy inference: 9729 fps total: 8697 epoch: 935/2000\n",
+ "fps step: 15488 fps step and policy inference: 9157 fps total: 8196 epoch: 936/2000\n",
+ "fps step: 17152 fps step and policy inference: 10695 fps total: 9238 epoch: 937/2000\n",
+ "fps step: 16035 fps step and policy inference: 9713 fps total: 8428 epoch: 938/2000\n",
+ "fps step: 15559 fps step and policy inference: 9225 fps total: 8155 epoch: 939/2000\n",
+ "fps step: 16797 fps step and policy inference: 10698 fps total: 9465 epoch: 940/2000\n",
+ "fps step: 15308 fps step and policy inference: 8930 fps total: 7925 epoch: 941/2000\n",
+ "fps step: 15399 fps step and policy inference: 8786 fps total: 7603 epoch: 942/2000\n",
+ "fps step: 16312 fps step and policy inference: 9771 fps total: 8432 epoch: 943/2000\n",
+ "fps step: 15939 fps step and policy inference: 9591 fps total: 8613 epoch: 944/2000\n",
+ "fps step: 16886 fps step and policy inference: 10793 fps total: 9559 epoch: 945/2000\n",
+ "fps step: 17093 fps step and policy inference: 11380 fps total: 10005 epoch: 946/2000\n",
+ "fps step: 15011 fps step and policy inference: 8631 fps total: 7679 epoch: 947/2000\n",
+ "fps step: 15144 fps step and policy inference: 8684 fps total: 7651 epoch: 948/2000\n",
+ "fps step: 16213 fps step and policy inference: 9948 fps total: 8792 epoch: 949/2000\n",
+ "fps step: 15850 fps step and policy inference: 9617 fps total: 8556 epoch: 950/2000\n",
+ "fps step: 16195 fps step and policy inference: 10027 fps total: 8907 epoch: 951/2000\n",
+ "fps step: 15733 fps step and policy inference: 9250 fps total: 8068 epoch: 952/2000\n",
+ "fps step: 18042 fps step and policy inference: 12686 fps total: 10930 epoch: 953/2000\n",
+ "fps step: 16269 fps step and policy inference: 9942 fps total: 8818 epoch: 954/2000\n",
+ "fps step: 17071 fps step and policy inference: 10933 fps total: 9348 epoch: 955/2000\n",
+ "fps step: 15958 fps step and policy inference: 9610 fps total: 8476 epoch: 956/2000\n",
+ "fps step: 15830 fps step and policy inference: 9340 fps total: 8055 epoch: 957/2000\n",
+ "fps step: 17550 fps step and policy inference: 11762 fps total: 10324 epoch: 958/2000\n",
+ "fps step: 15293 fps step and policy inference: 8888 fps total: 7785 epoch: 959/2000\n",
+ "fps step: 15194 fps step and policy inference: 8645 fps total: 7593 epoch: 960/2000\n",
+ "fps step: 15500 fps step and policy inference: 8956 fps total: 8273 epoch: 961/2000\n",
+ "fps step: 18299 fps step and policy inference: 12889 fps total: 10969 epoch: 962/2000\n",
+ "fps step: 16997 fps step and policy inference: 10872 fps total: 9564 epoch: 963/2000\n",
+ "fps step: 16334 fps step and policy inference: 9836 fps total: 8562 epoch: 964/2000\n",
+ "fps step: 14796 fps step and policy inference: 8281 fps total: 7386 epoch: 965/2000\n",
+ "fps step: 16842 fps step and policy inference: 10608 fps total: 9146 epoch: 966/2000\n",
+ "fps step: 15517 fps step and policy inference: 9085 fps total: 8124 epoch: 967/2000\n",
+ "fps step: 15574 fps step and policy inference: 9072 fps total: 7811 epoch: 968/2000\n",
+ "fps step: 15474 fps step and policy inference: 9267 fps total: 8282 epoch: 969/2000\n",
+ "fps step: 16212 fps step and policy inference: 10092 fps total: 8976 epoch: 970/2000\n",
+ "fps step: 16772 fps step and policy inference: 10392 fps total: 8844 epoch: 971/2000\n",
+ "fps step: 16704 fps step and policy inference: 10241 fps total: 8790 epoch: 972/2000\n",
+ "fps step: 16183 fps step and policy inference: 9800 fps total: 8655 epoch: 973/2000\n",
+ "fps step: 14709 fps step and policy inference: 8322 fps total: 7361 epoch: 974/2000\n",
+ "fps step: 15807 fps step and policy inference: 9538 fps total: 8505 epoch: 975/2000\n",
+ "fps step: 15323 fps step and policy inference: 8950 fps total: 7987 epoch: 976/2000\n",
+ "fps step: 15384 fps step and policy inference: 8842 fps total: 7705 epoch: 977/2000\n",
+ "fps step: 17373 fps step and policy inference: 11620 fps total: 10205 epoch: 978/2000\n",
+ "fps step: 17090 fps step and policy inference: 10722 fps total: 9185 epoch: 979/2000\n",
+ "fps step: 16525 fps step and policy inference: 10110 fps total: 8965 epoch: 980/2000\n",
+ "fps step: 15567 fps step and policy inference: 9336 fps total: 8352 epoch: 981/2000\n",
+ "saving next best rewards: [5376.0947]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16584 fps step and policy inference: 10326 fps total: 8950 epoch: 982/2000\n",
+ "fps step: 16197 fps step and policy inference: 9889 fps total: 8543 epoch: 983/2000\n",
+ "fps step: 17493 fps step and policy inference: 11392 fps total: 9758 epoch: 984/2000\n",
+ "fps step: 15351 fps step and policy inference: 8927 fps total: 7796 epoch: 985/2000\n",
+ "saving next best rewards: [5384.0703]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 18056 fps step and policy inference: 12631 fps total: 10913 epoch: 986/2000\n",
+ "saving next best rewards: [5389.7104]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16290 fps step and policy inference: 9807 fps total: 8533 epoch: 987/2000\n",
+ "saving next best rewards: [5407.5757]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15232 fps step and policy inference: 8903 fps total: 7945 epoch: 988/2000\n",
+ "saving next best rewards: [5421.5806]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16814 fps step and policy inference: 10489 fps total: 8996 epoch: 989/2000\n",
+ "saving next best rewards: [5438.559]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16483 fps step and policy inference: 10297 fps total: 9164 epoch: 990/2000\n",
+ "fps step: 15507 fps step and policy inference: 9017 fps total: 7879 epoch: 991/2000\n",
+ "saving next best rewards: [5448.995]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15648 fps step and policy inference: 9290 fps total: 8262 epoch: 992/2000\n",
+ "saving next best rewards: [5457.6274]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14944 fps step and policy inference: 8380 fps total: 7362 epoch: 993/2000\n",
+ "saving next best rewards: [5485.7905]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16101 fps step and policy inference: 9680 fps total: 8352 epoch: 994/2000\n",
+ "saving next best rewards: [5497.416]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16259 fps step and policy inference: 9878 fps total: 8727 epoch: 995/2000\n",
+ "saving next best rewards: [5498.2646]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15982 fps step and policy inference: 9480 fps total: 8190 epoch: 996/2000\n",
+ "saving next best rewards: [5525.507]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15414 fps step and policy inference: 9031 fps total: 7882 epoch: 997/2000\n",
+ "saving next best rewards: [5542.9243]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17299 fps step and policy inference: 11141 fps total: 9552 epoch: 998/2000\n",
+ "saving next best rewards: [5557.8936]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16082 fps step and policy inference: 9763 fps total: 8514 epoch: 999/2000\n",
+ "saving next best rewards: [5580.673]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17522 fps step and policy inference: 11554 fps total: 9641 epoch: 1000/2000\n",
+ "saving next best rewards: [5595.4824]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15405 fps step and policy inference: 8942 fps total: 7851 epoch: 1001/2000\n",
+ "saving next best rewards: [5608.1284]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16260 fps step and policy inference: 10186 fps total: 9340 epoch: 1002/2000\n",
+ "fps step: 17225 fps step and policy inference: 11151 fps total: 9807 epoch: 1003/2000\n",
+ "fps step: 17180 fps step and policy inference: 11061 fps total: 9693 epoch: 1004/2000\n",
+ "fps step: 17631 fps step and policy inference: 11626 fps total: 10194 epoch: 1005/2000\n",
+ "saving next best rewards: [5616.475]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16451 fps step and policy inference: 10283 fps total: 9485 epoch: 1006/2000\n",
+ "saving next best rewards: [5626.394]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17494 fps step and policy inference: 11823 fps total: 10305 epoch: 1007/2000\n",
+ "saving next best rewards: [5639.864]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16563 fps step and policy inference: 10614 fps total: 9415 epoch: 1008/2000\n",
+ "fps step: 17863 fps step and policy inference: 12156 fps total: 10596 epoch: 1009/2000\n",
+ "fps step: 17311 fps step and policy inference: 11462 fps total: 10055 epoch: 1010/2000\n",
+ "fps step: 16869 fps step and policy inference: 10928 fps total: 9601 epoch: 1011/2000\n",
+ "fps step: 16730 fps step and policy inference: 10596 fps total: 9006 epoch: 1012/2000\n",
+ "saving next best rewards: [5659.851]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15470 fps step and policy inference: 9159 fps total: 8125 epoch: 1013/2000\n",
+ "saving next best rewards: [5669.6616]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16017 fps step and policy inference: 9482 fps total: 8261 epoch: 1014/2000\n",
+ "saving next best rewards: [5685.238]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15316 fps step and policy inference: 8777 fps total: 7641 epoch: 1015/2000\n",
+ "fps step: 18259 fps step and policy inference: 13042 fps total: 11253 epoch: 1016/2000\n",
+ "fps step: 15174 fps step and policy inference: 8653 fps total: 7537 epoch: 1017/2000\n",
+ "fps step: 16135 fps step and policy inference: 9968 fps total: 8909 epoch: 1018/2000\n",
+ "fps step: 17342 fps step and policy inference: 11518 fps total: 10045 epoch: 1019/2000\n",
+ "fps step: 17751 fps step and policy inference: 11519 fps total: 9777 epoch: 1020/2000\n",
+ "fps step: 15641 fps step and policy inference: 9184 fps total: 7803 epoch: 1021/2000\n",
+ "fps step: 15505 fps step and policy inference: 9220 fps total: 8094 epoch: 1022/2000\n",
+ "fps step: 15654 fps step and policy inference: 9231 fps total: 8041 epoch: 1023/2000\n",
+ "fps step: 16691 fps step and policy inference: 10464 fps total: 8939 epoch: 1024/2000\n",
+ "fps step: 15551 fps step and policy inference: 8949 fps total: 7704 epoch: 1025/2000\n",
+ "fps step: 15678 fps step and policy inference: 9100 fps total: 7905 epoch: 1026/2000\n",
+ "fps step: 14731 fps step and policy inference: 8195 fps total: 7249 epoch: 1027/2000\n",
+ "fps step: 15119 fps step and policy inference: 8724 fps total: 8159 epoch: 1028/2000\n",
+ "fps step: 17516 fps step and policy inference: 11669 fps total: 10079 epoch: 1029/2000\n",
+ "fps step: 15739 fps step and policy inference: 9343 fps total: 8326 epoch: 1030/2000\n",
+ "fps step: 16869 fps step and policy inference: 10443 fps total: 9023 epoch: 1031/2000\n",
+ "fps step: 17101 fps step and policy inference: 11185 fps total: 9834 epoch: 1032/2000\n",
+ "saving next best rewards: [5689.02]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15947 fps step and policy inference: 9321 fps total: 8113 epoch: 1033/2000\n",
+ "saving next best rewards: [5699.36]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14906 fps step and policy inference: 8317 fps total: 7265 epoch: 1034/2000\n",
+ "fps step: 15058 fps step and policy inference: 8519 fps total: 7890 epoch: 1035/2000\n",
+ "fps step: 17003 fps step and policy inference: 10807 fps total: 9372 epoch: 1036/2000\n",
+ "fps step: 18308 fps step and policy inference: 12802 fps total: 11079 epoch: 1037/2000\n",
+ "fps step: 18045 fps step and policy inference: 12397 fps total: 10757 epoch: 1038/2000\n",
+ "fps step: 15987 fps step and policy inference: 9490 fps total: 8189 epoch: 1039/2000\n",
+ "fps step: 16334 fps step and policy inference: 10041 fps total: 8740 epoch: 1040/2000\n",
+ "fps step: 14899 fps step and policy inference: 8314 fps total: 7309 epoch: 1041/2000\n",
+ "fps step: 15758 fps step and policy inference: 9271 fps total: 8181 epoch: 1042/2000\n",
+ "fps step: 16345 fps step and policy inference: 10213 fps total: 9397 epoch: 1043/2000\n",
+ "fps step: 16024 fps step and policy inference: 9658 fps total: 8579 epoch: 1044/2000\n",
+ "fps step: 18125 fps step and policy inference: 12747 fps total: 11056 epoch: 1045/2000\n",
+ "fps step: 16452 fps step and policy inference: 10127 fps total: 8833 epoch: 1046/2000\n",
+ "fps step: 15514 fps step and policy inference: 9196 fps total: 8119 epoch: 1047/2000\n",
+ "fps step: 18149 fps step and policy inference: 12436 fps total: 10399 epoch: 1048/2000\n",
+ "fps step: 16237 fps step and policy inference: 9923 fps total: 8836 epoch: 1049/2000\n",
+ "fps step: 15437 fps step and policy inference: 8844 fps total: 7747 epoch: 1050/2000\n",
+ "fps step: 15172 fps step and policy inference: 8815 fps total: 7712 epoch: 1051/2000\n",
+ "fps step: 15503 fps step and policy inference: 8867 fps total: 7679 epoch: 1052/2000\n",
+ "fps step: 16433 fps step and policy inference: 10354 fps total: 9179 epoch: 1053/2000\n",
+ "fps step: 17524 fps step and policy inference: 11830 fps total: 10317 epoch: 1054/2000\n",
+ "fps step: 15430 fps step and policy inference: 8969 fps total: 7895 epoch: 1055/2000\n",
+ "fps step: 17086 fps step and policy inference: 11033 fps total: 9678 epoch: 1056/2000\n",
+ "fps step: 15027 fps step and policy inference: 8565 fps total: 7644 epoch: 1057/2000\n",
+ "fps step: 15786 fps step and policy inference: 9544 fps total: 8504 epoch: 1058/2000\n",
+ "fps step: 16216 fps step and policy inference: 9831 fps total: 8642 epoch: 1059/2000\n",
+ "fps step: 15669 fps step and policy inference: 9049 fps total: 7786 epoch: 1060/2000\n",
+ "fps step: 16635 fps step and policy inference: 10564 fps total: 9318 epoch: 1061/2000\n",
+ "fps step: 15776 fps step and policy inference: 9230 fps total: 7923 epoch: 1062/2000\n",
+ "fps step: 16345 fps step and policy inference: 9868 fps total: 8762 epoch: 1063/2000\n",
+ "fps step: 15420 fps step and policy inference: 8918 fps total: 7791 epoch: 1064/2000\n",
+ "fps step: 16888 fps step and policy inference: 10598 fps total: 9260 epoch: 1065/2000\n",
+ "fps step: 16545 fps step and policy inference: 10028 fps total: 8677 epoch: 1066/2000\n",
+ "fps step: 15188 fps step and policy inference: 8676 fps total: 7552 epoch: 1067/2000\n",
+ "fps step: 17586 fps step and policy inference: 11595 fps total: 9972 epoch: 1068/2000\n",
+ "fps step: 17350 fps step and policy inference: 11022 fps total: 9175 epoch: 1069/2000\n",
+ "fps step: 15412 fps step and policy inference: 9045 fps total: 8090 epoch: 1070/2000\n",
+ "fps step: 16321 fps step and policy inference: 9825 fps total: 8502 epoch: 1071/2000\n",
+ "fps step: 15773 fps step and policy inference: 9489 fps total: 8496 epoch: 1072/2000\n",
+ "fps step: 14929 fps step and policy inference: 8472 fps total: 7565 epoch: 1073/2000\n",
+ "fps step: 16791 fps step and policy inference: 10581 fps total: 9360 epoch: 1074/2000\n",
+ "fps step: 16867 fps step and policy inference: 10555 fps total: 9267 epoch: 1075/2000\n",
+ "fps step: 15994 fps step and policy inference: 9746 fps total: 8717 epoch: 1076/2000\n",
+ "fps step: 14946 fps step and policy inference: 8465 fps total: 7432 epoch: 1077/2000\n",
+ "fps step: 15553 fps step and policy inference: 9032 fps total: 7693 epoch: 1078/2000\n",
+ "fps step: 15498 fps step and policy inference: 9095 fps total: 8171 epoch: 1079/2000\n",
+ "fps step: 16595 fps step and policy inference: 10264 fps total: 8965 epoch: 1080/2000\n",
+ "fps step: 15583 fps step and policy inference: 9037 fps total: 7731 epoch: 1081/2000\n",
+ "fps step: 16545 fps step and policy inference: 10222 fps total: 8723 epoch: 1082/2000\n",
+ "fps step: 15488 fps step and policy inference: 9137 fps total: 7977 epoch: 1083/2000\n",
+ "fps step: 15532 fps step and policy inference: 9279 fps total: 8104 epoch: 1084/2000\n",
+ "fps step: 15466 fps step and policy inference: 8993 fps total: 7770 epoch: 1085/2000\n",
+ "fps step: 16045 fps step and policy inference: 9508 fps total: 8245 epoch: 1086/2000\n",
+ "fps step: 16431 fps step and policy inference: 9932 fps total: 8607 epoch: 1087/2000\n",
+ "fps step: 17254 fps step and policy inference: 11408 fps total: 10056 epoch: 1088/2000\n",
+ "fps step: 16065 fps step and policy inference: 9433 fps total: 8304 epoch: 1089/2000\n",
+ "fps step: 16288 fps step and policy inference: 9845 fps total: 8438 epoch: 1090/2000\n",
+ "fps step: 15198 fps step and policy inference: 8656 fps total: 7586 epoch: 1091/2000\n",
+ "fps step: 14782 fps step and policy inference: 8352 fps total: 7271 epoch: 1092/2000\n",
+ "fps step: 16851 fps step and policy inference: 10811 fps total: 9579 epoch: 1093/2000\n",
+ "fps step: 17396 fps step and policy inference: 11705 fps total: 10293 epoch: 1094/2000\n",
+ "fps step: 18192 fps step and policy inference: 13032 fps total: 11291 epoch: 1095/2000\n",
+ "fps step: 14986 fps step and policy inference: 8580 fps total: 7572 epoch: 1096/2000\n",
+ "fps step: 15242 fps step and policy inference: 8809 fps total: 7675 epoch: 1097/2000\n",
+ "fps step: 15641 fps step and policy inference: 9437 fps total: 8431 epoch: 1098/2000\n",
+ "fps step: 14918 fps step and policy inference: 8418 fps total: 7394 epoch: 1099/2000\n",
+ "fps step: 16833 fps step and policy inference: 10562 fps total: 9341 epoch: 1100/2000\n",
+ "fps step: 17637 fps step and policy inference: 11888 fps total: 10418 epoch: 1101/2000\n",
+ "fps step: 17854 fps step and policy inference: 12107 fps total: 10531 epoch: 1102/2000\n",
+ "fps step: 17107 fps step and policy inference: 11184 fps total: 9833 epoch: 1103/2000\n",
+ "fps step: 15322 fps step and policy inference: 8921 fps total: 8008 epoch: 1104/2000\n",
+ "fps step: 15506 fps step and policy inference: 9077 fps total: 7893 epoch: 1105/2000\n",
+ "fps step: 17614 fps step and policy inference: 11760 fps total: 9947 epoch: 1106/2000\n",
+ "fps step: 17553 fps step and policy inference: 11964 fps total: 10415 epoch: 1107/2000\n",
+ "fps step: 14859 fps step and policy inference: 8440 fps total: 7435 epoch: 1108/2000\n",
+ "fps step: 15445 fps step and policy inference: 9270 fps total: 8158 epoch: 1109/2000\n",
+ "fps step: 16160 fps step and policy inference: 10196 fps total: 9036 epoch: 1110/2000\n",
+ "fps step: 14792 fps step and policy inference: 8684 fps total: 7851 epoch: 1111/2000\n",
+ "fps step: 14602 fps step and policy inference: 8354 fps total: 7392 epoch: 1112/2000\n",
+ "fps step: 15093 fps step and policy inference: 8939 fps total: 7639 epoch: 1113/2000\n",
+ "fps step: 16136 fps step and policy inference: 10159 fps total: 8681 epoch: 1114/2000\n",
+ "fps step: 14595 fps step and policy inference: 8456 fps total: 7491 epoch: 1115/2000\n",
+ "fps step: 15838 fps step and policy inference: 9913 fps total: 8782 epoch: 1116/2000\n",
+ "fps step: 14809 fps step and policy inference: 8616 fps total: 7550 epoch: 1117/2000\n",
+ "fps step: 15868 fps step and policy inference: 9663 fps total: 8311 epoch: 1118/2000\n",
+ "fps step: 15601 fps step and policy inference: 9543 fps total: 8788 epoch: 1119/2000\n",
+ "fps step: 15485 fps step and policy inference: 9124 fps total: 8014 epoch: 1120/2000\n",
+ "fps step: 16178 fps step and policy inference: 10002 fps total: 8924 epoch: 1121/2000\n",
+ "fps step: 16693 fps step and policy inference: 10422 fps total: 9172 epoch: 1122/2000\n",
+ "fps step: 15565 fps step and policy inference: 8987 fps total: 7754 epoch: 1123/2000\n",
+ "fps step: 17436 fps step and policy inference: 11272 fps total: 9797 epoch: 1124/2000\n",
+ "fps step: 15550 fps step and policy inference: 8939 fps total: 7820 epoch: 1125/2000\n",
+ "fps step: 15693 fps step and policy inference: 9304 fps total: 8272 epoch: 1126/2000\n",
+ "fps step: 14949 fps step and policy inference: 8414 fps total: 7395 epoch: 1127/2000\n",
+ "fps step: 17298 fps step and policy inference: 11444 fps total: 10075 epoch: 1128/2000\n",
+ "fps step: 15786 fps step and policy inference: 9529 fps total: 8470 epoch: 1129/2000\n",
+ "fps step: 16985 fps step and policy inference: 10620 fps total: 9132 epoch: 1130/2000\n",
+ "fps step: 15949 fps step and policy inference: 9625 fps total: 8570 epoch: 1131/2000\n",
+ "fps step: 15316 fps step and policy inference: 8874 fps total: 7904 epoch: 1132/2000\n",
+ "fps step: 17225 fps step and policy inference: 11114 fps total: 10009 epoch: 1133/2000\n",
+ "fps step: 16326 fps step and policy inference: 9901 fps total: 9049 epoch: 1134/2000\n",
+ "fps step: 15318 fps step and policy inference: 8778 fps total: 7651 epoch: 1135/2000\n",
+ "fps step: 18125 fps step and policy inference: 12749 fps total: 10961 epoch: 1136/2000\n",
+ "fps step: 15883 fps step and policy inference: 9546 fps total: 8472 epoch: 1137/2000\n",
+ "fps step: 15614 fps step and policy inference: 8927 fps total: 7818 epoch: 1138/2000\n",
+ "fps step: 15629 fps step and policy inference: 9511 fps total: 8492 epoch: 1139/2000\n",
+ "fps step: 17930 fps step and policy inference: 12496 fps total: 10917 epoch: 1140/2000\n",
+ "fps step: 16032 fps step and policy inference: 9319 fps total: 8180 epoch: 1141/2000\n",
+ "fps step: 17794 fps step and policy inference: 11755 fps total: 10037 epoch: 1142/2000\n",
+ "fps step: 15023 fps step and policy inference: 8481 fps total: 7468 epoch: 1143/2000\n",
+ "fps step: 16004 fps step and policy inference: 9715 fps total: 8624 epoch: 1144/2000\n",
+ "fps step: 16548 fps step and policy inference: 9846 fps total: 8462 epoch: 1145/2000\n",
+ "fps step: 16813 fps step and policy inference: 10463 fps total: 9193 epoch: 1146/2000\n",
+ "fps step: 15057 fps step and policy inference: 8466 fps total: 7373 epoch: 1147/2000\n",
+ "fps step: 16157 fps step and policy inference: 9530 fps total: 8219 epoch: 1148/2000\n",
+ "fps step: 16588 fps step and policy inference: 10315 fps total: 8870 epoch: 1149/2000\n",
+ "fps step: 15340 fps step and policy inference: 8695 fps total: 7620 epoch: 1150/2000\n",
+ "fps step: 15369 fps step and policy inference: 9077 fps total: 8180 epoch: 1151/2000\n",
+ "fps step: 15876 fps step and policy inference: 9563 fps total: 8501 epoch: 1152/2000\n",
+ "fps step: 16154 fps step and policy inference: 9505 fps total: 8232 epoch: 1153/2000\n",
+ "fps step: 15500 fps step and policy inference: 8973 fps total: 7811 epoch: 1154/2000\n",
+ "fps step: 16802 fps step and policy inference: 10778 fps total: 9538 epoch: 1155/2000\n",
+ "fps step: 15404 fps step and policy inference: 9000 fps total: 8032 epoch: 1156/2000\n",
+ "fps step: 15755 fps step and policy inference: 9584 fps total: 8325 epoch: 1157/2000\n",
+ "fps step: 15230 fps step and policy inference: 8894 fps total: 7781 epoch: 1158/2000\n",
+ "fps step: 15478 fps step and policy inference: 8919 fps total: 7795 epoch: 1159/2000\n",
+ "fps step: 17478 fps step and policy inference: 11935 fps total: 10394 epoch: 1160/2000\n",
+ "fps step: 14876 fps step and policy inference: 8515 fps total: 7637 epoch: 1161/2000\n",
+ "fps step: 16817 fps step and policy inference: 10728 fps total: 9496 epoch: 1162/2000\n",
+ "fps step: 16478 fps step and policy inference: 10107 fps total: 8720 epoch: 1163/2000\n",
+ "fps step: 16037 fps step and policy inference: 9784 fps total: 8715 epoch: 1164/2000\n",
+ "fps step: 14373 fps step and policy inference: 8004 fps total: 7013 epoch: 1165/2000\n",
+ "fps step: 14791 fps step and policy inference: 8453 fps total: 7486 epoch: 1166/2000\n",
+ "fps step: 16042 fps step and policy inference: 9823 fps total: 8562 epoch: 1167/2000\n",
+ "fps step: 15766 fps step and policy inference: 9708 fps total: 8648 epoch: 1168/2000\n",
+ "fps step: 17431 fps step and policy inference: 11451 fps total: 9608 epoch: 1169/2000\n",
+ "fps step: 16694 fps step and policy inference: 10913 fps total: 9657 epoch: 1170/2000\n",
+ "fps step: 16783 fps step and policy inference: 10738 fps total: 9077 epoch: 1171/2000\n",
+ "fps step: 14891 fps step and policy inference: 8577 fps total: 7438 epoch: 1172/2000\n",
+ "fps step: 14350 fps step and policy inference: 8163 fps total: 7193 epoch: 1173/2000\n",
+ "fps step: 15450 fps step and policy inference: 9155 fps total: 7995 epoch: 1174/2000\n",
+ "fps step: 16916 fps step and policy inference: 11317 fps total: 9951 epoch: 1175/2000\n",
+ "fps step: 16192 fps step and policy inference: 10320 fps total: 9176 epoch: 1176/2000\n",
+ "fps step: 15103 fps step and policy inference: 8601 fps total: 7425 epoch: 1177/2000\n",
+ "fps step: 14822 fps step and policy inference: 8327 fps total: 7308 epoch: 1178/2000\n",
+ "fps step: 17387 fps step and policy inference: 11458 fps total: 9560 epoch: 1179/2000\n",
+ "fps step: 15334 fps step and policy inference: 9069 fps total: 7918 epoch: 1180/2000\n",
+ "fps step: 15121 fps step and policy inference: 8858 fps total: 7895 epoch: 1181/2000\n",
+ "fps step: 15905 fps step and policy inference: 9727 fps total: 8443 epoch: 1182/2000\n",
+ "fps step: 14617 fps step and policy inference: 8229 fps total: 7243 epoch: 1183/2000\n",
+ "fps step: 14808 fps step and policy inference: 8217 fps total: 7246 epoch: 1184/2000\n",
+ "fps step: 15877 fps step and policy inference: 9716 fps total: 8656 epoch: 1185/2000\n",
+ "fps step: 17450 fps step and policy inference: 11559 fps total: 10066 epoch: 1186/2000\n",
+ "fps step: 16708 fps step and policy inference: 10441 fps total: 8943 epoch: 1187/2000\n",
+ "fps step: 16864 fps step and policy inference: 10526 fps total: 9249 epoch: 1188/2000\n",
+ "fps step: 15115 fps step and policy inference: 8741 fps total: 7740 epoch: 1189/2000\n",
+ "fps step: 16323 fps step and policy inference: 9912 fps total: 8735 epoch: 1190/2000\n",
+ "fps step: 16446 fps step and policy inference: 10054 fps total: 8718 epoch: 1191/2000\n",
+ "fps step: 18155 fps step and policy inference: 12555 fps total: 11607 epoch: 1192/2000\n",
+ "fps step: 15905 fps step and policy inference: 9402 fps total: 8390 epoch: 1193/2000\n",
+ "fps step: 15750 fps step and policy inference: 9664 fps total: 8651 epoch: 1194/2000\n",
+ "fps step: 16076 fps step and policy inference: 9838 fps total: 8415 epoch: 1195/2000\n",
+ "fps step: 15844 fps step and policy inference: 9750 fps total: 8674 epoch: 1196/2000\n",
+ "fps step: 15568 fps step and policy inference: 9278 fps total: 8241 epoch: 1197/2000\n",
+ "fps step: 15766 fps step and policy inference: 9455 fps total: 8327 epoch: 1198/2000\n",
+ "fps step: 15322 fps step and policy inference: 8915 fps total: 7821 epoch: 1199/2000\n",
+ "fps step: 17487 fps step and policy inference: 11851 fps total: 10351 epoch: 1200/2000\n",
+ "fps step: 15159 fps step and policy inference: 8941 fps total: 7852 epoch: 1201/2000\n",
+ "fps step: 15287 fps step and policy inference: 9561 fps total: 8463 epoch: 1202/2000\n",
+ "fps step: 14637 fps step and policy inference: 8661 fps total: 7561 epoch: 1203/2000\n",
+ "fps step: 15008 fps step and policy inference: 9151 fps total: 8232 epoch: 1204/2000\n",
+ "fps step: 17945 fps step and policy inference: 13033 fps total: 11176 epoch: 1205/2000\n",
+ "fps step: 15796 fps step and policy inference: 9550 fps total: 8281 epoch: 1206/2000\n",
+ "fps step: 14645 fps step and policy inference: 8323 fps total: 7277 epoch: 1207/2000\n",
+ "fps step: 15258 fps step and policy inference: 8974 fps total: 7820 epoch: 1208/2000\n",
+ "fps step: 15020 fps step and policy inference: 8514 fps total: 7498 epoch: 1209/2000\n",
+ "fps step: 16621 fps step and policy inference: 10534 fps total: 8961 epoch: 1210/2000\n",
+ "fps step: 15601 fps step and policy inference: 9247 fps total: 8082 epoch: 1211/2000\n",
+ "fps step: 16849 fps step and policy inference: 10313 fps total: 8802 epoch: 1212/2000\n",
+ "fps step: 16114 fps step and policy inference: 9969 fps total: 8722 epoch: 1213/2000\n",
+ "fps step: 16055 fps step and policy inference: 9713 fps total: 8401 epoch: 1214/2000\n",
+ "fps step: 14755 fps step and policy inference: 8271 fps total: 7266 epoch: 1215/2000\n",
+ "fps step: 14999 fps step and policy inference: 8553 fps total: 7571 epoch: 1216/2000\n",
+ "fps step: 14994 fps step and policy inference: 8568 fps total: 7471 epoch: 1217/2000\n",
+ "fps step: 17440 fps step and policy inference: 11375 fps total: 9941 epoch: 1218/2000\n",
+ "fps step: 15553 fps step and policy inference: 9119 fps total: 8083 epoch: 1219/2000\n",
+ "fps step: 16246 fps step and policy inference: 10145 fps total: 9032 epoch: 1220/2000\n",
+ "fps step: 16830 fps step and policy inference: 10699 fps total: 9456 epoch: 1221/2000\n",
+ "fps step: 17109 fps step and policy inference: 10702 fps total: 9147 epoch: 1222/2000\n",
+ "fps step: 16313 fps step and policy inference: 10388 fps total: 9163 epoch: 1223/2000\n",
+ "fps step: 16195 fps step and policy inference: 9778 fps total: 8477 epoch: 1224/2000\n",
+ "fps step: 15954 fps step and policy inference: 9537 fps total: 8531 epoch: 1225/2000\n",
+ "fps step: 14795 fps step and policy inference: 8209 fps total: 7235 epoch: 1226/2000\n",
+ "fps step: 16123 fps step and policy inference: 9758 fps total: 8637 epoch: 1227/2000\n",
+ "fps step: 14879 fps step and policy inference: 8380 fps total: 7381 epoch: 1228/2000\n",
+ "fps step: 16968 fps step and policy inference: 10934 fps total: 9537 epoch: 1229/2000\n",
+ "fps step: 15249 fps step and policy inference: 8721 fps total: 7729 epoch: 1230/2000\n",
+ "fps step: 16065 fps step and policy inference: 9528 fps total: 8244 epoch: 1231/2000\n",
+ "fps step: 15244 fps step and policy inference: 8698 fps total: 7665 epoch: 1232/2000\n",
+ "fps step: 15961 fps step and policy inference: 9550 fps total: 8517 epoch: 1233/2000\n",
+ "fps step: 17143 fps step and policy inference: 11107 fps total: 9541 epoch: 1234/2000\n",
+ "fps step: 15634 fps step and policy inference: 9180 fps total: 8060 epoch: 1235/2000\n",
+ "fps step: 16037 fps step and policy inference: 9528 fps total: 8575 epoch: 1236/2000\n",
+ "fps step: 16049 fps step and policy inference: 9515 fps total: 8292 epoch: 1237/2000\n",
+ "fps step: 16638 fps step and policy inference: 10496 fps total: 9298 epoch: 1238/2000\n",
+ "fps step: 15120 fps step and policy inference: 8572 fps total: 7518 epoch: 1239/2000\n",
+ "fps step: 16864 fps step and policy inference: 10700 fps total: 9008 epoch: 1240/2000\n",
+ "fps step: 17145 fps step and policy inference: 11443 fps total: 10053 epoch: 1241/2000\n",
+ "fps step: 15506 fps step and policy inference: 9226 fps total: 8275 epoch: 1242/2000\n",
+ "fps step: 16698 fps step and policy inference: 10377 fps total: 8803 epoch: 1243/2000\n",
+ "fps step: 15556 fps step and policy inference: 9213 fps total: 7991 epoch: 1244/2000\n",
+ "fps step: 15546 fps step and policy inference: 9372 fps total: 8171 epoch: 1245/2000\n",
+ "fps step: 15590 fps step and policy inference: 9076 fps total: 7879 epoch: 1246/2000\n",
+ "fps step: 16639 fps step and policy inference: 10573 fps total: 9226 epoch: 1247/2000\n",
+ "fps step: 15055 fps step and policy inference: 8616 fps total: 7617 epoch: 1248/2000\n",
+ "fps step: 14717 fps step and policy inference: 8332 fps total: 7331 epoch: 1249/2000\n",
+ "fps step: 15430 fps step and policy inference: 9080 fps total: 7872 epoch: 1250/2000\n",
+ "fps step: 16396 fps step and policy inference: 10149 fps total: 8875 epoch: 1251/2000\n",
+ "fps step: 18222 fps step and policy inference: 11566 fps total: 9759 epoch: 1252/2000\n",
+ "fps step: 15962 fps step and policy inference: 9434 fps total: 8063 epoch: 1253/2000\n",
+ "fps step: 15479 fps step and policy inference: 9060 fps total: 8047 epoch: 1254/2000\n",
+ "fps step: 15659 fps step and policy inference: 9140 fps total: 7986 epoch: 1255/2000\n",
+ "fps step: 16281 fps step and policy inference: 9833 fps total: 8753 epoch: 1256/2000\n",
+ "fps step: 17220 fps step and policy inference: 11389 fps total: 10046 epoch: 1257/2000\n",
+ "fps step: 16691 fps step and policy inference: 10402 fps total: 9077 epoch: 1258/2000\n",
+ "fps step: 14713 fps step and policy inference: 8259 fps total: 7321 epoch: 1259/2000\n",
+ "fps step: 16020 fps step and policy inference: 9786 fps total: 8677 epoch: 1260/2000\n",
+ "fps step: 17228 fps step and policy inference: 11278 fps total: 9969 epoch: 1261/2000\n",
+ "fps step: 16760 fps step and policy inference: 10542 fps total: 9262 epoch: 1262/2000\n",
+ "fps step: 15474 fps step and policy inference: 9018 fps total: 7952 epoch: 1263/2000\n",
+ "fps step: 14995 fps step and policy inference: 8436 fps total: 7420 epoch: 1264/2000\n",
+ "fps step: 15099 fps step and policy inference: 8772 fps total: 7902 epoch: 1265/2000\n",
+ "fps step: 17928 fps step and policy inference: 12032 fps total: 9822 epoch: 1266/2000\n",
+ "fps step: 16914 fps step and policy inference: 10891 fps total: 9635 epoch: 1267/2000\n",
+ "fps step: 14976 fps step and policy inference: 8348 fps total: 7322 epoch: 1268/2000\n",
+ "fps step: 15475 fps step and policy inference: 8889 fps total: 7739 epoch: 1269/2000\n",
+ "fps step: 16078 fps step and policy inference: 9972 fps total: 8910 epoch: 1270/2000\n",
+ "fps step: 14995 fps step and policy inference: 8468 fps total: 7507 epoch: 1271/2000\n",
+ "fps step: 15460 fps step and policy inference: 9114 fps total: 8180 epoch: 1272/2000\n",
+ "fps step: 16121 fps step and policy inference: 9864 fps total: 8745 epoch: 1273/2000\n",
+ "fps step: 16840 fps step and policy inference: 10763 fps total: 9474 epoch: 1274/2000\n",
+ "fps step: 17999 fps step and policy inference: 12475 fps total: 10775 epoch: 1275/2000\n",
+ "fps step: 15356 fps step and policy inference: 8825 fps total: 7741 epoch: 1276/2000\n",
+ "fps step: 17754 fps step and policy inference: 11768 fps total: 10266 epoch: 1277/2000\n",
+ "fps step: 17724 fps step and policy inference: 12037 fps total: 10458 epoch: 1278/2000\n",
+ "fps step: 17195 fps step and policy inference: 11211 fps total: 9840 epoch: 1279/2000\n",
+ "fps step: 15588 fps step and policy inference: 9177 fps total: 8239 epoch: 1280/2000\n",
+ "fps step: 17132 fps step and policy inference: 11045 fps total: 9453 epoch: 1281/2000\n",
+ "fps step: 15071 fps step and policy inference: 8794 fps total: 7885 epoch: 1282/2000\n",
+ "fps step: 15232 fps step and policy inference: 8896 fps total: 7744 epoch: 1283/2000\n",
+ "fps step: 17254 fps step and policy inference: 11300 fps total: 9655 epoch: 1284/2000\n",
+ "fps step: 15238 fps step and policy inference: 8743 fps total: 7624 epoch: 1285/2000\n",
+ "fps step: 15476 fps step and policy inference: 9296 fps total: 8301 epoch: 1286/2000\n",
+ "fps step: 15332 fps step and policy inference: 8884 fps total: 7773 epoch: 1287/2000\n",
+ "fps step: 15589 fps step and policy inference: 9136 fps total: 7955 epoch: 1288/2000\n",
+ "fps step: 17670 fps step and policy inference: 12004 fps total: 10421 epoch: 1289/2000\n",
+ "fps step: 15636 fps step and policy inference: 9215 fps total: 8170 epoch: 1290/2000\n",
+ "fps step: 15691 fps step and policy inference: 9501 fps total: 8497 epoch: 1291/2000\n",
+ "fps step: 17699 fps step and policy inference: 12046 fps total: 10527 epoch: 1292/2000\n",
+ "fps step: 15982 fps step and policy inference: 9532 fps total: 8517 epoch: 1293/2000\n",
+ "fps step: 15544 fps step and policy inference: 9185 fps total: 8146 epoch: 1294/2000\n",
+ "fps step: 15973 fps step and policy inference: 9644 fps total: 8584 epoch: 1295/2000\n",
+ "fps step: 16921 fps step and policy inference: 10901 fps total: 9624 epoch: 1296/2000\n",
+ "fps step: 15286 fps step and policy inference: 8675 fps total: 7597 epoch: 1297/2000\n",
+ "fps step: 15389 fps step and policy inference: 9086 fps total: 8126 epoch: 1298/2000\n",
+ "fps step: 15428 fps step and policy inference: 8769 fps total: 7700 epoch: 1299/2000\n",
+ "fps step: 15218 fps step and policy inference: 8735 fps total: 7753 epoch: 1300/2000\n",
+ "saving next best rewards: [5705.8647]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16500 fps step and policy inference: 10089 fps total: 8660 epoch: 1301/2000\n",
+ "saving next best rewards: [5714.182]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17379 fps step and policy inference: 11684 fps total: 10208 epoch: 1302/2000\n",
+ "fps step: 16526 fps step and policy inference: 10085 fps total: 8753 epoch: 1303/2000\n",
+ "fps step: 18687 fps step and policy inference: 13651 fps total: 11667 epoch: 1304/2000\n",
+ "fps step: 17238 fps step and policy inference: 10912 fps total: 9087 epoch: 1305/2000\n",
+ "fps step: 17641 fps step and policy inference: 11591 fps total: 10152 epoch: 1306/2000\n",
+ "fps step: 16858 fps step and policy inference: 10484 fps total: 8872 epoch: 1307/2000\n",
+ "fps step: 16224 fps step and policy inference: 9876 fps total: 8558 epoch: 1308/2000\n",
+ "fps step: 15783 fps step and policy inference: 9173 fps total: 8036 epoch: 1309/2000\n",
+ "fps step: 17598 fps step and policy inference: 11723 fps total: 10318 epoch: 1310/2000\n",
+ "saving next best rewards: [5725.32]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15962 fps step and policy inference: 9795 fps total: 8699 epoch: 1311/2000\n",
+ "saving next best rewards: [5743.813]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16427 fps step and policy inference: 10166 fps total: 8980 epoch: 1312/2000\n",
+ "saving next best rewards: [5753.348]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16935 fps step and policy inference: 10648 fps total: 8974 epoch: 1313/2000\n",
+ "saving next best rewards: [5766.4243]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17617 fps step and policy inference: 12123 fps total: 10624 epoch: 1314/2000\n",
+ "saving next best rewards: [5777.6567]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17683 fps step and policy inference: 11191 fps total: 9819 epoch: 1315/2000\n",
+ "fps step: 14927 fps step and policy inference: 8346 fps total: 7350 epoch: 1316/2000\n",
+ "fps step: 15187 fps step and policy inference: 8966 fps total: 7712 epoch: 1317/2000\n",
+ "fps step: 16509 fps step and policy inference: 10100 fps total: 8622 epoch: 1318/2000\n",
+ "fps step: 15676 fps step and policy inference: 9053 fps total: 7903 epoch: 1319/2000\n",
+ "fps step: 16112 fps step and policy inference: 9566 fps total: 8244 epoch: 1320/2000\n",
+ "fps step: 14609 fps step and policy inference: 8213 fps total: 7280 epoch: 1321/2000\n",
+ "fps step: 16752 fps step and policy inference: 10583 fps total: 9370 epoch: 1322/2000\n",
+ "fps step: 16200 fps step and policy inference: 9921 fps total: 8542 epoch: 1323/2000\n",
+ "fps step: 15852 fps step and policy inference: 9365 fps total: 8230 epoch: 1324/2000\n",
+ "fps step: 17250 fps step and policy inference: 10759 fps total: 9330 epoch: 1325/2000\n",
+ "fps step: 16922 fps step and policy inference: 10603 fps total: 9134 epoch: 1326/2000\n",
+ "fps step: 15359 fps step and policy inference: 9041 fps total: 7811 epoch: 1327/2000\n",
+ "fps step: 16355 fps step and policy inference: 10407 fps total: 9273 epoch: 1328/2000\n",
+ "fps step: 15061 fps step and policy inference: 8497 fps total: 7467 epoch: 1329/2000\n",
+ "fps step: 16322 fps step and policy inference: 9843 fps total: 8578 epoch: 1330/2000\n",
+ "fps step: 15755 fps step and policy inference: 9279 fps total: 8096 epoch: 1331/2000\n",
+ "fps step: 15556 fps step and policy inference: 9064 fps total: 7911 epoch: 1332/2000\n",
+ "fps step: 15402 fps step and policy inference: 9009 fps total: 8070 epoch: 1333/2000\n",
+ "fps step: 14917 fps step and policy inference: 8345 fps total: 7421 epoch: 1334/2000\n",
+ "fps step: 16270 fps step and policy inference: 9760 fps total: 8536 epoch: 1335/2000\n",
+ "fps step: 15362 fps step and policy inference: 8875 fps total: 7807 epoch: 1336/2000\n",
+ "fps step: 16818 fps step and policy inference: 10414 fps total: 9027 epoch: 1337/2000\n",
+ "fps step: 18063 fps step and policy inference: 12697 fps total: 11039 epoch: 1338/2000\n",
+ "fps step: 16564 fps step and policy inference: 10094 fps total: 8990 epoch: 1339/2000\n",
+ "fps step: 15041 fps step and policy inference: 8629 fps total: 7567 epoch: 1340/2000\n",
+ "fps step: 14955 fps step and policy inference: 8609 fps total: 7575 epoch: 1341/2000\n",
+ "fps step: 14945 fps step and policy inference: 8672 fps total: 7557 epoch: 1342/2000\n",
+ "fps step: 15096 fps step and policy inference: 8974 fps total: 8046 epoch: 1343/2000\n",
+ "fps step: 14763 fps step and policy inference: 8404 fps total: 7402 epoch: 1344/2000\n",
+ "fps step: 15816 fps step and policy inference: 9627 fps total: 8571 epoch: 1345/2000\n",
+ "fps step: 15497 fps step and policy inference: 9275 fps total: 7985 epoch: 1346/2000\n",
+ "fps step: 15384 fps step and policy inference: 9441 fps total: 8450 epoch: 1347/2000\n",
+ "fps step: 16549 fps step and policy inference: 10717 fps total: 9521 epoch: 1348/2000\n",
+ "fps step: 15533 fps step and policy inference: 9341 fps total: 8116 epoch: 1349/2000\n",
+ "fps step: 14920 fps step and policy inference: 8716 fps total: 7644 epoch: 1350/2000\n",
+ "fps step: 14929 fps step and policy inference: 8813 fps total: 7796 epoch: 1351/2000\n",
+ "fps step: 16106 fps step and policy inference: 10127 fps total: 8763 epoch: 1352/2000\n",
+ "fps step: 14511 fps step and policy inference: 8404 fps total: 7499 epoch: 1353/2000\n",
+ "fps step: 15565 fps step and policy inference: 9454 fps total: 8255 epoch: 1354/2000\n",
+ "fps step: 16068 fps step and policy inference: 9399 fps total: 8011 epoch: 1355/2000\n",
+ "fps step: 16876 fps step and policy inference: 10779 fps total: 9461 epoch: 1356/2000\n",
+ "fps step: 17822 fps step and policy inference: 11796 fps total: 9818 epoch: 1357/2000\n",
+ "fps step: 15968 fps step and policy inference: 9498 fps total: 8314 epoch: 1358/2000\n",
+ "fps step: 16343 fps step and policy inference: 10008 fps total: 8914 epoch: 1359/2000\n",
+ "fps step: 15297 fps step and policy inference: 8718 fps total: 7705 epoch: 1360/2000\n",
+ "fps step: 16042 fps step and policy inference: 9588 fps total: 8411 epoch: 1361/2000\n",
+ "fps step: 16036 fps step and policy inference: 9626 fps total: 8282 epoch: 1362/2000\n",
+ "fps step: 15371 fps step and policy inference: 9072 fps total: 8173 epoch: 1363/2000\n",
+ "fps step: 18205 fps step and policy inference: 12832 fps total: 11090 epoch: 1364/2000\n",
+ "fps step: 15486 fps step and policy inference: 9171 fps total: 7988 epoch: 1365/2000\n",
+ "fps step: 15322 fps step and policy inference: 8970 fps total: 7817 epoch: 1366/2000\n",
+ "fps step: 15172 fps step and policy inference: 8564 fps total: 7562 epoch: 1367/2000\n",
+ "fps step: 16185 fps step and policy inference: 9555 fps total: 8251 epoch: 1368/2000\n",
+ "fps step: 15212 fps step and policy inference: 8801 fps total: 7893 epoch: 1369/2000\n",
+ "fps step: 15637 fps step and policy inference: 9267 fps total: 8276 epoch: 1370/2000\n",
+ "fps step: 15795 fps step and policy inference: 9436 fps total: 8439 epoch: 1371/2000\n",
+ "fps step: 17738 fps step and policy inference: 11624 fps total: 9883 epoch: 1372/2000\n",
+ "fps step: 15770 fps step and policy inference: 9142 fps total: 8033 epoch: 1373/2000\n",
+ "fps step: 15452 fps step and policy inference: 8813 fps total: 7589 epoch: 1374/2000\n",
+ "fps step: 15807 fps step and policy inference: 9289 fps total: 8204 epoch: 1375/2000\n",
+ "fps step: 15141 fps step and policy inference: 8521 fps total: 7469 epoch: 1376/2000\n",
+ "fps step: 16989 fps step and policy inference: 10989 fps total: 9623 epoch: 1377/2000\n",
+ "fps step: 16524 fps step and policy inference: 10204 fps total: 8907 epoch: 1378/2000\n",
+ "fps step: 15795 fps step and policy inference: 9407 fps total: 8344 epoch: 1379/2000\n",
+ "fps step: 16080 fps step and policy inference: 9802 fps total: 8749 epoch: 1380/2000\n",
+ "fps step: 15945 fps step and policy inference: 9637 fps total: 8582 epoch: 1381/2000\n",
+ "fps step: 15594 fps step and policy inference: 9011 fps total: 7854 epoch: 1382/2000\n",
+ "fps step: 15102 fps step and policy inference: 8502 fps total: 7524 epoch: 1383/2000\n",
+ "fps step: 16278 fps step and policy inference: 9745 fps total: 8216 epoch: 1384/2000\n",
+ "fps step: 15971 fps step and policy inference: 9856 fps total: 8771 epoch: 1385/2000\n",
+ "fps step: 16548 fps step and policy inference: 10253 fps total: 9092 epoch: 1386/2000\n",
+ "fps step: 16896 fps step and policy inference: 10577 fps total: 9083 epoch: 1387/2000\n",
+ "fps step: 15911 fps step and policy inference: 9506 fps total: 8392 epoch: 1388/2000\n",
+ "saving next best rewards: [5783.9854]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17205 fps step and policy inference: 11103 fps total: 9683 epoch: 1389/2000\n",
+ "saving next best rewards: [5814.4624]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15542 fps step and policy inference: 8987 fps total: 7881 epoch: 1390/2000\n",
+ "saving next best rewards: [5818.9424]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15636 fps step and policy inference: 9107 fps total: 7725 epoch: 1391/2000\n",
+ "saving next best rewards: [5824.4478]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17768 fps step and policy inference: 11817 fps total: 10040 epoch: 1392/2000\n",
+ "fps step: 15783 fps step and policy inference: 9345 fps total: 8134 epoch: 1393/2000\n",
+ "fps step: 15311 fps step and policy inference: 8842 fps total: 8119 epoch: 1394/2000\n",
+ "saving next best rewards: [5852.1675]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17211 fps step and policy inference: 11185 fps total: 9877 epoch: 1395/2000\n",
+ "saving next best rewards: [5861.4224]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 18917 fps step and policy inference: 13576 fps total: 11658 epoch: 1396/2000\n",
+ "fps step: 14837 fps step and policy inference: 8460 fps total: 7602 epoch: 1397/2000\n",
+ "fps step: 16774 fps step and policy inference: 10651 fps total: 9452 epoch: 1398/2000\n",
+ "fps step: 15833 fps step and policy inference: 9289 fps total: 8146 epoch: 1399/2000\n",
+ "fps step: 16614 fps step and policy inference: 10261 fps total: 8893 epoch: 1400/2000\n",
+ "saving next best rewards: [5863.5913]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16073 fps step and policy inference: 9702 fps total: 8437 epoch: 1401/2000\n",
+ "saving next best rewards: [5885.0923]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 18630 fps step and policy inference: 13297 fps total: 11414 epoch: 1402/2000\n",
+ "saving next best rewards: [5906.5146]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15031 fps step and policy inference: 8465 fps total: 7452 epoch: 1403/2000\n",
+ "fps step: 15765 fps step and policy inference: 9254 fps total: 8301 epoch: 1404/2000\n",
+ "fps step: 15696 fps step and policy inference: 9161 fps total: 8092 epoch: 1405/2000\n",
+ "fps step: 17077 fps step and policy inference: 10533 fps total: 9018 epoch: 1406/2000\n",
+ "saving next best rewards: [5917.3955]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15734 fps step and policy inference: 9188 fps total: 7983 epoch: 1407/2000\n",
+ "fps step: 15377 fps step and policy inference: 8864 fps total: 7726 epoch: 1408/2000\n",
+ "saving next best rewards: [5934.855]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17634 fps step and policy inference: 11901 fps total: 10072 epoch: 1409/2000\n",
+ "fps step: 17701 fps step and policy inference: 11625 fps total: 9478 epoch: 1410/2000\n",
+ "fps step: 16860 fps step and policy inference: 10593 fps total: 9127 epoch: 1411/2000\n",
+ "saving next best rewards: [5935.454]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15623 fps step and policy inference: 9224 fps total: 7869 epoch: 1412/2000\n",
+ "fps step: 16056 fps step and policy inference: 9644 fps total: 8398 epoch: 1413/2000\n",
+ "fps step: 16656 fps step and policy inference: 10599 fps total: 9325 epoch: 1414/2000\n",
+ "fps step: 15541 fps step and policy inference: 9020 fps total: 7924 epoch: 1415/2000\n",
+ "fps step: 18620 fps step and policy inference: 11890 fps total: 9915 epoch: 1416/2000\n",
+ "fps step: 15002 fps step and policy inference: 8405 fps total: 7393 epoch: 1417/2000\n",
+ "fps step: 17252 fps step and policy inference: 10886 fps total: 9202 epoch: 1418/2000\n",
+ "fps step: 16606 fps step and policy inference: 10295 fps total: 8812 epoch: 1419/2000\n",
+ "fps step: 15833 fps step and policy inference: 9516 fps total: 8521 epoch: 1420/2000\n",
+ "fps step: 17647 fps step and policy inference: 12036 fps total: 10526 epoch: 1421/2000\n",
+ "fps step: 14910 fps step and policy inference: 8592 fps total: 7704 epoch: 1422/2000\n",
+ "fps step: 15624 fps step and policy inference: 9203 fps total: 8012 epoch: 1423/2000\n",
+ "fps step: 16948 fps step and policy inference: 10939 fps total: 9577 epoch: 1424/2000\n",
+ "fps step: 16079 fps step and policy inference: 9679 fps total: 8269 epoch: 1425/2000\n",
+ "fps step: 15987 fps step and policy inference: 9577 fps total: 8212 epoch: 1426/2000\n",
+ "fps step: 16043 fps step and policy inference: 10160 fps total: 8989 epoch: 1427/2000\n",
+ "fps step: 15018 fps step and policy inference: 8457 fps total: 7410 epoch: 1428/2000\n",
+ "fps step: 15849 fps step and policy inference: 9667 fps total: 8663 epoch: 1429/2000\n",
+ "fps step: 15246 fps step and policy inference: 8781 fps total: 7615 epoch: 1430/2000\n",
+ "fps step: 15013 fps step and policy inference: 8523 fps total: 7465 epoch: 1431/2000\n",
+ "fps step: 16923 fps step and policy inference: 10731 fps total: 9142 epoch: 1432/2000\n",
+ "fps step: 15467 fps step and policy inference: 9055 fps total: 7923 epoch: 1433/2000\n",
+ "fps step: 16087 fps step and policy inference: 9683 fps total: 8358 epoch: 1434/2000\n",
+ "fps step: 15513 fps step and policy inference: 8881 fps total: 7731 epoch: 1435/2000\n",
+ "fps step: 18702 fps step and policy inference: 13535 fps total: 11720 epoch: 1436/2000\n",
+ "fps step: 19317 fps step and policy inference: 12893 fps total: 10892 epoch: 1437/2000\n",
+ "fps step: 17676 fps step and policy inference: 11881 fps total: 10317 epoch: 1438/2000\n",
+ "fps step: 16576 fps step and policy inference: 10258 fps total: 9346 epoch: 1439/2000\n",
+ "fps step: 17677 fps step and policy inference: 11888 fps total: 10666 epoch: 1440/2000\n",
+ "fps step: 18361 fps step and policy inference: 13170 fps total: 11321 epoch: 1441/2000\n",
+ "fps step: 15449 fps step and policy inference: 8938 fps total: 7894 epoch: 1442/2000\n",
+ "fps step: 15985 fps step and policy inference: 9591 fps total: 8235 epoch: 1443/2000\n",
+ "fps step: 16665 fps step and policy inference: 10649 fps total: 9107 epoch: 1444/2000\n",
+ "fps step: 18362 fps step and policy inference: 12805 fps total: 11092 epoch: 1445/2000\n",
+ "fps step: 16870 fps step and policy inference: 10641 fps total: 9202 epoch: 1446/2000\n",
+ "fps step: 17234 fps step and policy inference: 11050 fps total: 9554 epoch: 1447/2000\n",
+ "fps step: 15038 fps step and policy inference: 8532 fps total: 7430 epoch: 1448/2000\n",
+ "fps step: 17418 fps step and policy inference: 11553 fps total: 10100 epoch: 1449/2000\n",
+ "fps step: 15958 fps step and policy inference: 9460 fps total: 8174 epoch: 1450/2000\n",
+ "fps step: 16565 fps step and policy inference: 10264 fps total: 9118 epoch: 1451/2000\n",
+ "fps step: 17501 fps step and policy inference: 11697 fps total: 10106 epoch: 1452/2000\n",
+ "fps step: 16199 fps step and policy inference: 10034 fps total: 9231 epoch: 1453/2000\n",
+ "fps step: 17049 fps step and policy inference: 10784 fps total: 9252 epoch: 1454/2000\n",
+ "fps step: 16547 fps step and policy inference: 10422 fps total: 9244 epoch: 1455/2000\n",
+ "fps step: 15282 fps step and policy inference: 8844 fps total: 7912 epoch: 1456/2000\n",
+ "fps step: 16288 fps step and policy inference: 10047 fps total: 8924 epoch: 1457/2000\n",
+ "fps step: 14973 fps step and policy inference: 8609 fps total: 7652 epoch: 1458/2000\n",
+ "fps step: 15369 fps step and policy inference: 9070 fps total: 8132 epoch: 1459/2000\n",
+ "fps step: 16111 fps step and policy inference: 9551 fps total: 8235 epoch: 1460/2000\n",
+ "fps step: 17875 fps step and policy inference: 12312 fps total: 10746 epoch: 1461/2000\n",
+ "fps step: 16165 fps step and policy inference: 9791 fps total: 8687 epoch: 1462/2000\n",
+ "fps step: 16380 fps step and policy inference: 10110 fps total: 8906 epoch: 1463/2000\n",
+ "fps step: 16001 fps step and policy inference: 9439 fps total: 8158 epoch: 1464/2000\n",
+ "fps step: 16094 fps step and policy inference: 9564 fps total: 8258 epoch: 1465/2000\n",
+ "fps step: 16630 fps step and policy inference: 10452 fps total: 9260 epoch: 1466/2000\n",
+ "fps step: 15736 fps step and policy inference: 9413 fps total: 8374 epoch: 1467/2000\n",
+ "fps step: 15810 fps step and policy inference: 9240 fps total: 8010 epoch: 1468/2000\n",
+ "fps step: 16648 fps step and policy inference: 10262 fps total: 8740 epoch: 1469/2000\n",
+ "fps step: 15078 fps step and policy inference: 8521 fps total: 7475 epoch: 1470/2000\n",
+ "fps step: 15848 fps step and policy inference: 9615 fps total: 8558 epoch: 1471/2000\n",
+ "fps step: 17573 fps step and policy inference: 12035 fps total: 10514 epoch: 1472/2000\n",
+ "fps step: 16763 fps step and policy inference: 10735 fps total: 9541 epoch: 1473/2000\n",
+ "fps step: 17320 fps step and policy inference: 11432 fps total: 10021 epoch: 1474/2000\n",
+ "fps step: 16546 fps step and policy inference: 10076 fps total: 8534 epoch: 1475/2000\n",
+ "fps step: 15866 fps step and policy inference: 9721 fps total: 8871 epoch: 1476/2000\n",
+ "fps step: 14447 fps step and policy inference: 8292 fps total: 7318 epoch: 1477/2000\n",
+ "fps step: 15000 fps step and policy inference: 9016 fps total: 8094 epoch: 1478/2000\n",
+ "fps step: 14540 fps step and policy inference: 8447 fps total: 7358 epoch: 1479/2000\n",
+ "fps step: 14442 fps step and policy inference: 8165 fps total: 7230 epoch: 1480/2000\n",
+ "fps step: 15252 fps step and policy inference: 9108 fps total: 7911 epoch: 1481/2000\n",
+ "fps step: 14897 fps step and policy inference: 8580 fps total: 7563 epoch: 1482/2000\n",
+ "fps step: 15029 fps step and policy inference: 9009 fps total: 8034 epoch: 1483/2000\n",
+ "fps step: 15276 fps step and policy inference: 9041 fps total: 7958 epoch: 1484/2000\n",
+ "fps step: 16306 fps step and policy inference: 10522 fps total: 9340 epoch: 1485/2000\n",
+ "fps step: 17560 fps step and policy inference: 12171 fps total: 10532 epoch: 1486/2000\n",
+ "fps step: 15015 fps step and policy inference: 8937 fps total: 7981 epoch: 1487/2000\n",
+ "fps step: 16643 fps step and policy inference: 10525 fps total: 8987 epoch: 1488/2000\n",
+ "fps step: 17668 fps step and policy inference: 12138 fps total: 10281 epoch: 1489/2000\n",
+ "fps step: 16010 fps step and policy inference: 9963 fps total: 8877 epoch: 1490/2000\n",
+ "fps step: 16044 fps step and policy inference: 9457 fps total: 8186 epoch: 1491/2000\n",
+ "fps step: 16479 fps step and policy inference: 10298 fps total: 9049 epoch: 1492/2000\n",
+ "fps step: 15771 fps step and policy inference: 9215 fps total: 8063 epoch: 1493/2000\n",
+ "fps step: 15883 fps step and policy inference: 9281 fps total: 8583 epoch: 1494/2000\n",
+ "fps step: 17670 fps step and policy inference: 11551 fps total: 9860 epoch: 1495/2000\n",
+ "fps step: 17562 fps step and policy inference: 11302 fps total: 9776 epoch: 1496/2000\n",
+ "fps step: 15857 fps step and policy inference: 9522 fps total: 8449 epoch: 1497/2000\n",
+ "fps step: 17181 fps step and policy inference: 10944 fps total: 9569 epoch: 1498/2000\n",
+ "fps step: 16552 fps step and policy inference: 10191 fps total: 9043 epoch: 1499/2000\n",
+ "fps step: 16627 fps step and policy inference: 10218 fps total: 8828 epoch: 1500/2000\n",
+ "fps step: 15803 fps step and policy inference: 9150 fps total: 7952 epoch: 1501/2000\n",
+ "fps step: 15211 fps step and policy inference: 8771 fps total: 7692 epoch: 1502/2000\n",
+ "fps step: 16163 fps step and policy inference: 9904 fps total: 8532 epoch: 1503/2000\n",
+ "fps step: 16818 fps step and policy inference: 10540 fps total: 9367 epoch: 1504/2000\n",
+ "fps step: 16371 fps step and policy inference: 9935 fps total: 8635 epoch: 1505/2000\n",
+ "fps step: 15125 fps step and policy inference: 8512 fps total: 7401 epoch: 1506/2000\n",
+ "fps step: 15127 fps step and policy inference: 8685 fps total: 7780 epoch: 1507/2000\n",
+ "fps step: 15204 fps step and policy inference: 8601 fps total: 7564 epoch: 1508/2000\n",
+ "fps step: 15681 fps step and policy inference: 9253 fps total: 8074 epoch: 1509/2000\n",
+ "fps step: 15678 fps step and policy inference: 9242 fps total: 8295 epoch: 1510/2000\n",
+ "fps step: 15680 fps step and policy inference: 9379 fps total: 8375 epoch: 1511/2000\n",
+ "fps step: 16703 fps step and policy inference: 10440 fps total: 9293 epoch: 1512/2000\n",
+ "fps step: 14978 fps step and policy inference: 8532 fps total: 7693 epoch: 1513/2000\n",
+ "fps step: 15051 fps step and policy inference: 8537 fps total: 7477 epoch: 1514/2000\n",
+ "fps step: 15919 fps step and policy inference: 9780 fps total: 8699 epoch: 1515/2000\n",
+ "fps step: 16398 fps step and policy inference: 10225 fps total: 9053 epoch: 1516/2000\n",
+ "fps step: 16791 fps step and policy inference: 10420 fps total: 8984 epoch: 1517/2000\n",
+ "fps step: 17092 fps step and policy inference: 10752 fps total: 9370 epoch: 1518/2000\n",
+ "fps step: 16095 fps step and policy inference: 9708 fps total: 8587 epoch: 1519/2000\n",
+ "fps step: 16114 fps step and policy inference: 9723 fps total: 8670 epoch: 1520/2000\n",
+ "fps step: 15299 fps step and policy inference: 8737 fps total: 7648 epoch: 1521/2000\n",
+ "fps step: 15606 fps step and policy inference: 8979 fps total: 7815 epoch: 1522/2000\n",
+ "fps step: 17239 fps step and policy inference: 11307 fps total: 9979 epoch: 1523/2000\n",
+ "fps step: 16211 fps step and policy inference: 9816 fps total: 8633 epoch: 1524/2000\n",
+ "fps step: 15767 fps step and policy inference: 9366 fps total: 8257 epoch: 1525/2000\n",
+ "fps step: 14955 fps step and policy inference: 8396 fps total: 7486 epoch: 1526/2000\n",
+ "fps step: 16200 fps step and policy inference: 9859 fps total: 8524 epoch: 1527/2000\n",
+ "fps step: 16607 fps step and policy inference: 10244 fps total: 8812 epoch: 1528/2000\n",
+ "fps step: 17421 fps step and policy inference: 10972 fps total: 9233 epoch: 1529/2000\n",
+ "fps step: 16712 fps step and policy inference: 10272 fps total: 9132 epoch: 1530/2000\n",
+ "fps step: 15872 fps step and policy inference: 9410 fps total: 8175 epoch: 1531/2000\n",
+ "fps step: 15656 fps step and policy inference: 9086 fps total: 7817 epoch: 1532/2000\n",
+ "fps step: 15728 fps step and policy inference: 9228 fps total: 8036 epoch: 1533/2000\n",
+ "fps step: 16831 fps step and policy inference: 10816 fps total: 9507 epoch: 1534/2000\n",
+ "fps step: 16598 fps step and policy inference: 10370 fps total: 9240 epoch: 1535/2000\n",
+ "fps step: 15617 fps step and policy inference: 9087 fps total: 7915 epoch: 1536/2000\n",
+ "fps step: 16839 fps step and policy inference: 10488 fps total: 8959 epoch: 1537/2000\n",
+ "fps step: 16453 fps step and policy inference: 10008 fps total: 8690 epoch: 1538/2000\n",
+ "fps step: 15875 fps step and policy inference: 9721 fps total: 8715 epoch: 1539/2000\n",
+ "fps step: 16641 fps step and policy inference: 10253 fps total: 8835 epoch: 1540/2000\n",
+ "fps step: 17910 fps step and policy inference: 12301 fps total: 10687 epoch: 1541/2000\n",
+ "fps step: 16487 fps step and policy inference: 10084 fps total: 8731 epoch: 1542/2000\n",
+ "fps step: 16330 fps step and policy inference: 9824 fps total: 8502 epoch: 1543/2000\n",
+ "fps step: 15842 fps step and policy inference: 9552 fps total: 8393 epoch: 1544/2000\n",
+ "fps step: 17035 fps step and policy inference: 10750 fps total: 9466 epoch: 1545/2000\n",
+ "fps step: 15219 fps step and policy inference: 8714 fps total: 7678 epoch: 1546/2000\n",
+ "fps step: 15642 fps step and policy inference: 9134 fps total: 8163 epoch: 1547/2000\n",
+ "fps step: 17616 fps step and policy inference: 11744 fps total: 10308 epoch: 1548/2000\n",
+ "fps step: 16435 fps step and policy inference: 9968 fps total: 8661 epoch: 1549/2000\n",
+ "fps step: 14971 fps step and policy inference: 8743 fps total: 7883 epoch: 1550/2000\n",
+ "fps step: 16125 fps step and policy inference: 9626 fps total: 8582 epoch: 1551/2000\n",
+ "fps step: 17887 fps step and policy inference: 12422 fps total: 10762 epoch: 1552/2000\n",
+ "fps step: 15865 fps step and policy inference: 9392 fps total: 8043 epoch: 1553/2000\n",
+ "fps step: 16031 fps step and policy inference: 9707 fps total: 8289 epoch: 1554/2000\n",
+ "fps step: 15198 fps step and policy inference: 8760 fps total: 7803 epoch: 1555/2000\n",
+ "fps step: 17293 fps step and policy inference: 11403 fps total: 9900 epoch: 1556/2000\n",
+ "fps step: 15685 fps step and policy inference: 9369 fps total: 8116 epoch: 1557/2000\n",
+ "fps step: 17274 fps step and policy inference: 10812 fps total: 9222 epoch: 1558/2000\n",
+ "fps step: 15682 fps step and policy inference: 9312 fps total: 8162 epoch: 1559/2000\n",
+ "fps step: 15164 fps step and policy inference: 8583 fps total: 7366 epoch: 1560/2000\n",
+ "fps step: 17094 fps step and policy inference: 11182 fps total: 9830 epoch: 1561/2000\n",
+ "fps step: 20033 fps step and policy inference: 13825 fps total: 11863 epoch: 1562/2000\n",
+ "fps step: 15572 fps step and policy inference: 9081 fps total: 7964 epoch: 1563/2000\n",
+ "fps step: 15024 fps step and policy inference: 8432 fps total: 7392 epoch: 1564/2000\n",
+ "fps step: 16550 fps step and policy inference: 10157 fps total: 8608 epoch: 1565/2000\n",
+ "fps step: 15860 fps step and policy inference: 9334 fps total: 8326 epoch: 1566/2000\n",
+ "fps step: 15735 fps step and policy inference: 9157 fps total: 7993 epoch: 1567/2000\n",
+ "fps step: 16524 fps step and policy inference: 10228 fps total: 9306 epoch: 1568/2000\n",
+ "fps step: 17454 fps step and policy inference: 11683 fps total: 10237 epoch: 1569/2000\n",
+ "fps step: 15040 fps step and policy inference: 8454 fps total: 7483 epoch: 1570/2000\n",
+ "fps step: 15509 fps step and policy inference: 9143 fps total: 7964 epoch: 1571/2000\n",
+ "fps step: 17300 fps step and policy inference: 11338 fps total: 9947 epoch: 1572/2000\n",
+ "fps step: 17557 fps step and policy inference: 11623 fps total: 10198 epoch: 1573/2000\n",
+ "fps step: 17144 fps step and policy inference: 10663 fps total: 9293 epoch: 1574/2000\n",
+ "fps step: 16038 fps step and policy inference: 9715 fps total: 8683 epoch: 1575/2000\n",
+ "fps step: 16861 fps step and policy inference: 10383 fps total: 8888 epoch: 1576/2000\n",
+ "fps step: 18279 fps step and policy inference: 12456 fps total: 10725 epoch: 1577/2000\n",
+ "fps step: 16757 fps step and policy inference: 10660 fps total: 9430 epoch: 1578/2000\n",
+ "fps step: 16818 fps step and policy inference: 10778 fps total: 9516 epoch: 1579/2000\n",
+ "fps step: 14941 fps step and policy inference: 8315 fps total: 7285 epoch: 1580/2000\n",
+ "fps step: 17441 fps step and policy inference: 11305 fps total: 9549 epoch: 1581/2000\n",
+ "fps step: 15711 fps step and policy inference: 9204 fps total: 8032 epoch: 1582/2000\n",
+ "fps step: 16808 fps step and policy inference: 10203 fps total: 9085 epoch: 1583/2000\n",
+ "fps step: 17035 fps step and policy inference: 10760 fps total: 9466 epoch: 1584/2000\n",
+ "fps step: 15730 fps step and policy inference: 9377 fps total: 8374 epoch: 1585/2000\n",
+ "fps step: 17839 fps step and policy inference: 12142 fps total: 10578 epoch: 1586/2000\n",
+ "fps step: 16988 fps step and policy inference: 10915 fps total: 9625 epoch: 1587/2000\n",
+ "fps step: 17898 fps step and policy inference: 11833 fps total: 10400 epoch: 1588/2000\n",
+ "fps step: 16611 fps step and policy inference: 10193 fps total: 8713 epoch: 1589/2000\n",
+ "fps step: 16131 fps step and policy inference: 9523 fps total: 8311 epoch: 1590/2000\n",
+ "fps step: 15540 fps step and policy inference: 8893 fps total: 7768 epoch: 1591/2000\n",
+ "fps step: 16268 fps step and policy inference: 9952 fps total: 8471 epoch: 1592/2000\n",
+ "fps step: 16712 fps step and policy inference: 10250 fps total: 8810 epoch: 1593/2000\n",
+ "fps step: 15039 fps step and policy inference: 8431 fps total: 7867 epoch: 1594/2000\n",
+ "fps step: 18795 fps step and policy inference: 13326 fps total: 11465 epoch: 1595/2000\n",
+ "fps step: 17554 fps step and policy inference: 11430 fps total: 9965 epoch: 1596/2000\n",
+ "fps step: 14938 fps step and policy inference: 8310 fps total: 7385 epoch: 1597/2000\n",
+ "fps step: 15709 fps step and policy inference: 9388 fps total: 8391 epoch: 1598/2000\n",
+ "fps step: 17563 fps step and policy inference: 11709 fps total: 10045 epoch: 1599/2000\n",
+ "fps step: 16025 fps step and policy inference: 9515 fps total: 8244 epoch: 1600/2000\n",
+ "fps step: 16319 fps step and policy inference: 9828 fps total: 8520 epoch: 1601/2000\n",
+ "fps step: 17701 fps step and policy inference: 12225 fps total: 10653 epoch: 1602/2000\n",
+ "fps step: 16656 fps step and policy inference: 10234 fps total: 8830 epoch: 1603/2000\n",
+ "fps step: 15334 fps step and policy inference: 8782 fps total: 7687 epoch: 1604/2000\n",
+ "fps step: 16096 fps step and policy inference: 9874 fps total: 8806 epoch: 1605/2000\n",
+ "fps step: 15507 fps step and policy inference: 9051 fps total: 7942 epoch: 1606/2000\n",
+ "fps step: 16509 fps step and policy inference: 10319 fps total: 9072 epoch: 1607/2000\n",
+ "fps step: 17013 fps step and policy inference: 10989 fps total: 9255 epoch: 1608/2000\n",
+ "fps step: 14435 fps step and policy inference: 8091 fps total: 7285 epoch: 1609/2000\n",
+ "fps step: 15257 fps step and policy inference: 8776 fps total: 7691 epoch: 1610/2000\n",
+ "fps step: 17241 fps step and policy inference: 11254 fps total: 9896 epoch: 1611/2000\n",
+ "fps step: 15994 fps step and policy inference: 9483 fps total: 8237 epoch: 1612/2000\n",
+ "fps step: 16284 fps step and policy inference: 10044 fps total: 8867 epoch: 1613/2000\n",
+ "fps step: 15948 fps step and policy inference: 9310 fps total: 7956 epoch: 1614/2000\n",
+ "fps step: 15433 fps step and policy inference: 9158 fps total: 7968 epoch: 1615/2000\n",
+ "fps step: 15443 fps step and policy inference: 9063 fps total: 7956 epoch: 1616/2000\n",
+ "fps step: 16415 fps step and policy inference: 10565 fps total: 9354 epoch: 1617/2000\n",
+ "fps step: 15961 fps step and policy inference: 9951 fps total: 8725 epoch: 1618/2000\n",
+ "fps step: 15254 fps step and policy inference: 8816 fps total: 7706 epoch: 1619/2000\n",
+ "fps step: 15319 fps step and policy inference: 8824 fps total: 7715 epoch: 1620/2000\n",
+ "fps step: 15515 fps step and policy inference: 8903 fps total: 7763 epoch: 1621/2000\n",
+ "fps step: 16010 fps step and policy inference: 9356 fps total: 8129 epoch: 1622/2000\n",
+ "fps step: 16614 fps step and policy inference: 10050 fps total: 8520 epoch: 1623/2000\n",
+ "fps step: 15603 fps step and policy inference: 9097 fps total: 7877 epoch: 1624/2000\n",
+ "fps step: 18049 fps step and policy inference: 12131 fps total: 10570 epoch: 1625/2000\n",
+ "fps step: 17041 fps step and policy inference: 10644 fps total: 9226 epoch: 1626/2000\n",
+ "fps step: 17221 fps step and policy inference: 10903 fps total: 9031 epoch: 1627/2000\n",
+ "fps step: 16123 fps step and policy inference: 9733 fps total: 8715 epoch: 1628/2000\n",
+ "fps step: 15362 fps step and policy inference: 8709 fps total: 7610 epoch: 1629/2000\n",
+ "fps step: 16798 fps step and policy inference: 10791 fps total: 9482 epoch: 1630/2000\n",
+ "fps step: 16430 fps step and policy inference: 9823 fps total: 8517 epoch: 1631/2000\n",
+ "fps step: 16903 fps step and policy inference: 10593 fps total: 9089 epoch: 1632/2000\n",
+ "fps step: 14955 fps step and policy inference: 8447 fps total: 7470 epoch: 1633/2000\n",
+ "fps step: 18055 fps step and policy inference: 12512 fps total: 10902 epoch: 1634/2000\n",
+ "fps step: 15662 fps step and policy inference: 9060 fps total: 7725 epoch: 1635/2000\n",
+ "saving next best rewards: [5946.7793]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15113 fps step and policy inference: 8624 fps total: 7584 epoch: 1636/2000\n",
+ "saving next best rewards: [5959.422]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15749 fps step and policy inference: 9185 fps total: 7884 epoch: 1637/2000\n",
+ "saving next best rewards: [5962.838]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 18029 fps step and policy inference: 11868 fps total: 9971 epoch: 1638/2000\n",
+ "saving next best rewards: [5974.355]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 14865 fps step and policy inference: 8310 fps total: 7346 epoch: 1639/2000\n",
+ "fps step: 14951 fps step and policy inference: 8460 fps total: 7562 epoch: 1640/2000\n",
+ "fps step: 16245 fps step and policy inference: 10067 fps total: 8992 epoch: 1641/2000\n",
+ "fps step: 14813 fps step and policy inference: 8315 fps total: 7381 epoch: 1642/2000\n",
+ "fps step: 15564 fps step and policy inference: 9041 fps total: 7968 epoch: 1643/2000\n",
+ "fps step: 16320 fps step and policy inference: 9621 fps total: 8346 epoch: 1644/2000\n",
+ "fps step: 15843 fps step and policy inference: 9383 fps total: 8277 epoch: 1645/2000\n",
+ "fps step: 17980 fps step and policy inference: 12179 fps total: 10563 epoch: 1646/2000\n",
+ "fps step: 18202 fps step and policy inference: 12279 fps total: 10387 epoch: 1647/2000\n",
+ "fps step: 15977 fps step and policy inference: 9425 fps total: 8050 epoch: 1648/2000\n",
+ "fps step: 15374 fps step and policy inference: 8766 fps total: 7735 epoch: 1649/2000\n",
+ "fps step: 16617 fps step and policy inference: 10247 fps total: 8939 epoch: 1650/2000\n",
+ "fps step: 16096 fps step and policy inference: 9668 fps total: 8347 epoch: 1651/2000\n",
+ "fps step: 15194 fps step and policy inference: 8642 fps total: 7560 epoch: 1652/2000\n",
+ "fps step: 16932 fps step and policy inference: 10679 fps total: 9192 epoch: 1653/2000\n",
+ "fps step: 15958 fps step and policy inference: 9607 fps total: 8437 epoch: 1654/2000\n",
+ "fps step: 16782 fps step and policy inference: 10414 fps total: 8749 epoch: 1655/2000\n",
+ "fps step: 17412 fps step and policy inference: 11179 fps total: 9792 epoch: 1656/2000\n",
+ "fps step: 16244 fps step and policy inference: 9684 fps total: 8327 epoch: 1657/2000\n",
+ "fps step: 16103 fps step and policy inference: 9782 fps total: 8731 epoch: 1658/2000\n",
+ "fps step: 16577 fps step and policy inference: 10233 fps total: 8905 epoch: 1659/2000\n",
+ "fps step: 17562 fps step and policy inference: 11072 fps total: 9811 epoch: 1660/2000\n",
+ "fps step: 18457 fps step and policy inference: 13031 fps total: 11321 epoch: 1661/2000\n",
+ "fps step: 15018 fps step and policy inference: 8620 fps total: 7729 epoch: 1662/2000\n",
+ "fps step: 15492 fps step and policy inference: 9046 fps total: 8066 epoch: 1663/2000\n",
+ "fps step: 17363 fps step and policy inference: 11368 fps total: 9992 epoch: 1664/2000\n",
+ "fps step: 15982 fps step and policy inference: 9500 fps total: 8154 epoch: 1665/2000\n",
+ "fps step: 17610 fps step and policy inference: 11021 fps total: 9340 epoch: 1666/2000\n",
+ "fps step: 17105 fps step and policy inference: 10878 fps total: 9281 epoch: 1667/2000\n",
+ "fps step: 17555 fps step and policy inference: 11884 fps total: 10222 epoch: 1668/2000\n",
+ "fps step: 15493 fps step and policy inference: 8873 fps total: 7796 epoch: 1669/2000\n",
+ "fps step: 18140 fps step and policy inference: 12637 fps total: 10977 epoch: 1670/2000\n",
+ "fps step: 16887 fps step and policy inference: 10387 fps total: 8824 epoch: 1671/2000\n",
+ "fps step: 15160 fps step and policy inference: 8502 fps total: 7473 epoch: 1672/2000\n",
+ "fps step: 15630 fps step and policy inference: 9208 fps total: 7915 epoch: 1673/2000\n",
+ "fps step: 15356 fps step and policy inference: 8945 fps total: 7970 epoch: 1674/2000\n",
+ "fps step: 17011 fps step and policy inference: 10965 fps total: 9684 epoch: 1675/2000\n",
+ "fps step: 15638 fps step and policy inference: 9106 fps total: 8159 epoch: 1676/2000\n",
+ "fps step: 15071 fps step and policy inference: 8501 fps total: 7440 epoch: 1677/2000\n",
+ "fps step: 15706 fps step and policy inference: 9138 fps total: 7876 epoch: 1678/2000\n",
+ "fps step: 16409 fps step and policy inference: 10095 fps total: 8710 epoch: 1679/2000\n",
+ "fps step: 15506 fps step and policy inference: 9162 fps total: 8208 epoch: 1680/2000\n",
+ "fps step: 15719 fps step and policy inference: 9286 fps total: 8113 epoch: 1681/2000\n",
+ "fps step: 16227 fps step and policy inference: 9730 fps total: 8636 epoch: 1682/2000\n",
+ "fps step: 17605 fps step and policy inference: 11797 fps total: 10319 epoch: 1683/2000\n",
+ "fps step: 18224 fps step and policy inference: 12750 fps total: 10990 epoch: 1684/2000\n",
+ "fps step: 17133 fps step and policy inference: 10718 fps total: 9058 epoch: 1685/2000\n",
+ "fps step: 17276 fps step and policy inference: 11190 fps total: 9819 epoch: 1686/2000\n",
+ "fps step: 16302 fps step and policy inference: 9721 fps total: 8693 epoch: 1687/2000\n",
+ "fps step: 18051 fps step and policy inference: 12336 fps total: 10772 epoch: 1688/2000\n",
+ "fps step: 17278 fps step and policy inference: 10779 fps total: 9179 epoch: 1689/2000\n",
+ "fps step: 15850 fps step and policy inference: 9573 fps total: 8580 epoch: 1690/2000\n",
+ "fps step: 15560 fps step and policy inference: 9107 fps total: 8065 epoch: 1691/2000\n",
+ "fps step: 14980 fps step and policy inference: 8495 fps total: 7446 epoch: 1692/2000\n",
+ "fps step: 16189 fps step and policy inference: 9704 fps total: 8529 epoch: 1693/2000\n",
+ "fps step: 15582 fps step and policy inference: 8959 fps total: 7793 epoch: 1694/2000\n",
+ "fps step: 17585 fps step and policy inference: 11747 fps total: 10269 epoch: 1695/2000\n",
+ "fps step: 18292 fps step and policy inference: 12623 fps total: 10973 epoch: 1696/2000\n",
+ "fps step: 15958 fps step and policy inference: 9319 fps total: 8078 epoch: 1697/2000\n",
+ "fps step: 15314 fps step and policy inference: 8635 fps total: 7589 epoch: 1698/2000\n",
+ "fps step: 16268 fps step and policy inference: 9971 fps total: 8890 epoch: 1699/2000\n",
+ "fps step: 17558 fps step and policy inference: 11149 fps total: 9483 epoch: 1700/2000\n",
+ "fps step: 16116 fps step and policy inference: 9812 fps total: 8750 epoch: 1701/2000\n",
+ "fps step: 17873 fps step and policy inference: 12351 fps total: 10809 epoch: 1702/2000\n",
+ "fps step: 16148 fps step and policy inference: 9615 fps total: 8271 epoch: 1703/2000\n",
+ "fps step: 15937 fps step and policy inference: 9371 fps total: 7945 epoch: 1704/2000\n",
+ "fps step: 17286 fps step and policy inference: 11289 fps total: 9920 epoch: 1705/2000\n",
+ "fps step: 16043 fps step and policy inference: 9618 fps total: 8392 epoch: 1706/2000\n",
+ "fps step: 16692 fps step and policy inference: 10412 fps total: 9094 epoch: 1707/2000\n",
+ "fps step: 16700 fps step and policy inference: 10185 fps total: 8718 epoch: 1708/2000\n",
+ "fps step: 15720 fps step and policy inference: 9238 fps total: 8174 epoch: 1709/2000\n",
+ "fps step: 16741 fps step and policy inference: 10679 fps total: 9384 epoch: 1710/2000\n",
+ "fps step: 16364 fps step and policy inference: 10072 fps total: 8916 epoch: 1711/2000\n",
+ "fps step: 15638 fps step and policy inference: 9104 fps total: 7864 epoch: 1712/2000\n",
+ "fps step: 15354 fps step and policy inference: 8708 fps total: 7671 epoch: 1713/2000\n",
+ "fps step: 16764 fps step and policy inference: 10645 fps total: 9372 epoch: 1714/2000\n",
+ "fps step: 15485 fps step and policy inference: 8927 fps total: 7858 epoch: 1715/2000\n",
+ "fps step: 15298 fps step and policy inference: 8833 fps total: 7942 epoch: 1716/2000\n",
+ "fps step: 15657 fps step and policy inference: 9116 fps total: 7861 epoch: 1717/2000\n",
+ "fps step: 17296 fps step and policy inference: 11157 fps total: 9836 epoch: 1718/2000\n",
+ "fps step: 15051 fps step and policy inference: 8428 fps total: 7268 epoch: 1719/2000\n",
+ "fps step: 17726 fps step and policy inference: 11590 fps total: 9974 epoch: 1720/2000\n",
+ "fps step: 16892 fps step and policy inference: 10512 fps total: 8947 epoch: 1721/2000\n",
+ "fps step: 15418 fps step and policy inference: 9088 fps total: 8408 epoch: 1722/2000\n",
+ "fps step: 16578 fps step and policy inference: 10459 fps total: 8995 epoch: 1723/2000\n",
+ "fps step: 14587 fps step and policy inference: 8255 fps total: 7264 epoch: 1724/2000\n",
+ "fps step: 14712 fps step and policy inference: 8335 fps total: 7364 epoch: 1725/2000\n",
+ "fps step: 15733 fps step and policy inference: 9487 fps total: 8110 epoch: 1726/2000\n",
+ "fps step: 16131 fps step and policy inference: 10229 fps total: 9009 epoch: 1727/2000\n",
+ "fps step: 16917 fps step and policy inference: 11165 fps total: 9792 epoch: 1728/2000\n",
+ "fps step: 16967 fps step and policy inference: 10979 fps total: 9433 epoch: 1729/2000\n",
+ "fps step: 15455 fps step and policy inference: 9064 fps total: 7878 epoch: 1730/2000\n",
+ "fps step: 15333 fps step and policy inference: 9057 fps total: 7927 epoch: 1731/2000\n",
+ "fps step: 15129 fps step and policy inference: 8766 fps total: 7672 epoch: 1732/2000\n",
+ "fps step: 16718 fps step and policy inference: 10709 fps total: 9507 epoch: 1733/2000\n",
+ "fps step: 16705 fps step and policy inference: 10510 fps total: 9185 epoch: 1734/2000\n",
+ "fps step: 15700 fps step and policy inference: 9215 fps total: 7822 epoch: 1735/2000\n",
+ "fps step: 16868 fps step and policy inference: 10486 fps total: 9015 epoch: 1736/2000\n",
+ "fps step: 15225 fps step and policy inference: 8607 fps total: 7531 epoch: 1737/2000\n",
+ "fps step: 16422 fps step and policy inference: 9895 fps total: 8482 epoch: 1738/2000\n",
+ "fps step: 17378 fps step and policy inference: 11293 fps total: 9982 epoch: 1739/2000\n",
+ "fps step: 17351 fps step and policy inference: 11258 fps total: 9677 epoch: 1740/2000\n",
+ "fps step: 16166 fps step and policy inference: 9878 fps total: 8805 epoch: 1741/2000\n",
+ "fps step: 15390 fps step and policy inference: 8702 fps total: 7666 epoch: 1742/2000\n",
+ "fps step: 17968 fps step and policy inference: 12292 fps total: 10688 epoch: 1743/2000\n",
+ "fps step: 16467 fps step and policy inference: 10395 fps total: 9178 epoch: 1744/2000\n",
+ "fps step: 16764 fps step and policy inference: 10635 fps total: 9366 epoch: 1745/2000\n",
+ "fps step: 17846 fps step and policy inference: 12025 fps total: 10506 epoch: 1746/2000\n",
+ "fps step: 16157 fps step and policy inference: 9955 fps total: 8842 epoch: 1747/2000\n",
+ "fps step: 15596 fps step and policy inference: 9151 fps total: 7959 epoch: 1748/2000\n",
+ "fps step: 17756 fps step and policy inference: 11752 fps total: 10301 epoch: 1749/2000\n",
+ "fps step: 16913 fps step and policy inference: 10387 fps total: 9060 epoch: 1750/2000\n",
+ "saving next best rewards: [5978.291]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15492 fps step and policy inference: 8901 fps total: 7765 epoch: 1751/2000\n",
+ "saving next best rewards: [5996.822]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16131 fps step and policy inference: 9594 fps total: 8497 epoch: 1752/2000\n",
+ "fps step: 16025 fps step and policy inference: 9698 fps total: 8649 epoch: 1753/2000\n",
+ "fps step: 16978 fps step and policy inference: 10680 fps total: 9104 epoch: 1754/2000\n",
+ "fps step: 15887 fps step and policy inference: 9279 fps total: 7853 epoch: 1755/2000\n",
+ "fps step: 17379 fps step and policy inference: 11380 fps total: 9896 epoch: 1756/2000\n",
+ "fps step: 14903 fps step and policy inference: 8455 fps total: 7537 epoch: 1757/2000\n",
+ "fps step: 17195 fps step and policy inference: 10804 fps total: 9083 epoch: 1758/2000\n",
+ "saving next best rewards: [6015.5825]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15947 fps step and policy inference: 9471 fps total: 8483 epoch: 1759/2000\n",
+ "saving next best rewards: [6037.7803]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16456 fps step and policy inference: 10405 fps total: 9264 epoch: 1760/2000\n",
+ "fps step: 16963 fps step and policy inference: 10449 fps total: 8890 epoch: 1761/2000\n",
+ "fps step: 14983 fps step and policy inference: 8377 fps total: 7367 epoch: 1762/2000\n",
+ "fps step: 15934 fps step and policy inference: 9304 fps total: 7980 epoch: 1763/2000\n",
+ "fps step: 15066 fps step and policy inference: 8558 fps total: 7541 epoch: 1764/2000\n",
+ "fps step: 16036 fps step and policy inference: 9467 fps total: 8210 epoch: 1765/2000\n",
+ "saving next best rewards: [6043.9404]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17062 fps step and policy inference: 10844 fps total: 9119 epoch: 1766/2000\n",
+ "saving next best rewards: [6056.987]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 17094 fps step and policy inference: 11186 fps total: 9832 epoch: 1767/2000\n",
+ "fps step: 17128 fps step and policy inference: 11195 fps total: 9840 epoch: 1768/2000\n",
+ "fps step: 16540 fps step and policy inference: 10255 fps total: 9060 epoch: 1769/2000\n",
+ "fps step: 15367 fps step and policy inference: 8730 fps total: 7553 epoch: 1770/2000\n",
+ "saving next best rewards: [6066.9556]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16859 fps step and policy inference: 10645 fps total: 9401 epoch: 1771/2000\n",
+ "fps step: 17317 fps step and policy inference: 11301 fps total: 9934 epoch: 1772/2000\n",
+ "saving next best rewards: [6084.475]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 18326 fps step and policy inference: 12916 fps total: 11179 epoch: 1773/2000\n",
+ "saving next best rewards: [6102.375]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 16568 fps step and policy inference: 10324 fps total: 8968 epoch: 1774/2000\n",
+ "saving next best rewards: [6113.601]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15969 fps step and policy inference: 9506 fps total: 8222 epoch: 1775/2000\n",
+ "saving next best rewards: [6123.4243]\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/Ant-v4_envpool.pth'\n",
+ "fps step: 15781 fps step and policy inference: 9225 fps total: 8086 epoch: 1776/2000\n",
+ "fps step: 15195 fps step and policy inference: 8597 fps total: 7620 epoch: 1777/2000\n",
+ "fps step: 17035 fps step and policy inference: 11089 fps total: 9766 epoch: 1778/2000\n",
+ "fps step: 15579 fps step and policy inference: 8993 fps total: 7691 epoch: 1779/2000\n",
+ "fps step: 15974 fps step and policy inference: 9354 fps total: 8016 epoch: 1780/2000\n",
+ "fps step: 16076 fps step and policy inference: 9793 fps total: 8752 epoch: 1781/2000\n",
+ "fps step: 17703 fps step and policy inference: 11926 fps total: 10307 epoch: 1782/2000\n",
+ "fps step: 16191 fps step and policy inference: 9638 fps total: 8325 epoch: 1783/2000\n",
+ "fps step: 17742 fps step and policy inference: 11904 fps total: 10439 epoch: 1784/2000\n",
+ "fps step: 15581 fps step and policy inference: 9193 fps total: 8209 epoch: 1785/2000\n",
+ "fps step: 15775 fps step and policy inference: 9239 fps total: 8045 epoch: 1786/2000\n",
+ "fps step: 15743 fps step and policy inference: 9029 fps total: 7852 epoch: 1787/2000\n",
+ "fps step: 15245 fps step and policy inference: 8735 fps total: 7669 epoch: 1788/2000\n",
+ "fps step: 16823 fps step and policy inference: 10507 fps total: 8990 epoch: 1789/2000\n",
+ "fps step: 15505 fps step and policy inference: 9038 fps total: 7668 epoch: 1790/2000\n",
+ "fps step: 14764 fps step and policy inference: 8332 fps total: 7189 epoch: 1791/2000\n",
+ "fps step: 15276 fps step and policy inference: 9134 fps total: 8218 epoch: 1792/2000\n",
+ "fps step: 15619 fps step and policy inference: 9227 fps total: 8034 epoch: 1793/2000\n",
+ "fps step: 15127 fps step and policy inference: 8623 fps total: 7555 epoch: 1794/2000\n",
+ "fps step: 16117 fps step and policy inference: 10077 fps total: 9217 epoch: 1795/2000\n",
+ "fps step: 17547 fps step and policy inference: 11651 fps total: 10180 epoch: 1796/2000\n",
+ "fps step: 16817 fps step and policy inference: 10415 fps total: 9050 epoch: 1797/2000\n",
+ "fps step: 14662 fps step and policy inference: 8304 fps total: 7298 epoch: 1798/2000\n",
+ "fps step: 15778 fps step and policy inference: 9479 fps total: 8275 epoch: 1799/2000\n",
+ "fps step: 14772 fps step and policy inference: 8385 fps total: 7372 epoch: 1800/2000\n",
+ "fps step: 16224 fps step and policy inference: 10094 fps total: 8678 epoch: 1801/2000\n",
+ "fps step: 16710 fps step and policy inference: 10591 fps total: 9270 epoch: 1802/2000\n",
+ "fps step: 15048 fps step and policy inference: 8924 fps total: 8014 epoch: 1803/2000\n",
+ "fps step: 15856 fps step and policy inference: 9651 fps total: 8347 epoch: 1804/2000\n",
+ "fps step: 16732 fps step and policy inference: 10952 fps total: 9508 epoch: 1805/2000\n",
+ "fps step: 16495 fps step and policy inference: 10917 fps total: 9625 epoch: 1806/2000\n",
+ "fps step: 17427 fps step and policy inference: 11777 fps total: 9968 epoch: 1807/2000\n",
+ "fps step: 15266 fps step and policy inference: 8884 fps total: 7762 epoch: 1808/2000\n",
+ "fps step: 17591 fps step and policy inference: 12357 fps total: 10626 epoch: 1809/2000\n",
+ "fps step: 14557 fps step and policy inference: 8257 fps total: 7287 epoch: 1810/2000\n",
+ "fps step: 15659 fps step and policy inference: 9391 fps total: 8348 epoch: 1811/2000\n",
+ "fps step: 15888 fps step and policy inference: 9670 fps total: 8646 epoch: 1812/2000\n",
+ "fps step: 16262 fps step and policy inference: 9832 fps total: 8567 epoch: 1813/2000\n",
+ "fps step: 15126 fps step and policy inference: 8446 fps total: 7432 epoch: 1814/2000\n",
+ "fps step: 18301 fps step and policy inference: 12266 fps total: 10302 epoch: 1815/2000\n",
+ "fps step: 15976 fps step and policy inference: 9352 fps total: 8089 epoch: 1816/2000\n",
+ "fps step: 17693 fps step and policy inference: 11527 fps total: 9823 epoch: 1817/2000\n",
+ "fps step: 15136 fps step and policy inference: 8513 fps total: 7520 epoch: 1818/2000\n",
+ "fps step: 16033 fps step and policy inference: 9528 fps total: 8193 epoch: 1819/2000\n",
+ "fps step: 15983 fps step and policy inference: 9345 fps total: 8065 epoch: 1820/2000\n",
+ "fps step: 18431 fps step and policy inference: 12478 fps total: 10825 epoch: 1821/2000\n",
+ "fps step: 18031 fps step and policy inference: 12187 fps total: 10387 epoch: 1822/2000\n",
+ "fps step: 15918 fps step and policy inference: 9661 fps total: 8587 epoch: 1823/2000\n",
+ "fps step: 16284 fps step and policy inference: 9696 fps total: 8231 epoch: 1824/2000\n",
+ "fps step: 15080 fps step and policy inference: 8449 fps total: 7392 epoch: 1825/2000\n",
+ "fps step: 15700 fps step and policy inference: 9221 fps total: 8181 epoch: 1826/2000\n",
+ "fps step: 16937 fps step and policy inference: 11131 fps total: 9718 epoch: 1827/2000\n",
+ "fps step: 15021 fps step and policy inference: 8674 fps total: 7807 epoch: 1828/2000\n",
+ "fps step: 17578 fps step and policy inference: 11513 fps total: 9684 epoch: 1829/2000\n",
+ "fps step: 17476 fps step and policy inference: 11783 fps total: 10285 epoch: 1830/2000\n",
+ "fps step: 18479 fps step and policy inference: 12714 fps total: 11035 epoch: 1831/2000\n",
+ "fps step: 16669 fps step and policy inference: 10250 fps total: 9030 epoch: 1832/2000\n",
+ "fps step: 16689 fps step and policy inference: 10258 fps total: 9174 epoch: 1833/2000\n",
+ "fps step: 16498 fps step and policy inference: 10192 fps total: 8882 epoch: 1834/2000\n",
+ "fps step: 16024 fps step and policy inference: 9598 fps total: 8550 epoch: 1835/2000\n",
+ "fps step: 16546 fps step and policy inference: 10250 fps total: 9013 epoch: 1836/2000\n",
+ "fps step: 17144 fps step and policy inference: 11161 fps total: 9804 epoch: 1837/2000\n",
+ "fps step: 15965 fps step and policy inference: 9540 fps total: 8345 epoch: 1838/2000\n",
+ "fps step: 17591 fps step and policy inference: 11430 fps total: 9722 epoch: 1839/2000\n",
+ "fps step: 17445 fps step and policy inference: 11675 fps total: 10212 epoch: 1840/2000\n",
+ "fps step: 16871 fps step and policy inference: 10448 fps total: 8925 epoch: 1841/2000\n",
+ "fps step: 15864 fps step and policy inference: 9587 fps total: 8577 epoch: 1842/2000\n",
+ "fps step: 17216 fps step and policy inference: 11370 fps total: 9982 epoch: 1843/2000\n",
+ "fps step: 17095 fps step and policy inference: 10927 fps total: 10013 epoch: 1844/2000\n",
+ "fps step: 16856 fps step and policy inference: 10439 fps total: 8896 epoch: 1845/2000\n",
+ "fps step: 14693 fps step and policy inference: 8090 fps total: 7173 epoch: 1846/2000\n",
+ "fps step: 15393 fps step and policy inference: 8822 fps total: 7732 epoch: 1847/2000\n",
+ "fps step: 16470 fps step and policy inference: 10259 fps total: 9072 epoch: 1848/2000\n",
+ "fps step: 16754 fps step and policy inference: 10405 fps total: 8962 epoch: 1849/2000\n",
+ "fps step: 17266 fps step and policy inference: 11092 fps total: 9736 epoch: 1850/2000\n",
+ "fps step: 16132 fps step and policy inference: 9446 fps total: 8177 epoch: 1851/2000\n",
+ "fps step: 16753 fps step and policy inference: 10553 fps total: 9274 epoch: 1852/2000\n",
+ "fps step: 17670 fps step and policy inference: 11691 fps total: 9931 epoch: 1853/2000\n",
+ "fps step: 15613 fps step and policy inference: 8960 fps total: 7897 epoch: 1854/2000\n",
+ "fps step: 15175 fps step and policy inference: 8637 fps total: 7493 epoch: 1855/2000\n",
+ "fps step: 16026 fps step and policy inference: 9653 fps total: 8407 epoch: 1856/2000\n",
+ "fps step: 15475 fps step and policy inference: 8810 fps total: 7692 epoch: 1857/2000\n",
+ "fps step: 17112 fps step and policy inference: 10940 fps total: 9640 epoch: 1858/2000\n",
+ "fps step: 17497 fps step and policy inference: 11202 fps total: 9542 epoch: 1859/2000\n",
+ "fps step: 17540 fps step and policy inference: 11735 fps total: 10295 epoch: 1860/2000\n",
+ "fps step: 16108 fps step and policy inference: 9661 fps total: 8332 epoch: 1861/2000\n",
+ "fps step: 15290 fps step and policy inference: 8673 fps total: 7620 epoch: 1862/2000\n",
+ "fps step: 18631 fps step and policy inference: 13435 fps total: 11554 epoch: 1863/2000\n",
+ "fps step: 15895 fps step and policy inference: 9554 fps total: 8454 epoch: 1864/2000\n",
+ "fps step: 16261 fps step and policy inference: 10074 fps total: 8950 epoch: 1865/2000\n",
+ "fps step: 15254 fps step and policy inference: 8781 fps total: 7850 epoch: 1866/2000\n",
+ "fps step: 17073 fps step and policy inference: 10884 fps total: 9181 epoch: 1867/2000\n",
+ "fps step: 16948 fps step and policy inference: 10704 fps total: 9199 epoch: 1868/2000\n",
+ "fps step: 16012 fps step and policy inference: 9766 fps total: 8733 epoch: 1869/2000\n",
+ "fps step: 15792 fps step and policy inference: 9223 fps total: 8036 epoch: 1870/2000\n",
+ "fps step: 16752 fps step and policy inference: 10140 fps total: 8709 epoch: 1871/2000\n",
+ "fps step: 16187 fps step and policy inference: 9711 fps total: 8360 epoch: 1872/2000\n",
+ "fps step: 15826 fps step and policy inference: 9226 fps total: 8131 epoch: 1873/2000\n",
+ "fps step: 17022 fps step and policy inference: 10880 fps total: 9484 epoch: 1874/2000\n",
+ "fps step: 17311 fps step and policy inference: 11367 fps total: 9898 epoch: 1875/2000\n",
+ "fps step: 16136 fps step and policy inference: 9462 fps total: 8213 epoch: 1876/2000\n",
+ "fps step: 15583 fps step and policy inference: 9131 fps total: 8150 epoch: 1877/2000\n",
+ "fps step: 17719 fps step and policy inference: 11861 fps total: 10105 epoch: 1878/2000\n",
+ "fps step: 18247 fps step and policy inference: 12355 fps total: 10709 epoch: 1879/2000\n",
+ "fps step: 17271 fps step and policy inference: 10916 fps total: 9257 epoch: 1880/2000\n",
+ "fps step: 16814 fps step and policy inference: 10826 fps total: 9478 epoch: 1881/2000\n",
+ "fps step: 14798 fps step and policy inference: 8348 fps total: 7343 epoch: 1882/2000\n",
+ "fps step: 17997 fps step and policy inference: 12597 fps total: 10925 epoch: 1883/2000\n",
+ "fps step: 15873 fps step and policy inference: 9641 fps total: 8576 epoch: 1884/2000\n",
+ "fps step: 14688 fps step and policy inference: 8183 fps total: 7113 epoch: 1885/2000\n",
+ "fps step: 15685 fps step and policy inference: 9512 fps total: 8284 epoch: 1886/2000\n",
+ "fps step: 16629 fps step and policy inference: 10507 fps total: 9354 epoch: 1887/2000\n",
+ "fps step: 15166 fps step and policy inference: 8543 fps total: 7503 epoch: 1888/2000\n",
+ "fps step: 15025 fps step and policy inference: 8438 fps total: 7336 epoch: 1889/2000\n",
+ "fps step: 16405 fps step and policy inference: 9902 fps total: 8560 epoch: 1890/2000\n",
+ "fps step: 17111 fps step and policy inference: 11070 fps total: 9452 epoch: 1891/2000\n",
+ "fps step: 16415 fps step and policy inference: 10099 fps total: 9012 epoch: 1892/2000\n",
+ "fps step: 16181 fps step and policy inference: 9693 fps total: 8408 epoch: 1893/2000\n",
+ "fps step: 15579 fps step and policy inference: 9014 fps total: 7828 epoch: 1894/2000\n",
+ "fps step: 15528 fps step and policy inference: 8960 fps total: 7666 epoch: 1895/2000\n",
+ "fps step: 17402 fps step and policy inference: 11488 fps total: 10326 epoch: 1896/2000\n",
+ "fps step: 15249 fps step and policy inference: 8646 fps total: 7561 epoch: 1897/2000\n",
+ "fps step: 15321 fps step and policy inference: 8746 fps total: 7894 epoch: 1898/2000\n",
+ "fps step: 14972 fps step and policy inference: 8500 fps total: 7471 epoch: 1899/2000\n",
+ "fps step: 15919 fps step and policy inference: 9429 fps total: 8388 epoch: 1900/2000\n",
+ "fps step: 15556 fps step and policy inference: 9240 fps total: 8286 epoch: 1901/2000\n",
+ "fps step: 15066 fps step and policy inference: 8543 fps total: 7568 epoch: 1902/2000\n",
+ "fps step: 16467 fps step and policy inference: 10152 fps total: 8871 epoch: 1903/2000\n",
+ "fps step: 16582 fps step and policy inference: 10238 fps total: 9091 epoch: 1904/2000\n",
+ "fps step: 16485 fps step and policy inference: 10212 fps total: 8778 epoch: 1905/2000\n",
+ "fps step: 14903 fps step and policy inference: 8324 fps total: 7303 epoch: 1906/2000\n",
+ "fps step: 15062 fps step and policy inference: 8553 fps total: 7552 epoch: 1907/2000\n",
+ "fps step: 15883 fps step and policy inference: 9384 fps total: 8053 epoch: 1908/2000\n",
+ "fps step: 15785 fps step and policy inference: 9152 fps total: 7972 epoch: 1909/2000\n",
+ "fps step: 15524 fps step and policy inference: 8903 fps total: 7675 epoch: 1910/2000\n",
+ "fps step: 16924 fps step and policy inference: 11093 fps total: 9510 epoch: 1911/2000\n",
+ "fps step: 15392 fps step and policy inference: 8954 fps total: 7840 epoch: 1912/2000\n",
+ "fps step: 15878 fps step and policy inference: 9762 fps total: 8893 epoch: 1913/2000\n",
+ "fps step: 14782 fps step and policy inference: 8248 fps total: 7242 epoch: 1914/2000\n",
+ "fps step: 16515 fps step and policy inference: 10048 fps total: 8707 epoch: 1915/2000\n",
+ "fps step: 16799 fps step and policy inference: 10564 fps total: 9356 epoch: 1916/2000\n",
+ "fps step: 16433 fps step and policy inference: 10311 fps total: 9136 epoch: 1917/2000\n",
+ "fps step: 15278 fps step and policy inference: 8968 fps total: 8014 epoch: 1918/2000\n",
+ "fps step: 15068 fps step and policy inference: 8537 fps total: 7548 epoch: 1919/2000\n",
+ "fps step: 16127 fps step and policy inference: 9750 fps total: 8574 epoch: 1920/2000\n",
+ "fps step: 16036 fps step and policy inference: 9710 fps total: 8630 epoch: 1921/2000\n",
+ "fps step: 16762 fps step and policy inference: 10332 fps total: 8657 epoch: 1922/2000\n",
+ "fps step: 17363 fps step and policy inference: 11668 fps total: 9917 epoch: 1923/2000\n",
+ "fps step: 15806 fps step and policy inference: 9123 fps total: 7793 epoch: 1924/2000\n",
+ "fps step: 16240 fps step and policy inference: 9580 fps total: 8305 epoch: 1925/2000\n",
+ "fps step: 17248 fps step and policy inference: 11162 fps total: 9821 epoch: 1926/2000\n",
+ "fps step: 15791 fps step and policy inference: 9150 fps total: 7991 epoch: 1927/2000\n",
+ "fps step: 16139 fps step and policy inference: 9745 fps total: 8721 epoch: 1928/2000\n",
+ "fps step: 17128 fps step and policy inference: 11167 fps total: 9778 epoch: 1929/2000\n",
+ "fps step: 15772 fps step and policy inference: 9242 fps total: 8004 epoch: 1930/2000\n",
+ "fps step: 15364 fps step and policy inference: 8986 fps total: 8066 epoch: 1931/2000\n",
+ "fps step: 16680 fps step and policy inference: 10099 fps total: 8671 epoch: 1932/2000\n",
+ "fps step: 15624 fps step and policy inference: 9112 fps total: 7953 epoch: 1933/2000\n",
+ "fps step: 15020 fps step and policy inference: 8422 fps total: 7422 epoch: 1934/2000\n",
+ "fps step: 16035 fps step and policy inference: 9419 fps total: 8353 epoch: 1935/2000\n",
+ "fps step: 15785 fps step and policy inference: 9048 fps total: 7943 epoch: 1936/2000\n",
+ "fps step: 17613 fps step and policy inference: 12118 fps total: 10546 epoch: 1937/2000\n",
+ "fps step: 17859 fps step and policy inference: 12147 fps total: 10638 epoch: 1938/2000\n",
+ "fps step: 16317 fps step and policy inference: 9983 fps total: 8857 epoch: 1939/2000\n",
+ "fps step: 15740 fps step and policy inference: 9223 fps total: 7997 epoch: 1940/2000\n",
+ "fps step: 15643 fps step and policy inference: 9226 fps total: 8237 epoch: 1941/2000\n",
+ "fps step: 16724 fps step and policy inference: 10090 fps total: 8574 epoch: 1942/2000\n",
+ "fps step: 16152 fps step and policy inference: 9669 fps total: 8511 epoch: 1943/2000\n",
+ "fps step: 15686 fps step and policy inference: 9237 fps total: 8049 epoch: 1944/2000\n",
+ "fps step: 17789 fps step and policy inference: 11645 fps total: 9831 epoch: 1945/2000\n",
+ "fps step: 15594 fps step and policy inference: 9065 fps total: 7966 epoch: 1946/2000\n",
+ "fps step: 16761 fps step and policy inference: 10456 fps total: 9252 epoch: 1947/2000\n",
+ "fps step: 17863 fps step and policy inference: 12116 fps total: 10612 epoch: 1948/2000\n",
+ "fps step: 17418 fps step and policy inference: 11256 fps total: 9773 epoch: 1949/2000\n",
+ "fps step: 15704 fps step and policy inference: 9178 fps total: 7962 epoch: 1950/2000\n",
+ "fps step: 17009 fps step and policy inference: 10548 fps total: 9054 epoch: 1951/2000\n",
+ "fps step: 16547 fps step and policy inference: 10161 fps total: 8818 epoch: 1952/2000\n",
+ "fps step: 17730 fps step and policy inference: 11812 fps total: 10223 epoch: 1953/2000\n",
+ "fps step: 15815 fps step and policy inference: 9204 fps total: 8016 epoch: 1954/2000\n",
+ "fps step: 16924 fps step and policy inference: 10572 fps total: 9135 epoch: 1955/2000\n",
+ "fps step: 15657 fps step and policy inference: 9070 fps total: 7822 epoch: 1956/2000\n",
+ "fps step: 15659 fps step and policy inference: 9100 fps total: 7928 epoch: 1957/2000\n",
+ "fps step: 17762 fps step and policy inference: 11749 fps total: 10034 epoch: 1958/2000\n",
+ "fps step: 16767 fps step and policy inference: 10471 fps total: 8921 epoch: 1959/2000\n",
+ "fps step: 16184 fps step and policy inference: 9640 fps total: 8370 epoch: 1960/2000\n",
+ "fps step: 16093 fps step and policy inference: 9691 fps total: 8492 epoch: 1961/2000\n",
+ "fps step: 14686 fps step and policy inference: 8219 fps total: 7250 epoch: 1962/2000\n",
+ "fps step: 16031 fps step and policy inference: 9446 fps total: 8237 epoch: 1963/2000\n",
+ "fps step: 15459 fps step and policy inference: 8969 fps total: 8045 epoch: 1964/2000\n",
+ "fps step: 15930 fps step and policy inference: 9409 fps total: 8406 epoch: 1965/2000\n",
+ "fps step: 16729 fps step and policy inference: 10240 fps total: 8761 epoch: 1966/2000\n",
+ "fps step: 17320 fps step and policy inference: 11292 fps total: 9898 epoch: 1967/2000\n",
+ "fps step: 15246 fps step and policy inference: 8730 fps total: 7774 epoch: 1968/2000\n",
+ "fps step: 16242 fps step and policy inference: 10035 fps total: 8955 epoch: 1969/2000\n",
+ "fps step: 17866 fps step and policy inference: 12248 fps total: 10668 epoch: 1970/2000\n",
+ "fps step: 16774 fps step and policy inference: 10411 fps total: 9031 epoch: 1971/2000\n",
+ "fps step: 16447 fps step and policy inference: 9955 fps total: 8850 epoch: 1972/2000\n",
+ "fps step: 18383 fps step and policy inference: 12807 fps total: 11100 epoch: 1973/2000\n",
+ "fps step: 16384 fps step and policy inference: 10038 fps total: 8944 epoch: 1974/2000\n",
+ "fps step: 17012 fps step and policy inference: 10453 fps total: 8864 epoch: 1975/2000\n",
+ "fps step: 15836 fps step and policy inference: 9284 fps total: 8173 epoch: 1976/2000\n",
+ "fps step: 17020 fps step and policy inference: 10953 fps total: 9619 epoch: 1977/2000\n",
+ "fps step: 15367 fps step and policy inference: 8764 fps total: 7650 epoch: 1978/2000\n",
+ "fps step: 14774 fps step and policy inference: 8318 fps total: 7444 epoch: 1979/2000\n",
+ "fps step: 17332 fps step and policy inference: 11393 fps total: 10041 epoch: 1980/2000\n",
+ "fps step: 17672 fps step and policy inference: 11664 fps total: 10166 epoch: 1981/2000\n",
+ "fps step: 17437 fps step and policy inference: 11231 fps total: 9385 epoch: 1982/2000\n",
+ "fps step: 17195 fps step and policy inference: 10785 fps total: 9284 epoch: 1983/2000\n",
+ "fps step: 15395 fps step and policy inference: 8862 fps total: 7827 epoch: 1984/2000\n",
+ "fps step: 16602 fps step and policy inference: 10456 fps total: 9265 epoch: 1985/2000\n",
+ "fps step: 17479 fps step and policy inference: 11410 fps total: 9778 epoch: 1986/2000\n",
+ "fps step: 16822 fps step and policy inference: 10198 fps total: 8771 epoch: 1987/2000\n",
+ "fps step: 16679 fps step and policy inference: 10657 fps total: 9463 epoch: 1988/2000\n",
+ "fps step: 15945 fps step and policy inference: 9260 fps total: 7957 epoch: 1989/2000\n",
+ "fps step: 14981 fps step and policy inference: 8439 fps total: 7394 epoch: 1990/2000\n",
+ "fps step: 14852 fps step and policy inference: 8307 fps total: 7291 epoch: 1991/2000\n",
+ "fps step: 15062 fps step and policy inference: 8719 fps total: 7742 epoch: 1992/2000\n",
+ "fps step: 18165 fps step and policy inference: 12764 fps total: 11114 epoch: 1993/2000\n",
+ "fps step: 15804 fps step and policy inference: 9298 fps total: 8063 epoch: 1994/2000\n",
+ "fps step: 16254 fps step and policy inference: 10157 fps total: 8672 epoch: 1995/2000\n",
+ "fps step: 15648 fps step and policy inference: 9567 fps total: 8496 epoch: 1996/2000\n",
+ "fps step: 17282 fps step and policy inference: 11562 fps total: 9808 epoch: 1997/2000\n",
+ "fps step: 15490 fps step and policy inference: 9021 fps total: 7845 epoch: 1998/2000\n",
+ "fps step: 16164 fps step and policy inference: 9883 fps total: 8634 epoch: 1999/2000\n",
+ "fps step: 15440 fps step and policy inference: 8908 fps total: 7836 epoch: 2000/2000\n",
+ "=> saving checkpoint 'runs/Ant-v4_envpool_23-08-31-39/nn/last_Ant-v4_envpoolep2000rew[5904.908].pth'\n",
+ "MAX EPOCHS NUM!\n"
+ ]
+ }
+ ],
+ "source": [
+ "runner = Runner()\n",
+ "runner.load(ant_config)\n",
+ "runner.run({\n",
+ " 'train': True,\n",
+ "})"
+ ]
+ }
+ ],
+ "metadata": {
+ "colab": {
+ "provenance": [],
+ "collapsed_sections": []
+ },
+ "kernelspec": {
+ "display_name": "Python 3",
+ "name": "python3"
+ },
+ "language_info": {
+ "name": "python"
+ },
+ "accelerator": "GPU",
+ "gpuClass": "standard"
+ },
+ "nbformat": 4,
+ "nbformat_minor": 0
+}
\ No newline at end of file
diff --git a/demo/envpool_demo_breakout.ipynb b/demo/envpool_demo_breakout.ipynb
new file mode 100644
index 00000000..8e960e35
--- /dev/null
+++ b/demo/envpool_demo_breakout.ipynb
@@ -0,0 +1,1887 @@
+{
+ "nbformat": 4,
+ "nbformat_minor": 0,
+ "metadata": {
+ "colab": {
+ "provenance": [],
+ "collapsed_sections": []
+ },
+ "kernelspec": {
+ "name": "python3",
+ "display_name": "Python 3"
+ },
+ "language_info": {
+ "name": "python"
+ }
+ },
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "source": [
+ "[](https://colab.research.google.com/github/sail-sg/envpool/blob/main/demo/envpool-demo-breakout.ipynb)"
+ ],
+ "metadata": {
+ "id": "3XJJ0SaDUeoR"
+ }
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "id": "lIyDa5q_zWih"
+ },
+ "outputs": [],
+ "source": [
+ "# OS: Ubuntu 18.04.4 LTS x86_64\n",
+ "# Kernel: 4.18.0-15-generic\n",
+ "# CPU: Intel(R) Core(TM) i9-10920X CPU (24) @ 3.50GHz\n",
+ "# GPU: NVIDIA GeForce RTX 2080 Ti\n",
+ "!apt-get install -y \\\n",
+ " libgl1-mesa-dev \\\n",
+ " libgl1-mesa-glx \\\n",
+ " libglew-dev \\\n",
+ " libosmesa6-dev \\\n",
+ " software-properties-common\n",
+ "\n",
+ "!apt-get install -y patchelf\n",
+ "!pip install git+https://github.com/Denys88/rl_games\n",
+ "!pip install envpool\n",
+ "!pip install gym\n",
+ "\n",
+ "!pip install free-mujoco-py\n",
+ "!apt-get install -y xvfb python-opengl ffmpeg > /dev/null 2>&1\n",
+ "!pip install imageio==2.4.1\n",
+ "!pip install -U colabgymrender"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "from rl_games.torch_runner import Runner"
+ ],
+ "metadata": {
+ "id": "LTVQdyGO0TA3"
+ },
+ "execution_count": null,
+ "outputs": []
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "## breakout_envpool config:\n",
+ "breakout_config = {'params': {'algo': {'name': 'a2c_discrete'}, \n",
+ " 'model': {'name': 'discrete_a2c'}, \n",
+ " 'network': {'name': 'actor_critic', \n",
+ " 'separate': False, \n",
+ " 'space': {'discrete': None}, \n",
+ " 'cnn': {'permute_input': False, \n",
+ " 'type': 'conv2d', \n",
+ " 'activation': 'relu', \n",
+ " 'initializer': {'name': 'orthogonal_initializer', 'gain': 1.41421356237}, \n",
+ " 'convs': [{'filters': 32, 'kernel_size': 8, 'strides': 4, 'padding': 0}, \n",
+ " {'filters': 64, 'kernel_size': 4, 'strides': 2, 'padding': 0}, \n",
+ " {'filters': 64, 'kernel_size': 3, 'strides': 1, 'padding': 0}]}, \n",
+ " 'mlp': {'units': [512], 'activation': 'relu', 'initializer': {'name': 'orthogonal_initializer', 'gain': 1.41421356237}}}, \n",
+ " 'config': {'name': 'Breakout_envpool', \n",
+ " 'env_name': 'envpool', \n",
+ " 'score_to_win': 900, \n",
+ " 'normalize_value': True, \n",
+ " 'normalize_input': False, \n",
+ " 'reward_shaper': {'scale_value': 1}, \n",
+ " 'normalize_advantage': True, \n",
+ " 'gamma': 0.999, \n",
+ " 'tau': 0.95, \n",
+ " 'learning_rate': '8e-4', \n",
+ " 'lr_schedule': 'None', \n",
+ " 'kl_threshold': 0.01, \n",
+ " 'grad_norm': 1.0, \n",
+ " 'entropy_coef': 0.01, \n",
+ " 'truncate_grads': True, \n",
+ " 'e_clip': 0.2, \n",
+ " 'clip_value': False, \n",
+ " 'num_actors': 64, \n",
+ " 'horizon_length': 128, \n",
+ " 'minibatch_size': 1024, \n",
+ " 'mini_epochs': 2, \n",
+ " 'critic_coef': 1, \n",
+ " 'max_epochs': 1500, \n",
+ " 'env_config': {'env_name': 'Breakout-v5', 'episodic_life': True, 'has_lives': True}, \n",
+ " 'player': {'render': False, 'games_num': 200, 'n_game_life': 5, 'determenistic': False}}}}"
+ ],
+ "metadata": {
+ "id": "76pPi11m0Uv1"
+ },
+ "execution_count": null,
+ "outputs": []
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "runner = Runner()\n",
+ "runner.load(breakout_config)\n",
+ "runner.run({\n",
+ " 'train': True,\n",
+ "})"
+ ],
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "id": "uMH9k4b13h-Y",
+ "outputId": "6da13308-00de-46b6-c61f-d64be6a8de9a"
+ },
+ "execution_count": null,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "self.seed = 1655992919\n",
+ "Started to train\n",
+ "current training device: cuda:0\n",
+ "conv_name: conv2d\n",
+ "build mlp: 3136\n",
+ "RunningMeanStd: (1,)\n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stderr",
+ "text": [
+ "/home/liubo/.local/lib/python3.8/site-packages/rl_games/common/experience.py:341: DeprecationWarning: `np.long` is a deprecated alias for `np.compat.long`. To silence this warning, use `np.compat.long` by itself. In the likely event your code does not need to work on Python 2 you can use the builtin `int` for which `np.compat.long` is itself an alias. Doing this will not modify any behaviour and is safe. When replacing `np.long`, you may wish to use e.g. `np.int64` or `np.int32` to specify the precision. If you wish to review your current use, check the release note link for additional information.\n",
+ "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+ " self.tensor_dict['actions'] = self._create_tensor_from_space(gym.spaces.Box(low=0, high=1,shape=self.actions_shape, dtype=np.long), obs_base_shape)\n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "fps step: 10570 fps step and policy inference: 8468 fps total: 6379 epoch: 1/1500\n",
+ "fps step: 6256 fps step and policy inference: 5105 fps total: 4586 epoch: 2/1500\n",
+ "fps step: 5151 fps step and policy inference: 4203 fps total: 3778 epoch: 3/1500\n",
+ "fps step: 4585 fps step and policy inference: 3745 fps total: 3404 epoch: 4/1500\n",
+ "fps step: 4710 fps step and policy inference: 3887 fps total: 3518 epoch: 5/1500\n",
+ "fps step: 4352 fps step and policy inference: 3540 fps total: 3244 epoch: 6/1500\n",
+ "fps step: 4295 fps step and policy inference: 3417 fps total: 3137 epoch: 7/1500\n",
+ "fps step: 4633 fps step and policy inference: 3813 fps total: 3458 epoch: 8/1500\n",
+ "fps step: 4562 fps step and policy inference: 3699 fps total: 3374 epoch: 9/1500\n",
+ "fps step: 4451 fps step and policy inference: 3580 fps total: 3263 epoch: 10/1500\n",
+ "fps step: 4653 fps step and policy inference: 3708 fps total: 3369 epoch: 11/1500\n",
+ "fps step: 4603 fps step and policy inference: 3745 fps total: 3402 epoch: 12/1500\n",
+ "fps step: 4949 fps step and policy inference: 4032 fps total: 3646 epoch: 13/1500\n",
+ "fps step: 4866 fps step and policy inference: 3908 fps total: 3536 epoch: 14/1500\n",
+ "fps step: 4718 fps step and policy inference: 3848 fps total: 3508 epoch: 15/1500\n",
+ "fps step: 5238 fps step and policy inference: 4213 fps total: 3794 epoch: 16/1500\n",
+ "fps step: 5145 fps step and policy inference: 4150 fps total: 3740 epoch: 17/1500\n",
+ "fps step: 4542 fps step and policy inference: 3666 fps total: 3352 epoch: 18/1500\n",
+ "fps step: 5430 fps step and policy inference: 4382 fps total: 3919 epoch: 19/1500\n",
+ "fps step: 5162 fps step and policy inference: 4199 fps total: 3786 epoch: 20/1500\n",
+ "fps step: 5189 fps step and policy inference: 4263 fps total: 3835 epoch: 21/1500\n",
+ "fps step: 5487 fps step and policy inference: 4460 fps total: 4008 epoch: 22/1500\n",
+ "fps step: 5124 fps step and policy inference: 4139 fps total: 3748 epoch: 23/1500\n",
+ "fps step: 5140 fps step and policy inference: 4215 fps total: 3788 epoch: 24/1500\n",
+ "fps step: 5337 fps step and policy inference: 4364 fps total: 3919 epoch: 25/1500\n",
+ "fps step: 5278 fps step and policy inference: 4314 fps total: 3861 epoch: 26/1500\n",
+ "fps step: 6026 fps step and policy inference: 4889 fps total: 4404 epoch: 27/1500\n",
+ "fps step: 5239 fps step and policy inference: 4216 fps total: 3797 epoch: 28/1500\n",
+ "fps step: 5334 fps step and policy inference: 4353 fps total: 3898 epoch: 29/1500\n",
+ "fps step: 5727 fps step and policy inference: 4572 fps total: 4112 epoch: 30/1500\n",
+ "fps step: 5334 fps step and policy inference: 4269 fps total: 3837 epoch: 31/1500\n",
+ "fps step: 5460 fps step and policy inference: 4324 fps total: 3873 epoch: 32/1500\n",
+ "fps step: 5663 fps step and policy inference: 4545 fps total: 4053 epoch: 33/1500\n",
+ "fps step: 6123 fps step and policy inference: 5012 fps total: 4507 epoch: 34/1500\n",
+ "fps step: 5270 fps step and policy inference: 4243 fps total: 3808 epoch: 35/1500\n",
+ "fps step: 6078 fps step and policy inference: 5046 fps total: 4534 epoch: 36/1500\n",
+ "fps step: 5839 fps step and policy inference: 4791 fps total: 4325 epoch: 37/1500\n",
+ "fps step: 5797 fps step and policy inference: 4618 fps total: 4120 epoch: 38/1500\n",
+ "fps step: 5706 fps step and policy inference: 4638 fps total: 4125 epoch: 39/1500\n",
+ "fps step: 5797 fps step and policy inference: 4683 fps total: 4167 epoch: 40/1500\n",
+ "fps step: 6208 fps step and policy inference: 5085 fps total: 4566 epoch: 41/1500\n",
+ "fps step: 5657 fps step and policy inference: 4566 fps total: 4074 epoch: 42/1500\n",
+ "fps step: 5616 fps step and policy inference: 4534 fps total: 4043 epoch: 43/1500\n",
+ "fps step: 6515 fps step and policy inference: 5398 fps total: 4814 epoch: 44/1500\n",
+ "fps step: 6311 fps step and policy inference: 5061 fps total: 4545 epoch: 45/1500\n",
+ "fps step: 5883 fps step and policy inference: 4792 fps total: 4327 epoch: 46/1500\n",
+ "fps step: 5655 fps step and policy inference: 4467 fps total: 4003 epoch: 47/1500\n",
+ "fps step: 6238 fps step and policy inference: 5053 fps total: 4540 epoch: 48/1500\n",
+ "fps step: 6486 fps step and policy inference: 5281 fps total: 4722 epoch: 49/1500\n",
+ "fps step: 6119 fps step and policy inference: 4994 fps total: 4492 epoch: 50/1500\n",
+ "fps step: 5720 fps step and policy inference: 4509 fps total: 4028 epoch: 51/1500\n",
+ "fps step: 6326 fps step and policy inference: 5135 fps total: 4605 epoch: 52/1500\n",
+ "fps step: 6715 fps step and policy inference: 5437 fps total: 4842 epoch: 53/1500\n",
+ "fps step: 5694 fps step and policy inference: 4564 fps total: 4137 epoch: 54/1500\n",
+ "fps step: 6665 fps step and policy inference: 5458 fps total: 4868 epoch: 55/1500\n",
+ "fps step: 6058 fps step and policy inference: 4906 fps total: 4414 epoch: 56/1500\n",
+ "fps step: 6244 fps step and policy inference: 5040 fps total: 4533 epoch: 57/1500\n",
+ "fps step: 7124 fps step and policy inference: 5816 fps total: 5145 epoch: 58/1500\n",
+ "fps step: 6160 fps step and policy inference: 5030 fps total: 4520 epoch: 59/1500\n",
+ "fps step: 6995 fps step and policy inference: 5669 fps total: 5032 epoch: 60/1500\n",
+ "fps step: 5875 fps step and policy inference: 4721 fps total: 4192 epoch: 61/1500\n",
+ "fps step: 6176 fps step and policy inference: 5014 fps total: 4506 epoch: 62/1500\n",
+ "fps step: 6189 fps step and policy inference: 4998 fps total: 4493 epoch: 63/1500\n",
+ "fps step: 6234 fps step and policy inference: 4946 fps total: 4451 epoch: 64/1500\n",
+ "fps step: 6176 fps step and policy inference: 4880 fps total: 4326 epoch: 65/1500\n",
+ "fps step: 6160 fps step and policy inference: 4857 fps total: 4379 epoch: 66/1500\n",
+ "fps step: 6672 fps step and policy inference: 5447 fps total: 4856 epoch: 67/1500\n",
+ "fps step: 6433 fps step and policy inference: 5146 fps total: 4611 epoch: 68/1500\n",
+ "fps step: 6169 fps step and policy inference: 5020 fps total: 4511 epoch: 69/1500\n",
+ "fps step: 6116 fps step and policy inference: 4815 fps total: 4344 epoch: 70/1500\n",
+ "fps step: 6763 fps step and policy inference: 5397 fps total: 4812 epoch: 71/1500\n",
+ "fps step: 6104 fps step and policy inference: 4990 fps total: 4487 epoch: 72/1500\n",
+ "fps step: 6465 fps step and policy inference: 5282 fps total: 4723 epoch: 73/1500\n",
+ "fps step: 6108 fps step and policy inference: 4919 fps total: 4409 epoch: 74/1500\n",
+ "fps step: 6285 fps step and policy inference: 5154 fps total: 4616 epoch: 75/1500\n",
+ "fps step: 6988 fps step and policy inference: 5584 fps total: 4962 epoch: 76/1500\n",
+ "fps step: 6018 fps step and policy inference: 4914 fps total: 4423 epoch: 77/1500\n",
+ "fps step: 6916 fps step and policy inference: 5665 fps total: 5026 epoch: 78/1500\n",
+ "fps step: 6665 fps step and policy inference: 5375 fps total: 4795 epoch: 79/1500\n",
+ "fps step: 5935 fps step and policy inference: 4719 fps total: 4266 epoch: 80/1500\n",
+ "fps step: 7243 fps step and policy inference: 5940 fps total: 5242 epoch: 81/1500\n",
+ "fps step: 6664 fps step and policy inference: 5535 fps total: 4924 epoch: 82/1500\n",
+ "fps step: 6697 fps step and policy inference: 5432 fps total: 4843 epoch: 83/1500\n",
+ "fps step: 6666 fps step and policy inference: 5429 fps total: 4841 epoch: 84/1500\n",
+ "fps step: 7193 fps step and policy inference: 5761 fps total: 5089 epoch: 85/1500\n",
+ "fps step: 7659 fps step and policy inference: 6317 fps total: 5535 epoch: 86/1500\n",
+ "fps step: 6390 fps step and policy inference: 5071 fps total: 4544 epoch: 87/1500\n",
+ "fps step: 7256 fps step and policy inference: 5936 fps total: 5239 epoch: 88/1500\n",
+ "fps step: 6280 fps step and policy inference: 5066 fps total: 4546 epoch: 89/1500\n",
+ "fps step: 7629 fps step and policy inference: 6175 fps total: 5423 epoch: 90/1500\n",
+ "fps step: 6391 fps step and policy inference: 5353 fps total: 4779 epoch: 91/1500\n",
+ "fps step: 7609 fps step and policy inference: 6225 fps total: 5460 epoch: 92/1500\n",
+ "fps step: 6983 fps step and policy inference: 5719 fps total: 5068 epoch: 93/1500\n",
+ "fps step: 5632 fps step and policy inference: 4472 fps total: 4012 epoch: 94/1500\n",
+ "fps step: 6746 fps step and policy inference: 5295 fps total: 4733 epoch: 95/1500\n",
+ "fps step: 6589 fps step and policy inference: 5298 fps total: 4722 epoch: 96/1500\n",
+ "fps step: 6488 fps step and policy inference: 5231 fps total: 4677 epoch: 97/1500\n",
+ "fps step: 7027 fps step and policy inference: 5642 fps total: 5005 epoch: 98/1500\n",
+ "fps step: 6535 fps step and policy inference: 5319 fps total: 4751 epoch: 99/1500\n",
+ "fps step: 7081 fps step and policy inference: 5729 fps total: 5075 epoch: 100/1500\n",
+ "saving next best rewards: [3.3788164]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 6777 fps step and policy inference: 5454 fps total: 4845 epoch: 101/1500\n",
+ "fps step: 6139 fps step and policy inference: 4923 fps total: 4436 epoch: 102/1500\n",
+ "fps step: 5874 fps step and policy inference: 4709 fps total: 4194 epoch: 103/1500\n",
+ "saving next best rewards: [3.4591439]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7791 fps step and policy inference: 6343 fps total: 5550 epoch: 104/1500\n",
+ "fps step: 7182 fps step and policy inference: 5891 fps total: 5201 epoch: 105/1500\n",
+ "saving next best rewards: [3.484466]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 6962 fps step and policy inference: 5708 fps total: 5058 epoch: 106/1500\n",
+ "saving next best rewards: [3.514262]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7590 fps step and policy inference: 6162 fps total: 5412 epoch: 107/1500\n",
+ "saving next best rewards: [3.7745383]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 6263 fps step and policy inference: 4969 fps total: 4388 epoch: 108/1500\n",
+ "saving next best rewards: [3.9690695]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 6568 fps step and policy inference: 5290 fps total: 4727 epoch: 109/1500\n",
+ "saving next best rewards: [3.978831]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7009 fps step and policy inference: 5568 fps total: 4948 epoch: 110/1500\n",
+ "fps step: 6523 fps step and policy inference: 5146 fps total: 4609 epoch: 111/1500\n",
+ "fps step: 6414 fps step and policy inference: 5132 fps total: 4593 epoch: 112/1500\n",
+ "fps step: 6837 fps step and policy inference: 5522 fps total: 4900 epoch: 113/1500\n",
+ "fps step: 5842 fps step and policy inference: 4659 fps total: 4159 epoch: 114/1500\n",
+ "fps step: 7144 fps step and policy inference: 5928 fps total: 5231 epoch: 115/1500\n",
+ "fps step: 7541 fps step and policy inference: 6157 fps total: 5407 epoch: 116/1500\n",
+ "fps step: 6888 fps step and policy inference: 5552 fps total: 4916 epoch: 117/1500\n",
+ "saving next best rewards: [3.9874258]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7054 fps step and policy inference: 5695 fps total: 5048 epoch: 118/1500\n",
+ "saving next best rewards: [4.0102034]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 6718 fps step and policy inference: 5427 fps total: 4830 epoch: 119/1500\n",
+ "fps step: 7249 fps step and policy inference: 6028 fps total: 5307 epoch: 120/1500\n",
+ "saving next best rewards: [4.2239246]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 5887 fps step and policy inference: 4715 fps total: 4210 epoch: 121/1500\n",
+ "saving next best rewards: [4.306622]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7111 fps step and policy inference: 5836 fps total: 5158 epoch: 122/1500\n",
+ "fps step: 7481 fps step and policy inference: 5999 fps total: 5285 epoch: 123/1500\n",
+ "saving next best rewards: [4.3125052]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7495 fps step and policy inference: 6056 fps total: 5317 epoch: 124/1500\n",
+ "fps step: 6234 fps step and policy inference: 5106 fps total: 4574 epoch: 125/1500\n",
+ "fps step: 7887 fps step and policy inference: 6478 fps total: 5653 epoch: 126/1500\n",
+ "saving next best rewards: [4.346729]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 6480 fps step and policy inference: 5158 fps total: 4615 epoch: 127/1500\n",
+ "saving next best rewards: [4.4043536]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 6830 fps step and policy inference: 5482 fps total: 4878 epoch: 128/1500\n",
+ "saving next best rewards: [4.5567203]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8386 fps step and policy inference: 6823 fps total: 5914 epoch: 129/1500\n",
+ "fps step: 7556 fps step and policy inference: 6186 fps total: 5429 epoch: 130/1500\n",
+ "saving next best rewards: [4.637531]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 6324 fps step and policy inference: 5132 fps total: 4601 epoch: 131/1500\n",
+ "saving next best rewards: [4.8382044]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7331 fps step and policy inference: 5911 fps total: 5204 epoch: 132/1500\n",
+ "fps step: 7365 fps step and policy inference: 5896 fps total: 5206 epoch: 133/1500\n",
+ "fps step: 6208 fps step and policy inference: 4933 fps total: 4438 epoch: 134/1500\n",
+ "fps step: 6429 fps step and policy inference: 5151 fps total: 4604 epoch: 135/1500\n",
+ "fps step: 6228 fps step and policy inference: 5016 fps total: 4502 epoch: 136/1500\n",
+ "fps step: 8148 fps step and policy inference: 6685 fps total: 5812 epoch: 137/1500\n",
+ "fps step: 7001 fps step and policy inference: 5710 fps total: 5059 epoch: 138/1500\n",
+ "fps step: 7071 fps step and policy inference: 5689 fps total: 5032 epoch: 139/1500\n",
+ "saving next best rewards: [4.911133]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8616 fps step and policy inference: 6935 fps total: 6002 epoch: 140/1500\n",
+ "fps step: 6542 fps step and policy inference: 5218 fps total: 4660 epoch: 141/1500\n",
+ "saving next best rewards: [5.0805364]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7193 fps step and policy inference: 5894 fps total: 5188 epoch: 142/1500\n",
+ "saving next best rewards: [5.205803]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 6548 fps step and policy inference: 5189 fps total: 4645 epoch: 143/1500\n",
+ "saving next best rewards: [5.2218266]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7158 fps step and policy inference: 5663 fps total: 5024 epoch: 144/1500\n",
+ "fps step: 7488 fps step and policy inference: 6066 fps total: 5322 epoch: 145/1500\n",
+ "saving next best rewards: [5.2892976]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7615 fps step and policy inference: 6204 fps total: 5442 epoch: 146/1500\n",
+ "fps step: 7464 fps step and policy inference: 5907 fps total: 5210 epoch: 147/1500\n",
+ "fps step: 7646 fps step and policy inference: 6237 fps total: 5467 epoch: 148/1500\n",
+ "fps step: 6977 fps step and policy inference: 5515 fps total: 4905 epoch: 149/1500\n",
+ "fps step: 6835 fps step and policy inference: 5517 fps total: 4907 epoch: 150/1500\n",
+ "fps step: 7534 fps step and policy inference: 6040 fps total: 5314 epoch: 151/1500\n",
+ "fps step: 7100 fps step and policy inference: 5825 fps total: 5151 epoch: 152/1500\n",
+ "saving next best rewards: [5.393074]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7563 fps step and policy inference: 6209 fps total: 5450 epoch: 153/1500\n",
+ "fps step: 8095 fps step and policy inference: 6478 fps total: 5659 epoch: 154/1500\n",
+ "fps step: 6844 fps step and policy inference: 5431 fps total: 4839 epoch: 155/1500\n",
+ "saving next best rewards: [5.5859575]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 6730 fps step and policy inference: 5282 fps total: 4716 epoch: 156/1500\n",
+ "saving next best rewards: [5.738788]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 6590 fps step and policy inference: 5361 fps total: 4788 epoch: 157/1500\n",
+ "fps step: 6679 fps step and policy inference: 5323 fps total: 4752 epoch: 158/1500\n",
+ "saving next best rewards: [5.78205]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7290 fps step and policy inference: 5929 fps total: 5229 epoch: 159/1500\n",
+ "fps step: 7743 fps step and policy inference: 6143 fps total: 5398 epoch: 160/1500\n",
+ "fps step: 7893 fps step and policy inference: 6505 fps total: 5673 epoch: 161/1500\n",
+ "fps step: 7202 fps step and policy inference: 5846 fps total: 5166 epoch: 162/1500\n",
+ "fps step: 8206 fps step and policy inference: 6546 fps total: 5705 epoch: 163/1500\n",
+ "fps step: 7455 fps step and policy inference: 6010 fps total: 5272 epoch: 164/1500\n",
+ "fps step: 6836 fps step and policy inference: 5600 fps total: 4973 epoch: 165/1500\n",
+ "fps step: 6308 fps step and policy inference: 5040 fps total: 4525 epoch: 166/1500\n",
+ "saving next best rewards: [5.960059]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8042 fps step and policy inference: 6524 fps total: 5685 epoch: 167/1500\n",
+ "fps step: 7641 fps step and policy inference: 6206 fps total: 5444 epoch: 168/1500\n",
+ "fps step: 7602 fps step and policy inference: 6126 fps total: 5384 epoch: 169/1500\n",
+ "fps step: 7092 fps step and policy inference: 5893 fps total: 5201 epoch: 170/1500\n",
+ "fps step: 7486 fps step and policy inference: 6088 fps total: 5354 epoch: 171/1500\n",
+ "fps step: 7198 fps step and policy inference: 5803 fps total: 5129 epoch: 172/1500\n",
+ "fps step: 6950 fps step and policy inference: 5694 fps total: 5045 epoch: 173/1500\n",
+ "fps step: 8097 fps step and policy inference: 6673 fps total: 5805 epoch: 174/1500\n",
+ "fps step: 6361 fps step and policy inference: 5205 fps total: 4659 epoch: 175/1500\n",
+ "saving next best rewards: [6.125639]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7470 fps step and policy inference: 5999 fps total: 5284 epoch: 176/1500\n",
+ "fps step: 8442 fps step and policy inference: 6767 fps total: 5873 epoch: 177/1500\n",
+ "fps step: 7086 fps step and policy inference: 5771 fps total: 5107 epoch: 178/1500\n",
+ "fps step: 6665 fps step and policy inference: 5295 fps total: 4728 epoch: 179/1500\n",
+ "saving next best rewards: [6.1550446]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7307 fps step and policy inference: 5924 fps total: 5225 epoch: 180/1500\n",
+ "saving next best rewards: [6.248376]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7069 fps step and policy inference: 5673 fps total: 5030 epoch: 181/1500\n",
+ "fps step: 7258 fps step and policy inference: 5853 fps total: 5170 epoch: 182/1500\n",
+ "fps step: 8899 fps step and policy inference: 7356 fps total: 6314 epoch: 183/1500\n",
+ "fps step: 7724 fps step and policy inference: 6271 fps total: 5480 epoch: 184/1500\n",
+ "fps step: 7472 fps step and policy inference: 5825 fps total: 5139 epoch: 185/1500\n",
+ "saving next best rewards: [6.646997]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 6860 fps step and policy inference: 5487 fps total: 4866 epoch: 186/1500\n",
+ "saving next best rewards: [6.6519623]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8128 fps step and policy inference: 6634 fps total: 5769 epoch: 187/1500\n",
+ "saving next best rewards: [6.6604214]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7081 fps step and policy inference: 5641 fps total: 4992 epoch: 188/1500\n",
+ "fps step: 7817 fps step and policy inference: 6290 fps total: 5504 epoch: 189/1500\n",
+ "saving next best rewards: [6.841798]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7261 fps step and policy inference: 5906 fps total: 5211 epoch: 190/1500\n",
+ "fps step: 7617 fps step and policy inference: 6080 fps total: 5348 epoch: 191/1500\n",
+ "fps step: 6909 fps step and policy inference: 5510 fps total: 4894 epoch: 192/1500\n",
+ "fps step: 6469 fps step and policy inference: 5124 fps total: 4590 epoch: 193/1500\n",
+ "fps step: 6499 fps step and policy inference: 5206 fps total: 4649 epoch: 194/1500\n",
+ "fps step: 7885 fps step and policy inference: 6387 fps total: 5568 epoch: 195/1500\n",
+ "fps step: 7252 fps step and policy inference: 5861 fps total: 5176 epoch: 196/1500\n",
+ "fps step: 7768 fps step and policy inference: 6307 fps total: 5522 epoch: 197/1500\n",
+ "fps step: 7746 fps step and policy inference: 6373 fps total: 5573 epoch: 198/1500\n",
+ "fps step: 7961 fps step and policy inference: 6492 fps total: 5663 epoch: 199/1500\n",
+ "saving next best rewards: [6.9124894]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7044 fps step and policy inference: 5676 fps total: 5031 epoch: 200/1500\n",
+ "saving next best rewards: [7.3546133]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8854 fps step and policy inference: 7245 fps total: 6231 epoch: 201/1500\n",
+ "fps step: 7656 fps step and policy inference: 6385 fps total: 5581 epoch: 202/1500\n",
+ "fps step: 8300 fps step and policy inference: 6746 fps total: 5858 epoch: 203/1500\n",
+ "fps step: 8169 fps step and policy inference: 6718 fps total: 5838 epoch: 204/1500\n",
+ "fps step: 7656 fps step and policy inference: 6170 fps total: 5413 epoch: 205/1500\n",
+ "fps step: 7198 fps step and policy inference: 5889 fps total: 5186 epoch: 206/1500\n",
+ "saving next best rewards: [7.4263897]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7604 fps step and policy inference: 6035 fps total: 5295 epoch: 207/1500\n",
+ "fps step: 8665 fps step and policy inference: 6979 fps total: 6031 epoch: 208/1500\n",
+ "saving next best rewards: [7.488877]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8519 fps step and policy inference: 6848 fps total: 5932 epoch: 209/1500\n",
+ "fps step: 6974 fps step and policy inference: 5602 fps total: 4967 epoch: 210/1500\n",
+ "saving next best rewards: [7.9715753]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7190 fps step and policy inference: 5691 fps total: 5039 epoch: 211/1500\n",
+ "saving next best rewards: [8.114768]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 6804 fps step and policy inference: 5539 fps total: 4926 epoch: 212/1500\n",
+ "fps step: 7254 fps step and policy inference: 5865 fps total: 5181 epoch: 213/1500\n",
+ "fps step: 7317 fps step and policy inference: 5904 fps total: 5200 epoch: 214/1500\n",
+ "fps step: 6749 fps step and policy inference: 5343 fps total: 4765 epoch: 215/1500\n",
+ "fps step: 8876 fps step and policy inference: 7127 fps total: 6141 epoch: 216/1500\n",
+ "fps step: 9484 fps step and policy inference: 7634 fps total: 6517 epoch: 217/1500\n",
+ "fps step: 8081 fps step and policy inference: 6512 fps total: 5682 epoch: 218/1500\n",
+ "fps step: 6462 fps step and policy inference: 5205 fps total: 4655 epoch: 219/1500\n",
+ "fps step: 8100 fps step and policy inference: 6535 fps total: 5697 epoch: 220/1500\n",
+ "fps step: 8335 fps step and policy inference: 6754 fps total: 5851 epoch: 221/1500\n",
+ "saving next best rewards: [8.237596]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7231 fps step and policy inference: 5729 fps total: 5078 epoch: 222/1500\n",
+ "saving next best rewards: [8.379279]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 6931 fps step and policy inference: 5543 fps total: 4931 epoch: 223/1500\n",
+ "fps step: 8209 fps step and policy inference: 6645 fps total: 5781 epoch: 224/1500\n",
+ "fps step: 7844 fps step and policy inference: 6197 fps total: 5428 epoch: 225/1500\n",
+ "fps step: 8087 fps step and policy inference: 6502 fps total: 5642 epoch: 226/1500\n",
+ "saving next best rewards: [8.456155]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 9046 fps step and policy inference: 7322 fps total: 6292 epoch: 227/1500\n",
+ "saving next best rewards: [8.653183]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7418 fps step and policy inference: 5894 fps total: 5203 epoch: 228/1500\n",
+ "saving next best rewards: [8.85155]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7018 fps step and policy inference: 5663 fps total: 5026 epoch: 229/1500\n",
+ "saving next best rewards: [9.224875]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8143 fps step and policy inference: 6501 fps total: 5671 epoch: 230/1500\n",
+ "saving next best rewards: [9.334941]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7727 fps step and policy inference: 6104 fps total: 5374 epoch: 231/1500\n",
+ "saving next best rewards: [9.358768]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8034 fps step and policy inference: 6487 fps total: 5659 epoch: 232/1500\n",
+ "fps step: 7653 fps step and policy inference: 6277 fps total: 5504 epoch: 233/1500\n",
+ "fps step: 7362 fps step and policy inference: 5972 fps total: 5269 epoch: 234/1500\n",
+ "fps step: 7119 fps step and policy inference: 5686 fps total: 5042 epoch: 235/1500\n",
+ "fps step: 8333 fps step and policy inference: 6822 fps total: 5919 epoch: 236/1500\n",
+ "fps step: 7071 fps step and policy inference: 5598 fps total: 4959 epoch: 237/1500\n",
+ "fps step: 8191 fps step and policy inference: 6680 fps total: 5807 epoch: 238/1500\n",
+ "fps step: 7491 fps step and policy inference: 6025 fps total: 5299 epoch: 239/1500\n",
+ "fps step: 7400 fps step and policy inference: 5929 fps total: 5227 epoch: 240/1500\n",
+ "fps step: 7577 fps step and policy inference: 6224 fps total: 5458 epoch: 241/1500\n",
+ "fps step: 7248 fps step and policy inference: 5869 fps total: 5187 epoch: 242/1500\n",
+ "fps step: 7636 fps step and policy inference: 6156 fps total: 5402 epoch: 243/1500\n",
+ "fps step: 7345 fps step and policy inference: 5973 fps total: 5259 epoch: 244/1500\n",
+ "fps step: 7539 fps step and policy inference: 6052 fps total: 5325 epoch: 245/1500\n",
+ "fps step: 7911 fps step and policy inference: 6469 fps total: 5645 epoch: 246/1500\n",
+ "fps step: 8394 fps step and policy inference: 6773 fps total: 5875 epoch: 247/1500\n",
+ "fps step: 8220 fps step and policy inference: 6710 fps total: 5830 epoch: 248/1500\n",
+ "fps step: 8519 fps step and policy inference: 6889 fps total: 5945 epoch: 249/1500\n",
+ "fps step: 6638 fps step and policy inference: 5375 fps total: 4797 epoch: 250/1500\n",
+ "fps step: 9000 fps step and policy inference: 7377 fps total: 6326 epoch: 251/1500\n",
+ "fps step: 7815 fps step and policy inference: 6339 fps total: 5546 epoch: 252/1500\n",
+ "fps step: 7825 fps step and policy inference: 6267 fps total: 5494 epoch: 253/1500\n",
+ "fps step: 7556 fps step and policy inference: 6053 fps total: 5322 epoch: 254/1500\n",
+ "fps step: 7388 fps step and policy inference: 5932 fps total: 5230 epoch: 255/1500\n",
+ "fps step: 8351 fps step and policy inference: 6752 fps total: 5863 epoch: 256/1500\n",
+ "fps step: 9525 fps step and policy inference: 7701 fps total: 6563 epoch: 257/1500\n",
+ "fps step: 8508 fps step and policy inference: 6904 fps total: 5974 epoch: 258/1500\n",
+ "saving next best rewards: [9.792122]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7830 fps step and policy inference: 6215 fps total: 5444 epoch: 259/1500\n",
+ "saving next best rewards: [9.908857]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8811 fps step and policy inference: 7066 fps total: 6080 epoch: 260/1500\n",
+ "saving next best rewards: [10.201598]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 6793 fps step and policy inference: 5422 fps total: 4832 epoch: 261/1500\n",
+ "saving next best rewards: [10.590834]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7339 fps step and policy inference: 5930 fps total: 5214 epoch: 262/1500\n",
+ "fps step: 7875 fps step and policy inference: 6355 fps total: 5559 epoch: 263/1500\n",
+ "fps step: 6757 fps step and policy inference: 5437 fps total: 4839 epoch: 264/1500\n",
+ "fps step: 7875 fps step and policy inference: 6296 fps total: 5513 epoch: 265/1500\n",
+ "fps step: 8787 fps step and policy inference: 6923 fps total: 5982 epoch: 266/1500\n",
+ "fps step: 8161 fps step and policy inference: 6516 fps total: 5683 epoch: 267/1500\n",
+ "fps step: 7039 fps step and policy inference: 5492 fps total: 4883 epoch: 268/1500\n",
+ "fps step: 8355 fps step and policy inference: 6725 fps total: 5839 epoch: 269/1500\n",
+ "fps step: 8209 fps step and policy inference: 6702 fps total: 5823 epoch: 270/1500\n",
+ "fps step: 8270 fps step and policy inference: 6723 fps total: 5840 epoch: 271/1500\n",
+ "saving next best rewards: [11.15418]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7482 fps step and policy inference: 5925 fps total: 5227 epoch: 272/1500\n",
+ "fps step: 7455 fps step and policy inference: 5946 fps total: 5243 epoch: 273/1500\n",
+ "fps step: 7827 fps step and policy inference: 6275 fps total: 5483 epoch: 274/1500\n",
+ "fps step: 7325 fps step and policy inference: 5811 fps total: 5133 epoch: 275/1500\n",
+ "fps step: 9285 fps step and policy inference: 7497 fps total: 6413 epoch: 276/1500\n",
+ "fps step: 9308 fps step and policy inference: 7608 fps total: 6497 epoch: 277/1500\n",
+ "fps step: 8415 fps step and policy inference: 6751 fps total: 5857 epoch: 278/1500\n",
+ "fps step: 7088 fps step and policy inference: 5634 fps total: 4998 epoch: 279/1500\n",
+ "fps step: 8482 fps step and policy inference: 6936 fps total: 5997 epoch: 280/1500\n",
+ "fps step: 7273 fps step and policy inference: 5989 fps total: 5277 epoch: 281/1500\n",
+ "fps step: 6572 fps step and policy inference: 5206 fps total: 4658 epoch: 282/1500\n",
+ "fps step: 6833 fps step and policy inference: 5501 fps total: 4892 epoch: 283/1500\n",
+ "fps step: 7779 fps step and policy inference: 6225 fps total: 5459 epoch: 284/1500\n",
+ "fps step: 7285 fps step and policy inference: 5715 fps total: 5054 epoch: 285/1500\n",
+ "fps step: 7847 fps step and policy inference: 6252 fps total: 5483 epoch: 286/1500\n",
+ "fps step: 7644 fps step and policy inference: 5999 fps total: 5282 epoch: 287/1500\n",
+ "fps step: 9173 fps step and policy inference: 7294 fps total: 6265 epoch: 288/1500\n",
+ "fps step: 6999 fps step and policy inference: 5655 fps total: 5004 epoch: 289/1500\n",
+ "fps step: 8092 fps step and policy inference: 6575 fps total: 5712 epoch: 290/1500\n",
+ "fps step: 8869 fps step and policy inference: 7180 fps total: 6178 epoch: 291/1500\n",
+ "fps step: 8268 fps step and policy inference: 6842 fps total: 5929 epoch: 292/1500\n",
+ "saving next best rewards: [11.2650385]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7672 fps step and policy inference: 6213 fps total: 5449 epoch: 293/1500\n",
+ "fps step: 8263 fps step and policy inference: 6791 fps total: 5892 epoch: 294/1500\n",
+ "fps step: 7221 fps step and policy inference: 5842 fps total: 5163 epoch: 295/1500\n",
+ "fps step: 7411 fps step and policy inference: 5920 fps total: 5209 epoch: 296/1500\n",
+ "fps step: 8399 fps step and policy inference: 6798 fps total: 5887 epoch: 297/1500\n",
+ "fps step: 7841 fps step and policy inference: 6262 fps total: 5488 epoch: 298/1500\n",
+ "fps step: 7462 fps step and policy inference: 5976 fps total: 5265 epoch: 299/1500\n",
+ "fps step: 7543 fps step and policy inference: 5905 fps total: 5207 epoch: 300/1500\n",
+ "fps step: 8169 fps step and policy inference: 6682 fps total: 5803 epoch: 301/1500\n",
+ "fps step: 8770 fps step and policy inference: 7131 fps total: 6145 epoch: 302/1500\n",
+ "fps step: 7339 fps step and policy inference: 5876 fps total: 5189 epoch: 303/1500\n",
+ "fps step: 7668 fps step and policy inference: 6147 fps total: 5399 epoch: 304/1500\n",
+ "fps step: 8366 fps step and policy inference: 6800 fps total: 5898 epoch: 305/1500\n",
+ "saving next best rewards: [11.321135]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 6798 fps step and policy inference: 5429 fps total: 4834 epoch: 306/1500\n",
+ "saving next best rewards: [11.347021]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8069 fps step and policy inference: 6491 fps total: 5663 epoch: 307/1500\n",
+ "fps step: 7530 fps step and policy inference: 6112 fps total: 5374 epoch: 308/1500\n",
+ "fps step: 8680 fps step and policy inference: 7036 fps total: 6084 epoch: 309/1500\n",
+ "fps step: 8323 fps step and policy inference: 6670 fps total: 5799 epoch: 310/1500\n",
+ "fps step: 7143 fps step and policy inference: 5714 fps total: 5060 epoch: 311/1500\n",
+ "fps step: 6311 fps step and policy inference: 5076 fps total: 4552 epoch: 312/1500\n",
+ "fps step: 7310 fps step and policy inference: 5807 fps total: 5135 epoch: 313/1500\n",
+ "fps step: 8471 fps step and policy inference: 6841 fps total: 5932 epoch: 314/1500\n",
+ "fps step: 8673 fps step and policy inference: 7098 fps total: 6119 epoch: 315/1500\n",
+ "fps step: 7901 fps step and policy inference: 6411 fps total: 5604 epoch: 316/1500\n",
+ "fps step: 7909 fps step and policy inference: 6422 fps total: 5599 epoch: 317/1500\n",
+ "saving next best rewards: [11.678578]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8263 fps step and policy inference: 6776 fps total: 5881 epoch: 318/1500\n",
+ "saving next best rewards: [11.837453]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 9121 fps step and policy inference: 7325 fps total: 6278 epoch: 319/1500\n",
+ "fps step: 8322 fps step and policy inference: 6739 fps total: 5851 epoch: 320/1500\n",
+ "fps step: 7299 fps step and policy inference: 5866 fps total: 5169 epoch: 321/1500\n",
+ "fps step: 7832 fps step and policy inference: 6176 fps total: 5421 epoch: 322/1500\n",
+ "fps step: 7440 fps step and policy inference: 5792 fps total: 5120 epoch: 323/1500\n",
+ "fps step: 7552 fps step and policy inference: 6042 fps total: 5317 epoch: 324/1500\n",
+ "fps step: 8762 fps step and policy inference: 7087 fps total: 6111 epoch: 325/1500\n",
+ "saving next best rewards: [11.961185]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 6990 fps step and policy inference: 5702 fps total: 5051 epoch: 326/1500\n",
+ "saving next best rewards: [12.419674]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8376 fps step and policy inference: 6654 fps total: 5786 epoch: 327/1500\n",
+ "fps step: 7928 fps step and policy inference: 6353 fps total: 5560 epoch: 328/1500\n",
+ "fps step: 7714 fps step and policy inference: 6158 fps total: 5416 epoch: 329/1500\n",
+ "fps step: 9261 fps step and policy inference: 7599 fps total: 6487 epoch: 330/1500\n",
+ "fps step: 8176 fps step and policy inference: 6574 fps total: 5721 epoch: 331/1500\n",
+ "fps step: 8228 fps step and policy inference: 6533 fps total: 5692 epoch: 332/1500\n",
+ "saving next best rewards: [12.646646]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8513 fps step and policy inference: 6899 fps total: 5968 epoch: 333/1500\n",
+ "fps step: 8416 fps step and policy inference: 6753 fps total: 5860 epoch: 334/1500\n",
+ "fps step: 6616 fps step and policy inference: 5082 fps total: 4559 epoch: 335/1500\n",
+ "saving next best rewards: [12.842021]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 6852 fps step and policy inference: 5546 fps total: 4929 epoch: 336/1500\n",
+ "saving next best rewards: [13.191847]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7808 fps step and policy inference: 6205 fps total: 5443 epoch: 337/1500\n",
+ "fps step: 8848 fps step and policy inference: 7121 fps total: 6136 epoch: 338/1500\n",
+ "saving next best rewards: [13.808141]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8275 fps step and policy inference: 6519 fps total: 5685 epoch: 339/1500\n",
+ "fps step: 7351 fps step and policy inference: 5890 fps total: 5198 epoch: 340/1500\n",
+ "fps step: 7772 fps step and policy inference: 6229 fps total: 5464 epoch: 341/1500\n",
+ "fps step: 9105 fps step and policy inference: 7409 fps total: 6355 epoch: 342/1500\n",
+ "fps step: 8723 fps step and policy inference: 7022 fps total: 6065 epoch: 343/1500\n",
+ "fps step: 7649 fps step and policy inference: 6184 fps total: 5428 epoch: 344/1500\n",
+ "fps step: 7123 fps step and policy inference: 5663 fps total: 5022 epoch: 345/1500\n",
+ "fps step: 7590 fps step and policy inference: 6192 fps total: 5440 epoch: 346/1500\n",
+ "saving next best rewards: [14.107501]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7662 fps step and policy inference: 6157 fps total: 5411 epoch: 347/1500\n",
+ "saving next best rewards: [15.062423]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8137 fps step and policy inference: 6588 fps total: 5740 epoch: 348/1500\n",
+ "saving next best rewards: [15.098409]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7555 fps step and policy inference: 6108 fps total: 5374 epoch: 349/1500\n",
+ "fps step: 7776 fps step and policy inference: 6347 fps total: 5547 epoch: 350/1500\n",
+ "fps step: 9113 fps step and policy inference: 7370 fps total: 6324 epoch: 351/1500\n",
+ "fps step: 8326 fps step and policy inference: 6639 fps total: 5775 epoch: 352/1500\n",
+ "fps step: 7383 fps step and policy inference: 5947 fps total: 5232 epoch: 353/1500\n",
+ "fps step: 8558 fps step and policy inference: 6971 fps total: 6025 epoch: 354/1500\n",
+ "saving next best rewards: [15.110613]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8649 fps step and policy inference: 6794 fps total: 5890 epoch: 355/1500\n",
+ "saving next best rewards: [15.657738]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7661 fps step and policy inference: 6104 fps total: 5366 epoch: 356/1500\n",
+ "saving next best rewards: [16.597609]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7550 fps step and policy inference: 5996 fps total: 5281 epoch: 357/1500\n",
+ "fps step: 7828 fps step and policy inference: 6321 fps total: 5531 epoch: 358/1500\n",
+ "fps step: 8230 fps step and policy inference: 6722 fps total: 5838 epoch: 359/1500\n",
+ "fps step: 7240 fps step and policy inference: 5791 fps total: 5122 epoch: 360/1500\n",
+ "fps step: 8619 fps step and policy inference: 6934 fps total: 5997 epoch: 361/1500\n",
+ "fps step: 7996 fps step and policy inference: 6479 fps total: 5638 epoch: 362/1500\n",
+ "fps step: 7761 fps step and policy inference: 6127 fps total: 5384 epoch: 363/1500\n",
+ "fps step: 7161 fps step and policy inference: 5733 fps total: 5078 epoch: 364/1500\n",
+ "fps step: 8731 fps step and policy inference: 7068 fps total: 6097 epoch: 365/1500\n",
+ "fps step: 8752 fps step and policy inference: 6958 fps total: 6025 epoch: 366/1500\n",
+ "fps step: 7985 fps step and policy inference: 6606 fps total: 5752 epoch: 367/1500\n",
+ "fps step: 8491 fps step and policy inference: 6890 fps total: 5965 epoch: 368/1500\n",
+ "fps step: 9420 fps step and policy inference: 7542 fps total: 6444 epoch: 369/1500\n",
+ "fps step: 7486 fps step and policy inference: 6122 fps total: 5380 epoch: 370/1500\n",
+ "fps step: 7589 fps step and policy inference: 6134 fps total: 5390 epoch: 371/1500\n",
+ "fps step: 7899 fps step and policy inference: 6386 fps total: 5581 epoch: 372/1500\n",
+ "fps step: 8446 fps step and policy inference: 6725 fps total: 5841 epoch: 373/1500\n",
+ "fps step: 7730 fps step and policy inference: 6064 fps total: 5336 epoch: 374/1500\n",
+ "fps step: 7816 fps step and policy inference: 6218 fps total: 5455 epoch: 375/1500\n",
+ "fps step: 8185 fps step and policy inference: 6524 fps total: 5687 epoch: 376/1500\n",
+ "fps step: 8234 fps step and policy inference: 6675 fps total: 5804 epoch: 377/1500\n",
+ "fps step: 7397 fps step and policy inference: 5962 fps total: 5260 epoch: 378/1500\n",
+ "fps step: 8358 fps step and policy inference: 6717 fps total: 5834 epoch: 379/1500\n",
+ "fps step: 8349 fps step and policy inference: 6705 fps total: 5826 epoch: 380/1500\n",
+ "saving next best rewards: [16.893972]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7034 fps step and policy inference: 5535 fps total: 4926 epoch: 381/1500\n",
+ "saving next best rewards: [16.930658]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7697 fps step and policy inference: 6092 fps total: 5357 epoch: 382/1500\n",
+ "fps step: 7449 fps step and policy inference: 6014 fps total: 5295 epoch: 383/1500\n",
+ "fps step: 6691 fps step and policy inference: 5210 fps total: 4658 epoch: 384/1500\n",
+ "saving next best rewards: [17.175018]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7481 fps step and policy inference: 5916 fps total: 5220 epoch: 385/1500\n",
+ "fps step: 9373 fps step and policy inference: 7530 fps total: 6419 epoch: 386/1500\n",
+ "saving next best rewards: [17.544489]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7549 fps step and policy inference: 5928 fps total: 5217 epoch: 387/1500\n",
+ "saving next best rewards: [19.912893]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8511 fps step and policy inference: 6857 fps total: 5938 epoch: 388/1500\n",
+ "fps step: 8583 fps step and policy inference: 6965 fps total: 6021 epoch: 389/1500\n",
+ "fps step: 8667 fps step and policy inference: 6975 fps total: 6027 epoch: 390/1500\n",
+ "fps step: 8592 fps step and policy inference: 7079 fps total: 6106 epoch: 391/1500\n",
+ "fps step: 8741 fps step and policy inference: 6966 fps total: 6024 epoch: 392/1500\n",
+ "fps step: 8542 fps step and policy inference: 6929 fps total: 5994 epoch: 393/1500\n",
+ "fps step: 8855 fps step and policy inference: 7258 fps total: 6232 epoch: 394/1500\n",
+ "fps step: 7493 fps step and policy inference: 5979 fps total: 5273 epoch: 395/1500\n",
+ "saving next best rewards: [20.93684]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 9057 fps step and policy inference: 7295 fps total: 6265 epoch: 396/1500\n",
+ "fps step: 8998 fps step and policy inference: 7218 fps total: 6207 epoch: 397/1500\n",
+ "fps step: 8320 fps step and policy inference: 6620 fps total: 5766 epoch: 398/1500\n",
+ "saving next best rewards: [21.675663]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 9818 fps step and policy inference: 7927 fps total: 6706 epoch: 399/1500\n",
+ "saving next best rewards: [22.008396]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8851 fps step and policy inference: 7094 fps total: 6114 epoch: 400/1500\n",
+ "saving next best rewards: [22.433498]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7461 fps step and policy inference: 5927 fps total: 5228 epoch: 401/1500\n",
+ "fps step: 7897 fps step and policy inference: 6396 fps total: 5593 epoch: 402/1500\n",
+ "saving next best rewards: [22.817749]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8477 fps step and policy inference: 6871 fps total: 5951 epoch: 403/1500\n",
+ "fps step: 8586 fps step and policy inference: 7028 fps total: 6074 epoch: 404/1500\n",
+ "fps step: 7409 fps step and policy inference: 5832 fps total: 5154 epoch: 405/1500\n",
+ "fps step: 8404 fps step and policy inference: 6693 fps total: 5813 epoch: 406/1500\n",
+ "saving next best rewards: [24.379225]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8275 fps step and policy inference: 6577 fps total: 5728 epoch: 407/1500\n",
+ "fps step: 7815 fps step and policy inference: 6262 fps total: 5491 epoch: 408/1500\n",
+ "fps step: 10027 fps step and policy inference: 8008 fps total: 6784 epoch: 409/1500\n",
+ "fps step: 8058 fps step and policy inference: 6548 fps total: 5709 epoch: 410/1500\n",
+ "fps step: 7838 fps step and policy inference: 6241 fps total: 5469 epoch: 411/1500\n",
+ "fps step: 8389 fps step and policy inference: 6918 fps total: 5988 epoch: 412/1500\n",
+ "fps step: 9135 fps step and policy inference: 7320 fps total: 6285 epoch: 413/1500\n",
+ "fps step: 10168 fps step and policy inference: 8221 fps total: 6928 epoch: 414/1500\n",
+ "fps step: 7972 fps step and policy inference: 6421 fps total: 5609 epoch: 415/1500\n",
+ "fps step: 8491 fps step and policy inference: 6744 fps total: 5863 epoch: 416/1500\n",
+ "fps step: 9131 fps step and policy inference: 7321 fps total: 6278 epoch: 417/1500\n",
+ "fps step: 8436 fps step and policy inference: 6950 fps total: 6011 epoch: 418/1500\n",
+ "saving next best rewards: [26.52502]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8281 fps step and policy inference: 6645 fps total: 5775 epoch: 419/1500\n",
+ "saving next best rewards: [27.020956]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7950 fps step and policy inference: 6278 fps total: 5482 epoch: 420/1500\n",
+ "fps step: 9388 fps step and policy inference: 7560 fps total: 6459 epoch: 421/1500\n",
+ "fps step: 9767 fps step and policy inference: 7781 fps total: 6615 epoch: 422/1500\n",
+ "fps step: 8975 fps step and policy inference: 7129 fps total: 6142 epoch: 423/1500\n",
+ "fps step: 8700 fps step and policy inference: 6993 fps total: 6039 epoch: 424/1500\n",
+ "saving next best rewards: [30.730392]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8842 fps step and policy inference: 7184 fps total: 6183 epoch: 425/1500\n",
+ "fps step: 7469 fps step and policy inference: 5872 fps total: 5186 epoch: 426/1500\n",
+ "fps step: 8739 fps step and policy inference: 7049 fps total: 6082 epoch: 427/1500\n",
+ "saving next best rewards: [32.12111]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7933 fps step and policy inference: 6299 fps total: 5515 epoch: 428/1500\n",
+ "saving next best rewards: [32.28975]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8548 fps step and policy inference: 6777 fps total: 5839 epoch: 429/1500\n",
+ "fps step: 8861 fps step and policy inference: 7081 fps total: 6107 epoch: 430/1500\n",
+ "fps step: 8544 fps step and policy inference: 6917 fps total: 5986 epoch: 431/1500\n",
+ "fps step: 8453 fps step and policy inference: 6795 fps total: 5891 epoch: 432/1500\n",
+ "fps step: 7787 fps step and policy inference: 6123 fps total: 5369 epoch: 433/1500\n",
+ "fps step: 7415 fps step and policy inference: 5856 fps total: 5173 epoch: 434/1500\n",
+ "saving next best rewards: [35.929085]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7571 fps step and policy inference: 5917 fps total: 5219 epoch: 435/1500\n",
+ "fps step: 7460 fps step and policy inference: 5933 fps total: 5234 epoch: 436/1500\n",
+ "fps step: 9446 fps step and policy inference: 7614 fps total: 6501 epoch: 437/1500\n",
+ "fps step: 7470 fps step and policy inference: 5971 fps total: 5259 epoch: 438/1500\n",
+ "fps step: 9480 fps step and policy inference: 7633 fps total: 6515 epoch: 439/1500\n",
+ "fps step: 8468 fps step and policy inference: 6788 fps total: 5864 epoch: 440/1500\n",
+ "fps step: 7449 fps step and policy inference: 5916 fps total: 5222 epoch: 441/1500\n",
+ "fps step: 7689 fps step and policy inference: 6213 fps total: 5450 epoch: 442/1500\n",
+ "fps step: 8857 fps step and policy inference: 7200 fps total: 6198 epoch: 443/1500\n",
+ "fps step: 7807 fps step and policy inference: 6264 fps total: 5489 epoch: 444/1500\n",
+ "fps step: 8853 fps step and policy inference: 7157 fps total: 6164 epoch: 445/1500\n",
+ "fps step: 8279 fps step and policy inference: 6592 fps total: 5738 epoch: 446/1500\n",
+ "fps step: 7140 fps step and policy inference: 5627 fps total: 4883 epoch: 447/1500\n",
+ "fps step: 9511 fps step and policy inference: 7516 fps total: 6426 epoch: 448/1500\n",
+ "fps step: 9568 fps step and policy inference: 7749 fps total: 6598 epoch: 449/1500\n",
+ "fps step: 9386 fps step and policy inference: 7544 fps total: 6436 epoch: 450/1500\n",
+ "fps step: 7994 fps step and policy inference: 6412 fps total: 5603 epoch: 451/1500\n",
+ "fps step: 7461 fps step and policy inference: 6010 fps total: 5295 epoch: 452/1500\n",
+ "fps step: 9280 fps step and policy inference: 7533 fps total: 6441 epoch: 453/1500\n",
+ "fps step: 8339 fps step and policy inference: 6651 fps total: 5784 epoch: 454/1500\n",
+ "fps step: 8359 fps step and policy inference: 6751 fps total: 5851 epoch: 455/1500\n",
+ "fps step: 8664 fps step and policy inference: 7048 fps total: 6083 epoch: 456/1500\n",
+ "fps step: 8867 fps step and policy inference: 7109 fps total: 6130 epoch: 457/1500\n",
+ "fps step: 8286 fps step and policy inference: 6657 fps total: 5788 epoch: 458/1500\n",
+ "saving next best rewards: [37.00623]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8299 fps step and policy inference: 6691 fps total: 5798 epoch: 459/1500\n",
+ "saving next best rewards: [38.876015]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8861 fps step and policy inference: 7211 fps total: 6203 epoch: 460/1500\n",
+ "fps step: 8246 fps step and policy inference: 6563 fps total: 5719 epoch: 461/1500\n",
+ "fps step: 9647 fps step and policy inference: 7708 fps total: 6569 epoch: 462/1500\n",
+ "fps step: 9075 fps step and policy inference: 7264 fps total: 6241 epoch: 463/1500\n",
+ "fps step: 8117 fps step and policy inference: 6615 fps total: 5756 epoch: 464/1500\n",
+ "fps step: 9350 fps step and policy inference: 7510 fps total: 6423 epoch: 465/1500\n",
+ "fps step: 7249 fps step and policy inference: 5886 fps total: 5198 epoch: 466/1500\n",
+ "fps step: 8657 fps step and policy inference: 7042 fps total: 6080 epoch: 467/1500\n",
+ "fps step: 7956 fps step and policy inference: 6513 fps total: 5684 epoch: 468/1500\n",
+ "fps step: 9139 fps step and policy inference: 7305 fps total: 6273 epoch: 469/1500\n",
+ "fps step: 8697 fps step and policy inference: 7126 fps total: 6145 epoch: 470/1500\n",
+ "fps step: 8709 fps step and policy inference: 6998 fps total: 6031 epoch: 471/1500\n",
+ "fps step: 7910 fps step and policy inference: 6275 fps total: 5498 epoch: 472/1500\n",
+ "fps step: 8376 fps step and policy inference: 6804 fps total: 5899 epoch: 473/1500\n",
+ "saving next best rewards: [39.874485]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8244 fps step and policy inference: 6563 fps total: 5718 epoch: 474/1500\n",
+ "fps step: 8751 fps step and policy inference: 7001 fps total: 6050 epoch: 475/1500\n",
+ "fps step: 7518 fps step and policy inference: 6004 fps total: 5289 epoch: 476/1500\n",
+ "fps step: 8750 fps step and policy inference: 6933 fps total: 6002 epoch: 477/1500\n",
+ "fps step: 9240 fps step and policy inference: 7516 fps total: 6432 epoch: 478/1500\n",
+ "fps step: 8842 fps step and policy inference: 7185 fps total: 6170 epoch: 479/1500\n",
+ "fps step: 7882 fps step and policy inference: 6296 fps total: 5515 epoch: 480/1500\n",
+ "fps step: 9194 fps step and policy inference: 7516 fps total: 6429 epoch: 481/1500\n",
+ "saving next best rewards: [40.514366]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8459 fps step and policy inference: 6704 fps total: 5828 epoch: 482/1500\n",
+ "fps step: 8991 fps step and policy inference: 7160 fps total: 6164 epoch: 483/1500\n",
+ "fps step: 9252 fps step and policy inference: 7545 fps total: 6445 epoch: 484/1500\n",
+ "saving next best rewards: [44.334324]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7908 fps step and policy inference: 6186 fps total: 5427 epoch: 485/1500\n",
+ "fps step: 9034 fps step and policy inference: 7233 fps total: 6205 epoch: 486/1500\n",
+ "saving next best rewards: [48.218758]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 7958 fps step and policy inference: 6290 fps total: 5506 epoch: 487/1500\n",
+ "saving next best rewards: [54.320667]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8048 fps step and policy inference: 6426 fps total: 5614 epoch: 488/1500\n",
+ "fps step: 8453 fps step and policy inference: 6715 fps total: 5832 epoch: 489/1500\n",
+ "fps step: 8858 fps step and policy inference: 7050 fps total: 6088 epoch: 490/1500\n",
+ "fps step: 8182 fps step and policy inference: 6597 fps total: 5743 epoch: 491/1500\n",
+ "fps step: 8065 fps step and policy inference: 6427 fps total: 5614 epoch: 492/1500\n",
+ "fps step: 8237 fps step and policy inference: 6641 fps total: 5780 epoch: 493/1500\n",
+ "fps step: 8279 fps step and policy inference: 6636 fps total: 5774 epoch: 494/1500\n",
+ "fps step: 7772 fps step and policy inference: 6076 fps total: 5343 epoch: 495/1500\n",
+ "fps step: 9724 fps step and policy inference: 7753 fps total: 6574 epoch: 496/1500\n",
+ "fps step: 8277 fps step and policy inference: 6640 fps total: 5766 epoch: 497/1500\n",
+ "fps step: 7567 fps step and policy inference: 5980 fps total: 5270 epoch: 498/1500\n",
+ "fps step: 8572 fps step and policy inference: 6832 fps total: 5922 epoch: 499/1500\n",
+ "fps step: 8403 fps step and policy inference: 6699 fps total: 5830 epoch: 500/1500\n",
+ "fps step: 9213 fps step and policy inference: 7582 fps total: 6478 epoch: 501/1500\n",
+ "fps step: 8189 fps step and policy inference: 6607 fps total: 5750 epoch: 502/1500\n",
+ "fps step: 8497 fps step and policy inference: 6796 fps total: 5892 epoch: 503/1500\n",
+ "fps step: 7445 fps step and policy inference: 6016 fps total: 5305 epoch: 504/1500\n",
+ "fps step: 9246 fps step and policy inference: 7447 fps total: 6386 epoch: 505/1500\n",
+ "fps step: 8419 fps step and policy inference: 6670 fps total: 5799 epoch: 506/1500\n",
+ "fps step: 8943 fps step and policy inference: 7169 fps total: 6179 epoch: 507/1500\n",
+ "fps step: 8724 fps step and policy inference: 6959 fps total: 6001 epoch: 508/1500\n",
+ "fps step: 8592 fps step and policy inference: 6922 fps total: 5989 epoch: 509/1500\n",
+ "fps step: 9197 fps step and policy inference: 7424 fps total: 6361 epoch: 510/1500\n",
+ "fps step: 8259 fps step and policy inference: 6628 fps total: 5753 epoch: 511/1500\n",
+ "fps step: 8020 fps step and policy inference: 6435 fps total: 5621 epoch: 512/1500\n",
+ "fps step: 8032 fps step and policy inference: 6347 fps total: 5552 epoch: 513/1500\n",
+ "fps step: 9426 fps step and policy inference: 7514 fps total: 6428 epoch: 514/1500\n",
+ "fps step: 7895 fps step and policy inference: 6314 fps total: 5528 epoch: 515/1500\n",
+ "fps step: 9298 fps step and policy inference: 7576 fps total: 6456 epoch: 516/1500\n",
+ "fps step: 9499 fps step and policy inference: 7572 fps total: 6469 epoch: 517/1500\n",
+ "fps step: 8172 fps step and policy inference: 6528 fps total: 5691 epoch: 518/1500\n",
+ "fps step: 7840 fps step and policy inference: 6241 fps total: 5455 epoch: 519/1500\n",
+ "fps step: 7691 fps step and policy inference: 6041 fps total: 5318 epoch: 520/1500\n",
+ "fps step: 8040 fps step and policy inference: 6397 fps total: 5590 epoch: 521/1500\n",
+ "fps step: 7838 fps step and policy inference: 6289 fps total: 5510 epoch: 522/1500\n",
+ "fps step: 7513 fps step and policy inference: 5989 fps total: 5276 epoch: 523/1500\n",
+ "fps step: 8649 fps step and policy inference: 6933 fps total: 5998 epoch: 524/1500\n",
+ "fps step: 9073 fps step and policy inference: 7275 fps total: 6263 epoch: 525/1500\n",
+ "fps step: 8183 fps step and policy inference: 6593 fps total: 5739 epoch: 526/1500\n",
+ "fps step: 9781 fps step and policy inference: 7803 fps total: 6646 epoch: 527/1500\n",
+ "fps step: 7417 fps step and policy inference: 5827 fps total: 5155 epoch: 528/1500\n",
+ "fps step: 7926 fps step and policy inference: 6295 fps total: 5510 epoch: 529/1500\n",
+ "fps step: 8687 fps step and policy inference: 6933 fps total: 5997 epoch: 530/1500\n",
+ "fps step: 7692 fps step and policy inference: 6148 fps total: 5401 epoch: 531/1500\n",
+ "fps step: 7884 fps step and policy inference: 6332 fps total: 5547 epoch: 532/1500\n",
+ "fps step: 8530 fps step and policy inference: 7029 fps total: 6062 epoch: 533/1500\n",
+ "fps step: 9057 fps step and policy inference: 7429 fps total: 6365 epoch: 534/1500\n",
+ "fps step: 9249 fps step and policy inference: 7374 fps total: 6324 epoch: 535/1500\n",
+ "fps step: 8680 fps step and policy inference: 6893 fps total: 5965 epoch: 536/1500\n",
+ "fps step: 8265 fps step and policy inference: 6613 fps total: 5766 epoch: 537/1500\n",
+ "fps step: 8827 fps step and policy inference: 6919 fps total: 5982 epoch: 538/1500\n",
+ "fps step: 7534 fps step and policy inference: 5865 fps total: 5179 epoch: 539/1500\n",
+ "fps step: 8149 fps step and policy inference: 6613 fps total: 5760 epoch: 540/1500\n",
+ "fps step: 9224 fps step and policy inference: 7264 fps total: 6224 epoch: 541/1500\n",
+ "fps step: 9363 fps step and policy inference: 7524 fps total: 6432 epoch: 542/1500\n",
+ "fps step: 8115 fps step and policy inference: 6504 fps total: 5658 epoch: 543/1500\n",
+ "saving next best rewards: [55.079437]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8935 fps step and policy inference: 7212 fps total: 6186 epoch: 544/1500\n",
+ "saving next best rewards: [55.281284]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8638 fps step and policy inference: 6857 fps total: 5939 epoch: 545/1500\n",
+ "saving next best rewards: [59.68618]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8067 fps step and policy inference: 6619 fps total: 5760 epoch: 546/1500\n",
+ "fps step: 7521 fps step and policy inference: 6007 fps total: 5272 epoch: 547/1500\n",
+ "fps step: 8939 fps step and policy inference: 7307 fps total: 6279 epoch: 548/1500\n",
+ "fps step: 8468 fps step and policy inference: 6772 fps total: 5868 epoch: 549/1500\n",
+ "fps step: 8623 fps step and policy inference: 6990 fps total: 6044 epoch: 550/1500\n",
+ "fps step: 8662 fps step and policy inference: 7046 fps total: 6093 epoch: 551/1500\n",
+ "fps step: 9038 fps step and policy inference: 7275 fps total: 6240 epoch: 552/1500\n",
+ "fps step: 8844 fps step and policy inference: 7114 fps total: 6132 epoch: 553/1500\n",
+ "fps step: 7672 fps step and policy inference: 6129 fps total: 5388 epoch: 554/1500\n",
+ "fps step: 8864 fps step and policy inference: 7130 fps total: 6157 epoch: 555/1500\n",
+ "fps step: 8733 fps step and policy inference: 6989 fps total: 6021 epoch: 556/1500\n",
+ "fps step: 7664 fps step and policy inference: 6135 fps total: 5381 epoch: 557/1500\n",
+ "fps step: 8381 fps step and policy inference: 6589 fps total: 5724 epoch: 558/1500\n",
+ "fps step: 9978 fps step and policy inference: 7789 fps total: 6628 epoch: 559/1500\n",
+ "fps step: 8910 fps step and policy inference: 7188 fps total: 6170 epoch: 560/1500\n",
+ "fps step: 7010 fps step and policy inference: 5707 fps total: 5057 epoch: 561/1500\n",
+ "fps step: 7614 fps step and policy inference: 6058 fps total: 5319 epoch: 562/1500\n",
+ "fps step: 9817 fps step and policy inference: 7903 fps total: 6711 epoch: 563/1500\n",
+ "fps step: 9628 fps step and policy inference: 7690 fps total: 6562 epoch: 564/1500\n",
+ "fps step: 8814 fps step and policy inference: 6940 fps total: 5997 epoch: 565/1500\n",
+ "fps step: 8477 fps step and policy inference: 6759 fps total: 5866 epoch: 566/1500\n",
+ "fps step: 9032 fps step and policy inference: 7283 fps total: 6255 epoch: 567/1500\n",
+ "fps step: 8766 fps step and policy inference: 7061 fps total: 6093 epoch: 568/1500\n",
+ "fps step: 7527 fps step and policy inference: 6153 fps total: 5393 epoch: 569/1500\n",
+ "fps step: 8420 fps step and policy inference: 6790 fps total: 5885 epoch: 570/1500\n",
+ "fps step: 9162 fps step and policy inference: 7441 fps total: 6356 epoch: 571/1500\n",
+ "fps step: 7687 fps step and policy inference: 6072 fps total: 5335 epoch: 572/1500\n",
+ "fps step: 8524 fps step and policy inference: 6854 fps total: 5940 epoch: 573/1500\n",
+ "fps step: 7455 fps step and policy inference: 5961 fps total: 5253 epoch: 574/1500\n",
+ "saving next best rewards: [64.38379]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8273 fps step and policy inference: 6525 fps total: 5694 epoch: 575/1500\n",
+ "fps step: 9419 fps step and policy inference: 7625 fps total: 6507 epoch: 576/1500\n",
+ "fps step: 7718 fps step and policy inference: 6303 fps total: 5519 epoch: 577/1500\n",
+ "fps step: 9063 fps step and policy inference: 7238 fps total: 6236 epoch: 578/1500\n",
+ "fps step: 8574 fps step and policy inference: 6858 fps total: 5926 epoch: 579/1500\n",
+ "fps step: 9496 fps step and policy inference: 7551 fps total: 6454 epoch: 580/1500\n",
+ "fps step: 9048 fps step and policy inference: 7217 fps total: 6200 epoch: 581/1500\n",
+ "fps step: 9615 fps step and policy inference: 7694 fps total: 6559 epoch: 582/1500\n",
+ "fps step: 8096 fps step and policy inference: 6525 fps total: 5689 epoch: 583/1500\n",
+ "fps step: 8706 fps step and policy inference: 7006 fps total: 6060 epoch: 584/1500\n",
+ "fps step: 8321 fps step and policy inference: 6704 fps total: 5807 epoch: 585/1500\n",
+ "fps step: 9662 fps step and policy inference: 7866 fps total: 6683 epoch: 586/1500\n",
+ "fps step: 8413 fps step and policy inference: 6887 fps total: 5967 epoch: 587/1500\n",
+ "fps step: 9504 fps step and policy inference: 7556 fps total: 6458 epoch: 588/1500\n",
+ "fps step: 9380 fps step and policy inference: 7582 fps total: 6477 epoch: 589/1500\n",
+ "fps step: 8621 fps step and policy inference: 7021 fps total: 6063 epoch: 590/1500\n",
+ "fps step: 8110 fps step and policy inference: 6694 fps total: 5816 epoch: 591/1500\n",
+ "fps step: 9030 fps step and policy inference: 7198 fps total: 6195 epoch: 592/1500\n",
+ "fps step: 9140 fps step and policy inference: 7420 fps total: 6357 epoch: 593/1500\n",
+ "fps step: 9243 fps step and policy inference: 7465 fps total: 6390 epoch: 594/1500\n",
+ "fps step: 8410 fps step and policy inference: 6738 fps total: 5822 epoch: 595/1500\n",
+ "fps step: 9232 fps step and policy inference: 7404 fps total: 6350 epoch: 596/1500\n",
+ "fps step: 9501 fps step and policy inference: 7690 fps total: 6557 epoch: 597/1500\n",
+ "fps step: 7565 fps step and policy inference: 6069 fps total: 5307 epoch: 598/1500\n",
+ "fps step: 8532 fps step and policy inference: 6876 fps total: 5954 epoch: 599/1500\n",
+ "fps step: 7578 fps step and policy inference: 5972 fps total: 5267 epoch: 600/1500\n",
+ "fps step: 9419 fps step and policy inference: 7525 fps total: 6435 epoch: 601/1500\n",
+ "fps step: 9781 fps step and policy inference: 7877 fps total: 6669 epoch: 602/1500\n",
+ "fps step: 10391 fps step and policy inference: 8312 fps total: 7001 epoch: 603/1500\n",
+ "fps step: 8495 fps step and policy inference: 6757 fps total: 5864 epoch: 604/1500\n",
+ "saving next best rewards: [71.89007]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8581 fps step and policy inference: 6911 fps total: 5960 epoch: 605/1500\n",
+ "fps step: 7514 fps step and policy inference: 5863 fps total: 5178 epoch: 606/1500\n",
+ "fps step: 9110 fps step and policy inference: 7366 fps total: 6318 epoch: 607/1500\n",
+ "fps step: 8648 fps step and policy inference: 6973 fps total: 6028 epoch: 608/1500\n",
+ "fps step: 9223 fps step and policy inference: 7534 fps total: 6442 epoch: 609/1500\n",
+ "fps step: 9446 fps step and policy inference: 7416 fps total: 6356 epoch: 610/1500\n",
+ "fps step: 8702 fps step and policy inference: 6802 fps total: 5897 epoch: 611/1500\n",
+ "fps step: 8484 fps step and policy inference: 6747 fps total: 5859 epoch: 612/1500\n",
+ "fps step: 8425 fps step and policy inference: 6663 fps total: 5794 epoch: 613/1500\n",
+ "fps step: 8589 fps step and policy inference: 6851 fps total: 5936 epoch: 614/1500\n",
+ "fps step: 9111 fps step and policy inference: 7249 fps total: 6231 epoch: 615/1500\n",
+ "fps step: 7837 fps step and policy inference: 6306 fps total: 5522 epoch: 616/1500\n",
+ "fps step: 8963 fps step and policy inference: 7066 fps total: 6067 epoch: 617/1500\n",
+ "fps step: 8741 fps step and policy inference: 6911 fps total: 5959 epoch: 618/1500\n",
+ "fps step: 9315 fps step and policy inference: 7417 fps total: 6340 epoch: 619/1500\n",
+ "fps step: 8873 fps step and policy inference: 7192 fps total: 6186 epoch: 620/1500\n",
+ "fps step: 8877 fps step and policy inference: 7066 fps total: 6104 epoch: 621/1500\n",
+ "fps step: 8612 fps step and policy inference: 6942 fps total: 6002 epoch: 622/1500\n",
+ "fps step: 9717 fps step and policy inference: 7966 fps total: 6758 epoch: 623/1500\n",
+ "fps step: 7870 fps step and policy inference: 6324 fps total: 5537 epoch: 624/1500\n",
+ "fps step: 7166 fps step and policy inference: 5640 fps total: 5004 epoch: 625/1500\n",
+ "fps step: 8713 fps step and policy inference: 7056 fps total: 6085 epoch: 626/1500\n",
+ "fps step: 9435 fps step and policy inference: 7492 fps total: 6414 epoch: 627/1500\n",
+ "fps step: 8265 fps step and policy inference: 6505 fps total: 5678 epoch: 628/1500\n",
+ "fps step: 7673 fps step and policy inference: 6075 fps total: 5343 epoch: 629/1500\n",
+ "fps step: 7991 fps step and policy inference: 6344 fps total: 5551 epoch: 630/1500\n",
+ "fps step: 9066 fps step and policy inference: 7276 fps total: 6251 epoch: 631/1500\n",
+ "fps step: 8372 fps step and policy inference: 6670 fps total: 5802 epoch: 632/1500\n",
+ "fps step: 9506 fps step and policy inference: 7510 fps total: 6415 epoch: 633/1500\n",
+ "fps step: 9661 fps step and policy inference: 7738 fps total: 6590 epoch: 634/1500\n",
+ "fps step: 8498 fps step and policy inference: 6757 fps total: 5866 epoch: 635/1500\n",
+ "fps step: 8188 fps step and policy inference: 6634 fps total: 5773 epoch: 636/1500\n",
+ "fps step: 9468 fps step and policy inference: 7570 fps total: 6475 epoch: 637/1500\n",
+ "fps step: 8921 fps step and policy inference: 7173 fps total: 6177 epoch: 638/1500\n",
+ "fps step: 8945 fps step and policy inference: 7201 fps total: 6197 epoch: 639/1500\n",
+ "fps step: 8899 fps step and policy inference: 7145 fps total: 6131 epoch: 640/1500\n",
+ "fps step: 9375 fps step and policy inference: 7422 fps total: 6356 epoch: 641/1500\n",
+ "fps step: 7011 fps step and policy inference: 5510 fps total: 4882 epoch: 642/1500\n",
+ "fps step: 9032 fps step and policy inference: 7214 fps total: 6197 epoch: 643/1500\n",
+ "fps step: 9488 fps step and policy inference: 7678 fps total: 6546 epoch: 644/1500\n",
+ "fps step: 8668 fps step and policy inference: 6913 fps total: 5987 epoch: 645/1500\n",
+ "fps step: 9185 fps step and policy inference: 7436 fps total: 6370 epoch: 646/1500\n",
+ "fps step: 7887 fps step and policy inference: 6353 fps total: 5553 epoch: 647/1500\n",
+ "fps step: 9760 fps step and policy inference: 7786 fps total: 6629 epoch: 648/1500\n",
+ "fps step: 8064 fps step and policy inference: 6441 fps total: 5625 epoch: 649/1500\n",
+ "fps step: 8362 fps step and policy inference: 6577 fps total: 5730 epoch: 650/1500\n",
+ "fps step: 8477 fps step and policy inference: 6642 fps total: 5778 epoch: 651/1500\n",
+ "fps step: 8797 fps step and policy inference: 7216 fps total: 6207 epoch: 652/1500\n",
+ "fps step: 7860 fps step and policy inference: 6292 fps total: 5514 epoch: 653/1500\n",
+ "fps step: 9091 fps step and policy inference: 7197 fps total: 6204 epoch: 654/1500\n",
+ "fps step: 8999 fps step and policy inference: 7271 fps total: 6246 epoch: 655/1500\n",
+ "fps step: 8943 fps step and policy inference: 7291 fps total: 6241 epoch: 656/1500\n",
+ "fps step: 9433 fps step and policy inference: 7435 fps total: 6370 epoch: 657/1500\n",
+ "fps step: 9315 fps step and policy inference: 7477 fps total: 6400 epoch: 658/1500\n",
+ "fps step: 8077 fps step and policy inference: 6472 fps total: 5649 epoch: 659/1500\n",
+ "fps step: 8154 fps step and policy inference: 6483 fps total: 5642 epoch: 660/1500\n",
+ "fps step: 9816 fps step and policy inference: 7940 fps total: 6734 epoch: 661/1500\n",
+ "fps step: 8360 fps step and policy inference: 6594 fps total: 5735 epoch: 662/1500\n",
+ "fps step: 8366 fps step and policy inference: 6673 fps total: 5805 epoch: 663/1500\n",
+ "fps step: 9845 fps step and policy inference: 7845 fps total: 6667 epoch: 664/1500\n",
+ "fps step: 9456 fps step and policy inference: 7577 fps total: 6475 epoch: 665/1500\n",
+ "fps step: 9954 fps step and policy inference: 7991 fps total: 6772 epoch: 666/1500\n",
+ "fps step: 8594 fps step and policy inference: 6876 fps total: 5964 epoch: 667/1500\n",
+ "fps step: 9736 fps step and policy inference: 7911 fps total: 6719 epoch: 668/1500\n",
+ "fps step: 8541 fps step and policy inference: 7033 fps total: 6056 epoch: 669/1500\n",
+ "fps step: 8396 fps step and policy inference: 6751 fps total: 5856 epoch: 670/1500\n",
+ "fps step: 8605 fps step and policy inference: 6883 fps total: 5959 epoch: 671/1500\n",
+ "fps step: 8839 fps step and policy inference: 7090 fps total: 6114 epoch: 672/1500\n",
+ "fps step: 8511 fps step and policy inference: 6723 fps total: 5839 epoch: 673/1500\n",
+ "fps step: 8033 fps step and policy inference: 6450 fps total: 5630 epoch: 674/1500\n",
+ "fps step: 10442 fps step and policy inference: 8376 fps total: 7047 epoch: 675/1500\n",
+ "saving next best rewards: [76.11868]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8875 fps step and policy inference: 6979 fps total: 6031 epoch: 676/1500\n",
+ "fps step: 9246 fps step and policy inference: 7357 fps total: 6306 epoch: 677/1500\n",
+ "fps step: 7980 fps step and policy inference: 6327 fps total: 5538 epoch: 678/1500\n",
+ "fps step: 8674 fps step and policy inference: 6918 fps total: 5987 epoch: 679/1500\n",
+ "fps step: 8968 fps step and policy inference: 7238 fps total: 6215 epoch: 680/1500\n",
+ "fps step: 9565 fps step and policy inference: 7758 fps total: 6606 epoch: 681/1500\n",
+ "fps step: 8555 fps step and policy inference: 6743 fps total: 5846 epoch: 682/1500\n",
+ "fps step: 9919 fps step and policy inference: 7868 fps total: 6686 epoch: 683/1500\n",
+ "fps step: 8820 fps step and policy inference: 7051 fps total: 6077 epoch: 684/1500\n",
+ "fps step: 7856 fps step and policy inference: 6265 fps total: 5492 epoch: 685/1500\n",
+ "fps step: 9279 fps step and policy inference: 7479 fps total: 6400 epoch: 686/1500\n",
+ "fps step: 8129 fps step and policy inference: 6483 fps total: 5641 epoch: 687/1500\n",
+ "saving next best rewards: [78.45671]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8163 fps step and policy inference: 6368 fps total: 5567 epoch: 688/1500\n",
+ "fps step: 7332 fps step and policy inference: 5722 fps total: 5068 epoch: 689/1500\n",
+ "fps step: 9526 fps step and policy inference: 7623 fps total: 6506 epoch: 690/1500\n",
+ "fps step: 7970 fps step and policy inference: 6258 fps total: 5485 epoch: 691/1500\n",
+ "fps step: 9495 fps step and policy inference: 7672 fps total: 6533 epoch: 692/1500\n",
+ "fps step: 7069 fps step and policy inference: 5690 fps total: 5038 epoch: 693/1500\n",
+ "fps step: 7589 fps step and policy inference: 6062 fps total: 5334 epoch: 694/1500\n",
+ "fps step: 8184 fps step and policy inference: 6406 fps total: 5596 epoch: 695/1500\n",
+ "fps step: 10120 fps step and policy inference: 8132 fps total: 6876 epoch: 696/1500\n",
+ "fps step: 8413 fps step and policy inference: 6603 fps total: 5747 epoch: 697/1500\n",
+ "fps step: 9498 fps step and policy inference: 7667 fps total: 6538 epoch: 698/1500\n",
+ "fps step: 10031 fps step and policy inference: 8101 fps total: 6852 epoch: 699/1500\n",
+ "fps step: 9157 fps step and policy inference: 7204 fps total: 6203 epoch: 700/1500\n",
+ "fps step: 9689 fps step and policy inference: 7819 fps total: 6654 epoch: 701/1500\n",
+ "saving next best rewards: [80.01203]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 9019 fps step and policy inference: 7063 fps total: 6091 epoch: 702/1500\n",
+ "fps step: 9868 fps step and policy inference: 7952 fps total: 6745 epoch: 703/1500\n",
+ "saving next best rewards: [81.964355]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8962 fps step and policy inference: 7158 fps total: 6163 epoch: 704/1500\n",
+ "fps step: 7809 fps step and policy inference: 6276 fps total: 5497 epoch: 705/1500\n",
+ "fps step: 7155 fps step and policy inference: 5504 fps total: 4895 epoch: 706/1500\n",
+ "fps step: 9776 fps step and policy inference: 7961 fps total: 6752 epoch: 707/1500\n",
+ "fps step: 10044 fps step and policy inference: 8108 fps total: 6855 epoch: 708/1500\n",
+ "fps step: 8895 fps step and policy inference: 7153 fps total: 6160 epoch: 709/1500\n",
+ "fps step: 8444 fps step and policy inference: 6672 fps total: 5775 epoch: 710/1500\n",
+ "fps step: 10145 fps step and policy inference: 8075 fps total: 6820 epoch: 711/1500\n",
+ "fps step: 9060 fps step and policy inference: 7282 fps total: 6256 epoch: 712/1500\n",
+ "fps step: 9621 fps step and policy inference: 7641 fps total: 6506 epoch: 713/1500\n",
+ "fps step: 9219 fps step and policy inference: 7168 fps total: 6171 epoch: 714/1500\n",
+ "fps step: 9157 fps step and policy inference: 7317 fps total: 6278 epoch: 715/1500\n",
+ "fps step: 9256 fps step and policy inference: 7416 fps total: 6355 epoch: 716/1500\n",
+ "fps step: 9598 fps step and policy inference: 7762 fps total: 6608 epoch: 717/1500\n",
+ "fps step: 8847 fps step and policy inference: 7063 fps total: 6097 epoch: 718/1500\n",
+ "fps step: 9888 fps step and policy inference: 7959 fps total: 6750 epoch: 719/1500\n",
+ "saving next best rewards: [82.52823]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8989 fps step and policy inference: 6921 fps total: 5987 epoch: 720/1500\n",
+ "saving next best rewards: [91.756035]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8317 fps step and policy inference: 6752 fps total: 5843 epoch: 721/1500\n",
+ "fps step: 9268 fps step and policy inference: 7534 fps total: 6441 epoch: 722/1500\n",
+ "fps step: 8788 fps step and policy inference: 7108 fps total: 6127 epoch: 723/1500\n",
+ "fps step: 9137 fps step and policy inference: 7259 fps total: 6240 epoch: 724/1500\n",
+ "fps step: 9005 fps step and policy inference: 7345 fps total: 6291 epoch: 725/1500\n",
+ "fps step: 9774 fps step and policy inference: 7681 fps total: 6549 epoch: 726/1500\n",
+ "fps step: 8615 fps step and policy inference: 6899 fps total: 5950 epoch: 727/1500\n",
+ "fps step: 7712 fps step and policy inference: 6094 fps total: 5360 epoch: 728/1500\n",
+ "fps step: 8363 fps step and policy inference: 6669 fps total: 5803 epoch: 729/1500\n",
+ "fps step: 8740 fps step and policy inference: 6934 fps total: 5996 epoch: 730/1500\n",
+ "fps step: 8638 fps step and policy inference: 6834 fps total: 5923 epoch: 731/1500\n",
+ "fps step: 8140 fps step and policy inference: 6446 fps total: 5629 epoch: 732/1500\n",
+ "fps step: 8972 fps step and policy inference: 7074 fps total: 6068 epoch: 733/1500\n",
+ "fps step: 9504 fps step and policy inference: 7578 fps total: 6474 epoch: 734/1500\n",
+ "fps step: 9474 fps step and policy inference: 7615 fps total: 6501 epoch: 735/1500\n",
+ "fps step: 8663 fps step and policy inference: 6930 fps total: 5997 epoch: 736/1500\n",
+ "fps step: 9206 fps step and policy inference: 7469 fps total: 6393 epoch: 737/1500\n",
+ "fps step: 8926 fps step and policy inference: 7136 fps total: 6149 epoch: 738/1500\n",
+ "fps step: 8462 fps step and policy inference: 6710 fps total: 5830 epoch: 739/1500\n",
+ "fps step: 8597 fps step and policy inference: 6901 fps total: 5974 epoch: 740/1500\n",
+ "fps step: 9923 fps step and policy inference: 7944 fps total: 6740 epoch: 741/1500\n",
+ "fps step: 10468 fps step and policy inference: 8361 fps total: 7036 epoch: 742/1500\n",
+ "fps step: 9497 fps step and policy inference: 7557 fps total: 6440 epoch: 743/1500\n",
+ "fps step: 9924 fps step and policy inference: 7831 fps total: 6651 epoch: 744/1500\n",
+ "fps step: 8569 fps step and policy inference: 6817 fps total: 5906 epoch: 745/1500\n",
+ "fps step: 7559 fps step and policy inference: 5877 fps total: 5191 epoch: 746/1500\n",
+ "fps step: 9606 fps step and policy inference: 7576 fps total: 6480 epoch: 747/1500\n",
+ "fps step: 10050 fps step and policy inference: 7850 fps total: 6665 epoch: 748/1500\n",
+ "fps step: 8416 fps step and policy inference: 6896 fps total: 5963 epoch: 749/1500\n",
+ "fps step: 8620 fps step and policy inference: 6831 fps total: 5913 epoch: 750/1500\n",
+ "fps step: 9506 fps step and policy inference: 7563 fps total: 6462 epoch: 751/1500\n",
+ "fps step: 8062 fps step and policy inference: 6463 fps total: 5645 epoch: 752/1500\n",
+ "fps step: 9264 fps step and policy inference: 7428 fps total: 6365 epoch: 753/1500\n",
+ "fps step: 9859 fps step and policy inference: 7910 fps total: 6722 epoch: 754/1500\n",
+ "fps step: 9141 fps step and policy inference: 7316 fps total: 6284 epoch: 755/1500\n",
+ "fps step: 9794 fps step and policy inference: 7758 fps total: 6605 epoch: 756/1500\n",
+ "fps step: 7835 fps step and policy inference: 6260 fps total: 5495 epoch: 757/1500\n",
+ "fps step: 8180 fps step and policy inference: 6551 fps total: 5712 epoch: 758/1500\n",
+ "fps step: 9355 fps step and policy inference: 7536 fps total: 6442 epoch: 759/1500\n",
+ "fps step: 9595 fps step and policy inference: 7608 fps total: 6489 epoch: 760/1500\n",
+ "fps step: 8631 fps step and policy inference: 6891 fps total: 5951 epoch: 761/1500\n",
+ "fps step: 8862 fps step and policy inference: 7018 fps total: 6060 epoch: 762/1500\n",
+ "fps step: 8122 fps step and policy inference: 6370 fps total: 5574 epoch: 763/1500\n",
+ "fps step: 9466 fps step and policy inference: 7599 fps total: 6489 epoch: 764/1500\n",
+ "fps step: 8462 fps step and policy inference: 6618 fps total: 5766 epoch: 765/1500\n",
+ "fps step: 7340 fps step and policy inference: 5634 fps total: 4997 epoch: 766/1500\n",
+ "fps step: 9320 fps step and policy inference: 7375 fps total: 6323 epoch: 767/1500\n",
+ "fps step: 9522 fps step and policy inference: 7736 fps total: 6591 epoch: 768/1500\n",
+ "fps step: 8768 fps step and policy inference: 6950 fps total: 6021 epoch: 769/1500\n",
+ "fps step: 9522 fps step and policy inference: 7569 fps total: 6466 epoch: 770/1500\n",
+ "fps step: 9204 fps step and policy inference: 7242 fps total: 6215 epoch: 771/1500\n",
+ "fps step: 8819 fps step and policy inference: 7091 fps total: 6113 epoch: 772/1500\n",
+ "fps step: 10970 fps step and policy inference: 8753 fps total: 7312 epoch: 773/1500\n",
+ "fps step: 9941 fps step and policy inference: 8000 fps total: 6780 epoch: 774/1500\n",
+ "fps step: 8922 fps step and policy inference: 7115 fps total: 6129 epoch: 775/1500\n",
+ "fps step: 8590 fps step and policy inference: 6827 fps total: 5916 epoch: 776/1500\n",
+ "fps step: 9214 fps step and policy inference: 7269 fps total: 6223 epoch: 777/1500\n",
+ "fps step: 9427 fps step and policy inference: 7508 fps total: 6427 epoch: 778/1500\n",
+ "fps step: 8574 fps step and policy inference: 6912 fps total: 5984 epoch: 779/1500\n",
+ "fps step: 10033 fps step and policy inference: 8013 fps total: 6789 epoch: 780/1500\n",
+ "fps step: 9242 fps step and policy inference: 7494 fps total: 6413 epoch: 781/1500\n",
+ "fps step: 8334 fps step and policy inference: 6600 fps total: 5747 epoch: 782/1500\n",
+ "fps step: 10185 fps step and policy inference: 8143 fps total: 6855 epoch: 783/1500\n",
+ "fps step: 9767 fps step and policy inference: 7819 fps total: 6650 epoch: 784/1500\n",
+ "fps step: 10132 fps step and policy inference: 8034 fps total: 6804 epoch: 785/1500\n",
+ "fps step: 10114 fps step and policy inference: 8052 fps total: 6807 epoch: 786/1500\n",
+ "fps step: 9106 fps step and policy inference: 7204 fps total: 6195 epoch: 787/1500\n",
+ "fps step: 8841 fps step and policy inference: 6978 fps total: 6031 epoch: 788/1500\n",
+ "fps step: 7982 fps step and policy inference: 6301 fps total: 5521 epoch: 789/1500\n",
+ "fps step: 9088 fps step and policy inference: 7165 fps total: 6178 epoch: 790/1500\n",
+ "fps step: 8812 fps step and policy inference: 7063 fps total: 6082 epoch: 791/1500\n",
+ "fps step: 9068 fps step and policy inference: 7169 fps total: 6174 epoch: 792/1500\n",
+ "fps step: 8958 fps step and policy inference: 7242 fps total: 6228 epoch: 793/1500\n",
+ "fps step: 9176 fps step and policy inference: 7271 fps total: 6248 epoch: 794/1500\n",
+ "fps step: 10025 fps step and policy inference: 8013 fps total: 6789 epoch: 795/1500\n",
+ "fps step: 9692 fps step and policy inference: 7659 fps total: 6529 epoch: 796/1500\n",
+ "fps step: 8618 fps step and policy inference: 6783 fps total: 5884 epoch: 797/1500\n",
+ "fps step: 8676 fps step and policy inference: 6890 fps total: 5970 epoch: 798/1500\n",
+ "fps step: 8802 fps step and policy inference: 6970 fps total: 6028 epoch: 799/1500\n",
+ "fps step: 9480 fps step and policy inference: 7632 fps total: 6510 epoch: 800/1500\n",
+ "fps step: 8361 fps step and policy inference: 6532 fps total: 5699 epoch: 801/1500\n",
+ "fps step: 10353 fps step and policy inference: 8232 fps total: 6952 epoch: 802/1500\n",
+ "fps step: 9254 fps step and policy inference: 7322 fps total: 6266 epoch: 803/1500\n",
+ "fps step: 9839 fps step and policy inference: 7857 fps total: 6675 epoch: 804/1500\n",
+ "fps step: 9514 fps step and policy inference: 7582 fps total: 6454 epoch: 805/1500\n",
+ "fps step: 8123 fps step and policy inference: 6314 fps total: 5526 epoch: 806/1500\n",
+ "fps step: 10719 fps step and policy inference: 8477 fps total: 7128 epoch: 807/1500\n",
+ "fps step: 10095 fps step and policy inference: 8135 fps total: 6874 epoch: 808/1500\n",
+ "fps step: 9042 fps step and policy inference: 7203 fps total: 6201 epoch: 809/1500\n",
+ "fps step: 8732 fps step and policy inference: 7060 fps total: 6091 epoch: 810/1500\n",
+ "fps step: 8336 fps step and policy inference: 6641 fps total: 5782 epoch: 811/1500\n",
+ "fps step: 8655 fps step and policy inference: 6952 fps total: 6013 epoch: 812/1500\n",
+ "fps step: 9706 fps step and policy inference: 7732 fps total: 6586 epoch: 813/1500\n",
+ "fps step: 9768 fps step and policy inference: 7753 fps total: 6609 epoch: 814/1500\n",
+ "fps step: 9828 fps step and policy inference: 7963 fps total: 6751 epoch: 815/1500\n",
+ "fps step: 9226 fps step and policy inference: 7288 fps total: 6238 epoch: 816/1500\n",
+ "fps step: 10355 fps step and policy inference: 8267 fps total: 6979 epoch: 817/1500\n",
+ "fps step: 9513 fps step and policy inference: 7591 fps total: 6483 epoch: 818/1500\n",
+ "fps step: 9776 fps step and policy inference: 7802 fps total: 6640 epoch: 819/1500\n",
+ "fps step: 9075 fps step and policy inference: 7147 fps total: 6156 epoch: 820/1500\n",
+ "fps step: 10824 fps step and policy inference: 8586 fps total: 7194 epoch: 821/1500\n",
+ "fps step: 9203 fps step and policy inference: 7363 fps total: 6315 epoch: 822/1500\n",
+ "fps step: 8988 fps step and policy inference: 6977 fps total: 6025 epoch: 823/1500\n",
+ "fps step: 9607 fps step and policy inference: 7517 fps total: 6429 epoch: 824/1500\n",
+ "fps step: 9716 fps step and policy inference: 7649 fps total: 6527 epoch: 825/1500\n",
+ "fps step: 10036 fps step and policy inference: 7994 fps total: 6774 epoch: 826/1500\n",
+ "fps step: 7920 fps step and policy inference: 6321 fps total: 5533 epoch: 827/1500\n",
+ "fps step: 8693 fps step and policy inference: 6934 fps total: 5999 epoch: 828/1500\n",
+ "fps step: 8589 fps step and policy inference: 6818 fps total: 5912 epoch: 829/1500\n",
+ "fps step: 8079 fps step and policy inference: 6404 fps total: 5605 epoch: 830/1500\n",
+ "fps step: 9011 fps step and policy inference: 7214 fps total: 6206 epoch: 831/1500\n",
+ "fps step: 8482 fps step and policy inference: 6737 fps total: 5852 epoch: 832/1500\n",
+ "fps step: 8894 fps step and policy inference: 6954 fps total: 6002 epoch: 833/1500\n",
+ "fps step: 8273 fps step and policy inference: 6409 fps total: 5600 epoch: 834/1500\n",
+ "fps step: 9843 fps step and policy inference: 7942 fps total: 6738 epoch: 835/1500\n",
+ "fps step: 9125 fps step and policy inference: 7136 fps total: 6146 epoch: 836/1500\n",
+ "fps step: 9485 fps step and policy inference: 7584 fps total: 6480 epoch: 837/1500\n",
+ "fps step: 9824 fps step and policy inference: 7720 fps total: 6582 epoch: 838/1500\n",
+ "fps step: 8264 fps step and policy inference: 6549 fps total: 5690 epoch: 839/1500\n",
+ "fps step: 9821 fps step and policy inference: 7873 fps total: 6695 epoch: 840/1500\n",
+ "fps step: 9147 fps step and policy inference: 7230 fps total: 6217 epoch: 841/1500\n",
+ "fps step: 9356 fps step and policy inference: 7540 fps total: 6443 epoch: 842/1500\n",
+ "fps step: 9245 fps step and policy inference: 7482 fps total: 6409 epoch: 843/1500\n",
+ "fps step: 9343 fps step and policy inference: 7351 fps total: 6306 epoch: 844/1500\n",
+ "fps step: 9690 fps step and policy inference: 7770 fps total: 6613 epoch: 845/1500\n",
+ "fps step: 9599 fps step and policy inference: 7615 fps total: 6500 epoch: 846/1500\n",
+ "saving next best rewards: [92.32931]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 9040 fps step and policy inference: 7117 fps total: 6133 epoch: 847/1500\n",
+ "fps step: 10428 fps step and policy inference: 8168 fps total: 6900 epoch: 848/1500\n",
+ "saving next best rewards: [96.85858]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 9525 fps step and policy inference: 7410 fps total: 6330 epoch: 849/1500\n",
+ "fps step: 10140 fps step and policy inference: 8028 fps total: 6802 epoch: 850/1500\n",
+ "fps step: 8785 fps step and policy inference: 6964 fps total: 6006 epoch: 851/1500\n",
+ "fps step: 10268 fps step and policy inference: 8158 fps total: 6891 epoch: 852/1500\n",
+ "fps step: 7966 fps step and policy inference: 6383 fps total: 5581 epoch: 853/1500\n",
+ "fps step: 9682 fps step and policy inference: 7660 fps total: 6532 epoch: 854/1500\n",
+ "fps step: 9539 fps step and policy inference: 7637 fps total: 6521 epoch: 855/1500\n",
+ "fps step: 9770 fps step and policy inference: 7735 fps total: 6573 epoch: 856/1500\n",
+ "fps step: 7558 fps step and policy inference: 5979 fps total: 5255 epoch: 857/1500\n",
+ "fps step: 9180 fps step and policy inference: 7130 fps total: 6139 epoch: 858/1500\n",
+ "fps step: 8710 fps step and policy inference: 6892 fps total: 5966 epoch: 859/1500\n",
+ "fps step: 9261 fps step and policy inference: 7327 fps total: 6291 epoch: 860/1500\n",
+ "fps step: 8945 fps step and policy inference: 7063 fps total: 6186 epoch: 861/1500\n",
+ "fps step: 8849 fps step and policy inference: 7054 fps total: 6095 epoch: 862/1500\n",
+ "fps step: 8244 fps step and policy inference: 6527 fps total: 5690 epoch: 863/1500\n",
+ "fps step: 9803 fps step and policy inference: 7748 fps total: 6597 epoch: 864/1500\n",
+ "fps step: 9114 fps step and policy inference: 7175 fps total: 6181 epoch: 865/1500\n",
+ "fps step: 9222 fps step and policy inference: 7252 fps total: 6239 epoch: 866/1500\n",
+ "fps step: 9755 fps step and policy inference: 7835 fps total: 6666 epoch: 867/1500\n",
+ "fps step: 9744 fps step and policy inference: 7722 fps total: 6578 epoch: 868/1500\n",
+ "fps step: 9270 fps step and policy inference: 7383 fps total: 6331 epoch: 869/1500\n",
+ "fps step: 9701 fps step and policy inference: 7696 fps total: 6565 epoch: 870/1500\n",
+ "fps step: 7485 fps step and policy inference: 5791 fps total: 5122 epoch: 871/1500\n",
+ "fps step: 10483 fps step and policy inference: 8300 fps total: 6995 epoch: 872/1500\n",
+ "fps step: 10593 fps step and policy inference: 8432 fps total: 7086 epoch: 873/1500\n",
+ "saving next best rewards: [101.07825]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 8644 fps step and policy inference: 6819 fps total: 5911 epoch: 874/1500\n",
+ "fps step: 8294 fps step and policy inference: 6554 fps total: 5712 epoch: 875/1500\n",
+ "fps step: 9520 fps step and policy inference: 7534 fps total: 6426 epoch: 876/1500\n",
+ "fps step: 9624 fps step and policy inference: 7546 fps total: 6451 epoch: 877/1500\n",
+ "fps step: 9013 fps step and policy inference: 7285 fps total: 6259 epoch: 878/1500\n",
+ "fps step: 9348 fps step and policy inference: 7484 fps total: 6406 epoch: 879/1500\n",
+ "fps step: 9050 fps step and policy inference: 7109 fps total: 6127 epoch: 880/1500\n",
+ "fps step: 9486 fps step and policy inference: 7464 fps total: 6389 epoch: 881/1500\n",
+ "fps step: 10292 fps step and policy inference: 8244 fps total: 6925 epoch: 882/1500\n",
+ "fps step: 9398 fps step and policy inference: 7362 fps total: 6309 epoch: 883/1500\n",
+ "fps step: 10215 fps step and policy inference: 8053 fps total: 6811 epoch: 884/1500\n",
+ "fps step: 9257 fps step and policy inference: 7368 fps total: 6306 epoch: 885/1500\n",
+ "fps step: 10071 fps step and policy inference: 8015 fps total: 6790 epoch: 886/1500\n",
+ "fps step: 9119 fps step and policy inference: 7062 fps total: 6089 epoch: 887/1500\n",
+ "fps step: 9032 fps step and policy inference: 7198 fps total: 6189 epoch: 888/1500\n",
+ "fps step: 10041 fps step and policy inference: 7934 fps total: 6731 epoch: 889/1500\n",
+ "fps step: 9776 fps step and policy inference: 7787 fps total: 6626 epoch: 890/1500\n",
+ "fps step: 10257 fps step and policy inference: 7971 fps total: 6753 epoch: 891/1500\n",
+ "fps step: 8461 fps step and policy inference: 6564 fps total: 5721 epoch: 892/1500\n",
+ "fps step: 8563 fps step and policy inference: 6786 fps total: 5893 epoch: 893/1500\n",
+ "fps step: 8815 fps step and policy inference: 6925 fps total: 5990 epoch: 894/1500\n",
+ "fps step: 9248 fps step and policy inference: 7372 fps total: 6330 epoch: 895/1500\n",
+ "fps step: 8965 fps step and policy inference: 7017 fps total: 6057 epoch: 896/1500\n",
+ "fps step: 8797 fps step and policy inference: 7023 fps total: 6061 epoch: 897/1500\n",
+ "fps step: 10096 fps step and policy inference: 8115 fps total: 6855 epoch: 898/1500\n",
+ "fps step: 10843 fps step and policy inference: 8608 fps total: 7216 epoch: 899/1500\n",
+ "fps step: 10188 fps step and policy inference: 8073 fps total: 6831 epoch: 900/1500\n",
+ "fps step: 9566 fps step and policy inference: 7698 fps total: 6562 epoch: 901/1500\n",
+ "fps step: 8999 fps step and policy inference: 7183 fps total: 6182 epoch: 902/1500\n",
+ "fps step: 10054 fps step and policy inference: 8041 fps total: 6810 epoch: 903/1500\n",
+ "fps step: 9802 fps step and policy inference: 7876 fps total: 6690 epoch: 904/1500\n",
+ "fps step: 8499 fps step and policy inference: 6831 fps total: 5912 epoch: 905/1500\n",
+ "fps step: 9708 fps step and policy inference: 7842 fps total: 6665 epoch: 906/1500\n",
+ "fps step: 9387 fps step and policy inference: 7518 fps total: 6429 epoch: 907/1500\n",
+ "fps step: 9024 fps step and policy inference: 7246 fps total: 6227 epoch: 908/1500\n",
+ "fps step: 9514 fps step and policy inference: 7507 fps total: 6422 epoch: 909/1500\n",
+ "fps step: 10242 fps step and policy inference: 8106 fps total: 6856 epoch: 910/1500\n",
+ "fps step: 10285 fps step and policy inference: 8059 fps total: 6810 epoch: 911/1500\n",
+ "fps step: 10707 fps step and policy inference: 8484 fps total: 7124 epoch: 912/1500\n",
+ "fps step: 10104 fps step and policy inference: 8038 fps total: 6780 epoch: 913/1500\n",
+ "fps step: 9756 fps step and policy inference: 7647 fps total: 6522 epoch: 914/1500\n",
+ "fps step: 9865 fps step and policy inference: 7898 fps total: 6705 epoch: 915/1500\n",
+ "fps step: 9077 fps step and policy inference: 7351 fps total: 6287 epoch: 916/1500\n",
+ "fps step: 9931 fps step and policy inference: 7976 fps total: 6761 epoch: 917/1500\n",
+ "fps step: 9133 fps step and policy inference: 7362 fps total: 6314 epoch: 918/1500\n",
+ "fps step: 9070 fps step and policy inference: 7203 fps total: 6199 epoch: 919/1500\n",
+ "fps step: 9607 fps step and policy inference: 7683 fps total: 6551 epoch: 920/1500\n",
+ "fps step: 9215 fps step and policy inference: 7435 fps total: 6370 epoch: 921/1500\n",
+ "fps step: 9505 fps step and policy inference: 7517 fps total: 6415 epoch: 922/1500\n",
+ "fps step: 10072 fps step and policy inference: 8069 fps total: 6829 epoch: 923/1500\n",
+ "fps step: 8055 fps step and policy inference: 6261 fps total: 5488 epoch: 924/1500\n",
+ "fps step: 8263 fps step and policy inference: 6617 fps total: 5758 epoch: 925/1500\n",
+ "fps step: 10141 fps step and policy inference: 8004 fps total: 6789 epoch: 926/1500\n",
+ "fps step: 10038 fps step and policy inference: 7938 fps total: 6738 epoch: 927/1500\n",
+ "fps step: 9555 fps step and policy inference: 7478 fps total: 6404 epoch: 928/1500\n",
+ "fps step: 10842 fps step and policy inference: 8559 fps total: 7182 epoch: 929/1500\n",
+ "fps step: 10298 fps step and policy inference: 8167 fps total: 6908 epoch: 930/1500\n",
+ "fps step: 8552 fps step and policy inference: 6920 fps total: 5972 epoch: 931/1500\n",
+ "fps step: 8706 fps step and policy inference: 6918 fps total: 5970 epoch: 932/1500\n",
+ "fps step: 8625 fps step and policy inference: 6840 fps total: 5928 epoch: 933/1500\n",
+ "fps step: 10042 fps step and policy inference: 8014 fps total: 6772 epoch: 934/1500\n",
+ "fps step: 9812 fps step and policy inference: 7802 fps total: 6637 epoch: 935/1500\n",
+ "fps step: 9014 fps step and policy inference: 7237 fps total: 6222 epoch: 936/1500\n",
+ "fps step: 9379 fps step and policy inference: 7331 fps total: 6284 epoch: 937/1500\n",
+ "fps step: 9678 fps step and policy inference: 7625 fps total: 6505 epoch: 938/1500\n",
+ "fps step: 7909 fps step and policy inference: 6158 fps total: 5406 epoch: 939/1500\n",
+ "fps step: 9877 fps step and policy inference: 7944 fps total: 6740 epoch: 940/1500\n",
+ "fps step: 9934 fps step and policy inference: 7886 fps total: 6699 epoch: 941/1500\n",
+ "fps step: 9035 fps step and policy inference: 7102 fps total: 6123 epoch: 942/1500\n",
+ "fps step: 8898 fps step and policy inference: 7042 fps total: 6083 epoch: 943/1500\n",
+ "fps step: 7877 fps step and policy inference: 6158 fps total: 5391 epoch: 944/1500\n",
+ "fps step: 9885 fps step and policy inference: 7782 fps total: 6615 epoch: 945/1500\n",
+ "fps step: 9805 fps step and policy inference: 7838 fps total: 6663 epoch: 946/1500\n",
+ "fps step: 8823 fps step and policy inference: 7164 fps total: 6170 epoch: 947/1500\n",
+ "fps step: 9433 fps step and policy inference: 7599 fps total: 6490 epoch: 948/1500\n",
+ "fps step: 10049 fps step and policy inference: 8092 fps total: 6825 epoch: 949/1500\n",
+ "fps step: 9699 fps step and policy inference: 7857 fps total: 6677 epoch: 950/1500\n",
+ "fps step: 9953 fps step and policy inference: 8006 fps total: 6786 epoch: 951/1500\n",
+ "fps step: 10328 fps step and policy inference: 8232 fps total: 6946 epoch: 952/1500\n",
+ "fps step: 8166 fps step and policy inference: 6319 fps total: 5535 epoch: 953/1500\n",
+ "fps step: 10453 fps step and policy inference: 8178 fps total: 6907 epoch: 954/1500\n",
+ "fps step: 8423 fps step and policy inference: 6543 fps total: 5702 epoch: 955/1500\n",
+ "fps step: 11318 fps step and policy inference: 8963 fps total: 7431 epoch: 956/1500\n",
+ "fps step: 10183 fps step and policy inference: 8084 fps total: 6840 epoch: 957/1500\n",
+ "fps step: 9741 fps step and policy inference: 7682 fps total: 6549 epoch: 958/1500\n",
+ "fps step: 9367 fps step and policy inference: 7414 fps total: 6355 epoch: 959/1500\n",
+ "fps step: 10213 fps step and policy inference: 8037 fps total: 6821 epoch: 960/1500\n",
+ "fps step: 8700 fps step and policy inference: 6904 fps total: 5975 epoch: 961/1500\n",
+ "fps step: 8828 fps step and policy inference: 7026 fps total: 6067 epoch: 962/1500\n",
+ "fps step: 9819 fps step and policy inference: 7733 fps total: 6588 epoch: 963/1500\n",
+ "fps step: 8202 fps step and policy inference: 6565 fps total: 5712 epoch: 964/1500\n",
+ "fps step: 10033 fps step and policy inference: 7960 fps total: 6751 epoch: 965/1500\n",
+ "fps step: 9019 fps step and policy inference: 7121 fps total: 6114 epoch: 966/1500\n",
+ "fps step: 9847 fps step and policy inference: 7719 fps total: 6581 epoch: 967/1500\n",
+ "fps step: 9029 fps step and policy inference: 7098 fps total: 6123 epoch: 968/1500\n",
+ "fps step: 10433 fps step and policy inference: 8331 fps total: 6997 epoch: 969/1500\n",
+ "fps step: 10988 fps step and policy inference: 8765 fps total: 7332 epoch: 970/1500\n",
+ "fps step: 9145 fps step and policy inference: 7246 fps total: 6229 epoch: 971/1500\n",
+ "fps step: 8249 fps step and policy inference: 6519 fps total: 5685 epoch: 972/1500\n",
+ "fps step: 10215 fps step and policy inference: 8270 fps total: 6972 epoch: 973/1500\n",
+ "fps step: 9487 fps step and policy inference: 7594 fps total: 6488 epoch: 974/1500\n",
+ "fps step: 10533 fps step and policy inference: 8375 fps total: 7050 epoch: 975/1500\n",
+ "fps step: 8098 fps step and policy inference: 6294 fps total: 5501 epoch: 976/1500\n",
+ "fps step: 9675 fps step and policy inference: 7545 fps total: 6450 epoch: 977/1500\n",
+ "fps step: 8963 fps step and policy inference: 7049 fps total: 6091 epoch: 978/1500\n",
+ "fps step: 8936 fps step and policy inference: 7098 fps total: 6120 epoch: 979/1500\n",
+ "fps step: 8474 fps step and policy inference: 6664 fps total: 5793 epoch: 980/1500\n",
+ "fps step: 9071 fps step and policy inference: 7284 fps total: 6264 epoch: 981/1500\n",
+ "fps step: 9019 fps step and policy inference: 7261 fps total: 6252 epoch: 982/1500\n",
+ "fps step: 9980 fps step and policy inference: 7874 fps total: 6697 epoch: 983/1500\n",
+ "fps step: 9121 fps step and policy inference: 7328 fps total: 6281 epoch: 984/1500\n",
+ "fps step: 9425 fps step and policy inference: 7570 fps total: 6468 epoch: 985/1500\n",
+ "fps step: 8075 fps step and policy inference: 6473 fps total: 5665 epoch: 986/1500\n",
+ "fps step: 9762 fps step and policy inference: 7699 fps total: 6567 epoch: 987/1500\n",
+ "fps step: 9216 fps step and policy inference: 7317 fps total: 6283 epoch: 988/1500\n",
+ "fps step: 9073 fps step and policy inference: 7230 fps total: 6219 epoch: 989/1500\n",
+ "fps step: 9756 fps step and policy inference: 7871 fps total: 6692 epoch: 990/1500\n",
+ "fps step: 9747 fps step and policy inference: 7809 fps total: 6641 epoch: 991/1500\n",
+ "fps step: 9417 fps step and policy inference: 7412 fps total: 6351 epoch: 992/1500\n",
+ "fps step: 9522 fps step and policy inference: 7591 fps total: 6483 epoch: 993/1500\n",
+ "fps step: 9473 fps step and policy inference: 7515 fps total: 6428 epoch: 994/1500\n",
+ "fps step: 8745 fps step and policy inference: 6999 fps total: 6047 epoch: 995/1500\n",
+ "fps step: 9284 fps step and policy inference: 7335 fps total: 6295 epoch: 996/1500\n",
+ "fps step: 8394 fps step and policy inference: 6773 fps total: 5877 epoch: 997/1500\n",
+ "fps step: 9941 fps step and policy inference: 7850 fps total: 6671 epoch: 998/1500\n",
+ "fps step: 9118 fps step and policy inference: 7181 fps total: 6175 epoch: 999/1500\n",
+ "fps step: 8240 fps step and policy inference: 6403 fps total: 5596 epoch: 1000/1500\n",
+ "fps step: 9220 fps step and policy inference: 7364 fps total: 6316 epoch: 1001/1500\n",
+ "fps step: 8445 fps step and policy inference: 6700 fps total: 5821 epoch: 1002/1500\n",
+ "fps step: 10028 fps step and policy inference: 8024 fps total: 6798 epoch: 1003/1500\n",
+ "fps step: 10727 fps step and policy inference: 8547 fps total: 7170 epoch: 1004/1500\n",
+ "fps step: 9700 fps step and policy inference: 7669 fps total: 6553 epoch: 1005/1500\n",
+ "fps step: 10638 fps step and policy inference: 8385 fps total: 7056 epoch: 1006/1500\n",
+ "fps step: 8802 fps step and policy inference: 6898 fps total: 5971 epoch: 1007/1500\n",
+ "fps step: 9496 fps step and policy inference: 7525 fps total: 6415 epoch: 1008/1500\n",
+ "fps step: 8451 fps step and policy inference: 6737 fps total: 5848 epoch: 1009/1500\n",
+ "fps step: 9066 fps step and policy inference: 7113 fps total: 6128 epoch: 1010/1500\n",
+ "fps step: 9753 fps step and policy inference: 7883 fps total: 6697 epoch: 1011/1500\n",
+ "fps step: 9123 fps step and policy inference: 7288 fps total: 6274 epoch: 1012/1500\n",
+ "fps step: 11238 fps step and policy inference: 8968 fps total: 7467 epoch: 1013/1500\n",
+ "fps step: 9110 fps step and policy inference: 7259 fps total: 6239 epoch: 1014/1500\n",
+ "fps step: 8492 fps step and policy inference: 6625 fps total: 5765 epoch: 1015/1500\n",
+ "fps step: 8746 fps step and policy inference: 6896 fps total: 5970 epoch: 1016/1500\n",
+ "fps step: 8600 fps step and policy inference: 6828 fps total: 5919 epoch: 1017/1500\n",
+ "fps step: 10808 fps step and policy inference: 8601 fps total: 7207 epoch: 1018/1500\n",
+ "fps step: 10093 fps step and policy inference: 8067 fps total: 6824 epoch: 1019/1500\n",
+ "fps step: 9430 fps step and policy inference: 7291 fps total: 6249 epoch: 1020/1500\n",
+ "fps step: 10026 fps step and policy inference: 8042 fps total: 6792 epoch: 1021/1500\n",
+ "fps step: 9349 fps step and policy inference: 7402 fps total: 6337 epoch: 1022/1500\n",
+ "fps step: 9301 fps step and policy inference: 7429 fps total: 6366 epoch: 1023/1500\n",
+ "fps step: 9404 fps step and policy inference: 7528 fps total: 6423 epoch: 1024/1500\n",
+ "fps step: 7306 fps step and policy inference: 5668 fps total: 5005 epoch: 1025/1500\n",
+ "fps step: 9759 fps step and policy inference: 7774 fps total: 6615 epoch: 1026/1500\n",
+ "fps step: 9875 fps step and policy inference: 7860 fps total: 6678 epoch: 1027/1500\n",
+ "fps step: 10001 fps step and policy inference: 7899 fps total: 6707 epoch: 1028/1500\n",
+ "fps step: 10889 fps step and policy inference: 8710 fps total: 7280 epoch: 1029/1500\n",
+ "fps step: 8350 fps step and policy inference: 6569 fps total: 5722 epoch: 1030/1500\n",
+ "fps step: 9934 fps step and policy inference: 7901 fps total: 6710 epoch: 1031/1500\n",
+ "fps step: 9766 fps step and policy inference: 7741 fps total: 6590 epoch: 1032/1500\n",
+ "fps step: 10456 fps step and policy inference: 8309 fps total: 7006 epoch: 1033/1500\n",
+ "fps step: 9466 fps step and policy inference: 7523 fps total: 6435 epoch: 1034/1500\n",
+ "fps step: 8450 fps step and policy inference: 6558 fps total: 5719 epoch: 1035/1500\n",
+ "fps step: 7865 fps step and policy inference: 6089 fps total: 5344 epoch: 1036/1500\n",
+ "fps step: 8444 fps step and policy inference: 6562 fps total: 5717 epoch: 1037/1500\n",
+ "fps step: 9160 fps step and policy inference: 7293 fps total: 6259 epoch: 1038/1500\n",
+ "fps step: 9432 fps step and policy inference: 7634 fps total: 6516 epoch: 1039/1500\n",
+ "fps step: 8415 fps step and policy inference: 6616 fps total: 5759 epoch: 1040/1500\n",
+ "fps step: 10146 fps step and policy inference: 8013 fps total: 6786 epoch: 1041/1500\n",
+ "fps step: 8802 fps step and policy inference: 6999 fps total: 6045 epoch: 1042/1500\n",
+ "fps step: 9774 fps step and policy inference: 7858 fps total: 6677 epoch: 1043/1500\n",
+ "fps step: 8740 fps step and policy inference: 6930 fps total: 5977 epoch: 1044/1500\n",
+ "fps step: 9684 fps step and policy inference: 7749 fps total: 6588 epoch: 1045/1500\n",
+ "fps step: 10577 fps step and policy inference: 8426 fps total: 7086 epoch: 1046/1500\n",
+ "fps step: 9670 fps step and policy inference: 7694 fps total: 6567 epoch: 1047/1500\n",
+ "fps step: 9299 fps step and policy inference: 7328 fps total: 6292 epoch: 1048/1500\n",
+ "fps step: 9745 fps step and policy inference: 7751 fps total: 6613 epoch: 1049/1500\n",
+ "fps step: 9829 fps step and policy inference: 7862 fps total: 6681 epoch: 1050/1500\n",
+ "fps step: 8988 fps step and policy inference: 7158 fps total: 6165 epoch: 1051/1500\n",
+ "fps step: 9287 fps step and policy inference: 7542 fps total: 6450 epoch: 1052/1500\n",
+ "fps step: 9752 fps step and policy inference: 7717 fps total: 6575 epoch: 1053/1500\n",
+ "fps step: 8910 fps step and policy inference: 7022 fps total: 6067 epoch: 1054/1500\n",
+ "fps step: 9644 fps step and policy inference: 7652 fps total: 6528 epoch: 1055/1500\n",
+ "fps step: 7902 fps step and policy inference: 6127 fps total: 5378 epoch: 1056/1500\n",
+ "fps step: 8427 fps step and policy inference: 6660 fps total: 5794 epoch: 1057/1500\n",
+ "fps step: 9033 fps step and policy inference: 7037 fps total: 6078 epoch: 1058/1500\n",
+ "fps step: 8166 fps step and policy inference: 6538 fps total: 5699 epoch: 1059/1500\n",
+ "fps step: 8988 fps step and policy inference: 7125 fps total: 6169 epoch: 1060/1500\n",
+ "fps step: 9661 fps step and policy inference: 7606 fps total: 6497 epoch: 1061/1500\n",
+ "fps step: 9127 fps step and policy inference: 7290 fps total: 6264 epoch: 1062/1500\n",
+ "fps step: 9101 fps step and policy inference: 7279 fps total: 6256 epoch: 1063/1500\n",
+ "fps step: 9328 fps step and policy inference: 7423 fps total: 6359 epoch: 1064/1500\n",
+ "fps step: 8834 fps step and policy inference: 7100 fps total: 6109 epoch: 1065/1500\n",
+ "fps step: 9155 fps step and policy inference: 7387 fps total: 6326 epoch: 1066/1500\n",
+ "fps step: 8719 fps step and policy inference: 6861 fps total: 5944 epoch: 1067/1500\n",
+ "fps step: 8105 fps step and policy inference: 6354 fps total: 5552 epoch: 1068/1500\n",
+ "fps step: 9923 fps step and policy inference: 7889 fps total: 6705 epoch: 1069/1500\n",
+ "fps step: 9010 fps step and policy inference: 7211 fps total: 6208 epoch: 1070/1500\n",
+ "fps step: 9794 fps step and policy inference: 7917 fps total: 6720 epoch: 1071/1500\n",
+ "fps step: 10636 fps step and policy inference: 8517 fps total: 7156 epoch: 1072/1500\n",
+ "fps step: 8619 fps step and policy inference: 6764 fps total: 5881 epoch: 1073/1500\n",
+ "fps step: 9892 fps step and policy inference: 7854 fps total: 6680 epoch: 1074/1500\n",
+ "fps step: 9293 fps step and policy inference: 7284 fps total: 6259 epoch: 1075/1500\n",
+ "fps step: 9691 fps step and policy inference: 7739 fps total: 6584 epoch: 1076/1500\n",
+ "fps step: 9912 fps step and policy inference: 8008 fps total: 6788 epoch: 1077/1500\n",
+ "fps step: 7492 fps step and policy inference: 5791 fps total: 5109 epoch: 1078/1500\n",
+ "fps step: 8100 fps step and policy inference: 6296 fps total: 5512 epoch: 1079/1500\n",
+ "fps step: 8995 fps step and policy inference: 7114 fps total: 6137 epoch: 1080/1500\n",
+ "fps step: 10544 fps step and policy inference: 8386 fps total: 7055 epoch: 1081/1500\n",
+ "fps step: 8536 fps step and policy inference: 6708 fps total: 5827 epoch: 1082/1500\n",
+ "fps step: 8188 fps step and policy inference: 6550 fps total: 5709 epoch: 1083/1500\n",
+ "fps step: 9272 fps step and policy inference: 7309 fps total: 6276 epoch: 1084/1500\n",
+ "fps step: 9214 fps step and policy inference: 7296 fps total: 6268 epoch: 1085/1500\n",
+ "fps step: 9995 fps step and policy inference: 7926 fps total: 6727 epoch: 1086/1500\n",
+ "fps step: 10844 fps step and policy inference: 8466 fps total: 7092 epoch: 1087/1500\n",
+ "fps step: 9938 fps step and policy inference: 8003 fps total: 6783 epoch: 1088/1500\n",
+ "fps step: 11759 fps step and policy inference: 9282 fps total: 7655 epoch: 1089/1500\n",
+ "fps step: 9423 fps step and policy inference: 7496 fps total: 6414 epoch: 1090/1500\n",
+ "fps step: 10311 fps step and policy inference: 8157 fps total: 6882 epoch: 1091/1500\n",
+ "fps step: 9018 fps step and policy inference: 7090 fps total: 6096 epoch: 1092/1500\n",
+ "fps step: 9981 fps step and policy inference: 7848 fps total: 6648 epoch: 1093/1500\n",
+ "fps step: 8629 fps step and policy inference: 6911 fps total: 5981 epoch: 1094/1500\n",
+ "fps step: 8846 fps step and policy inference: 6967 fps total: 6007 epoch: 1095/1500\n",
+ "fps step: 9849 fps step and policy inference: 7892 fps total: 6700 epoch: 1096/1500\n",
+ "fps step: 10481 fps step and policy inference: 8231 fps total: 6933 epoch: 1097/1500\n",
+ "fps step: 9500 fps step and policy inference: 7501 fps total: 6405 epoch: 1098/1500\n",
+ "fps step: 10424 fps step and policy inference: 8167 fps total: 6898 epoch: 1099/1500\n",
+ "fps step: 9344 fps step and policy inference: 7286 fps total: 6256 epoch: 1100/1500\n",
+ "fps step: 9392 fps step and policy inference: 7421 fps total: 6358 epoch: 1101/1500\n",
+ "fps step: 10718 fps step and policy inference: 8528 fps total: 7154 epoch: 1102/1500\n",
+ "fps step: 10682 fps step and policy inference: 8449 fps total: 7099 epoch: 1103/1500\n",
+ "fps step: 8923 fps step and policy inference: 7166 fps total: 6171 epoch: 1104/1500\n",
+ "fps step: 9530 fps step and policy inference: 7535 fps total: 6440 epoch: 1105/1500\n",
+ "fps step: 9470 fps step and policy inference: 7365 fps total: 6319 epoch: 1106/1500\n",
+ "fps step: 9862 fps step and policy inference: 7733 fps total: 6592 epoch: 1107/1500\n",
+ "fps step: 9508 fps step and policy inference: 7614 fps total: 6501 epoch: 1108/1500\n",
+ "fps step: 8628 fps step and policy inference: 6734 fps total: 5832 epoch: 1109/1500\n",
+ "fps step: 8824 fps step and policy inference: 6941 fps total: 6004 epoch: 1110/1500\n",
+ "fps step: 10549 fps step and policy inference: 8326 fps total: 7011 epoch: 1111/1500\n",
+ "fps step: 9433 fps step and policy inference: 7470 fps total: 6394 epoch: 1112/1500\n",
+ "fps step: 9173 fps step and policy inference: 7287 fps total: 6260 epoch: 1113/1500\n",
+ "fps step: 9092 fps step and policy inference: 7183 fps total: 6186 epoch: 1114/1500\n",
+ "fps step: 9998 fps step and policy inference: 7971 fps total: 6745 epoch: 1115/1500\n",
+ "fps step: 8626 fps step and policy inference: 6792 fps total: 5892 epoch: 1116/1500\n",
+ "fps step: 9748 fps step and policy inference: 7804 fps total: 6638 epoch: 1117/1500\n",
+ "fps step: 9526 fps step and policy inference: 7679 fps total: 6543 epoch: 1118/1500\n",
+ "fps step: 9962 fps step and policy inference: 7803 fps total: 6633 epoch: 1119/1500\n",
+ "fps step: 8332 fps step and policy inference: 6610 fps total: 5754 epoch: 1120/1500\n",
+ "fps step: 9952 fps step and policy inference: 7939 fps total: 6737 epoch: 1121/1500\n",
+ "fps step: 9887 fps step and policy inference: 7828 fps total: 6637 epoch: 1122/1500\n",
+ "fps step: 10545 fps step and policy inference: 8403 fps total: 7065 epoch: 1123/1500\n",
+ "fps step: 10271 fps step and policy inference: 8141 fps total: 6880 epoch: 1124/1500\n",
+ "fps step: 9991 fps step and policy inference: 7902 fps total: 6709 epoch: 1125/1500\n",
+ "fps step: 9694 fps step and policy inference: 7688 fps total: 6557 epoch: 1126/1500\n",
+ "fps step: 10449 fps step and policy inference: 8386 fps total: 7029 epoch: 1127/1500\n",
+ "fps step: 8071 fps step and policy inference: 6315 fps total: 5533 epoch: 1128/1500\n",
+ "fps step: 10305 fps step and policy inference: 8238 fps total: 6961 epoch: 1129/1500\n",
+ "fps step: 8148 fps step and policy inference: 6573 fps total: 5727 epoch: 1130/1500\n",
+ "fps step: 8489 fps step and policy inference: 6714 fps total: 5833 epoch: 1131/1500\n",
+ "fps step: 8743 fps step and policy inference: 6886 fps total: 5966 epoch: 1132/1500\n",
+ "fps step: 9259 fps step and policy inference: 7467 fps total: 6378 epoch: 1133/1500\n",
+ "fps step: 8982 fps step and policy inference: 7070 fps total: 6100 epoch: 1134/1500\n",
+ "fps step: 9705 fps step and policy inference: 7748 fps total: 6597 epoch: 1135/1500\n",
+ "fps step: 9863 fps step and policy inference: 7809 fps total: 6639 epoch: 1136/1500\n",
+ "fps step: 9206 fps step and policy inference: 7251 fps total: 6234 epoch: 1137/1500\n",
+ "fps step: 10368 fps step and policy inference: 8197 fps total: 6902 epoch: 1138/1500\n",
+ "fps step: 9180 fps step and policy inference: 7280 fps total: 6255 epoch: 1139/1500\n",
+ "fps step: 10895 fps step and policy inference: 8589 fps total: 7198 epoch: 1140/1500\n",
+ "fps step: 10178 fps step and policy inference: 8200 fps total: 6923 epoch: 1141/1500\n",
+ "fps step: 9619 fps step and policy inference: 7523 fps total: 6429 epoch: 1142/1500\n",
+ "fps step: 9375 fps step and policy inference: 7534 fps total: 6447 epoch: 1143/1500\n",
+ "fps step: 9679 fps step and policy inference: 7646 fps total: 6525 epoch: 1144/1500\n",
+ "fps step: 9947 fps step and policy inference: 7802 fps total: 6613 epoch: 1145/1500\n",
+ "fps step: 9830 fps step and policy inference: 7795 fps total: 6628 epoch: 1146/1500\n",
+ "fps step: 10580 fps step and policy inference: 8479 fps total: 7123 epoch: 1147/1500\n",
+ "fps step: 9410 fps step and policy inference: 7468 fps total: 6393 epoch: 1148/1500\n",
+ "fps step: 10484 fps step and policy inference: 8268 fps total: 6977 epoch: 1149/1500\n",
+ "fps step: 9019 fps step and policy inference: 7092 fps total: 6118 epoch: 1150/1500\n",
+ "fps step: 9211 fps step and policy inference: 7352 fps total: 6309 epoch: 1151/1500\n",
+ "fps step: 9474 fps step and policy inference: 7531 fps total: 6439 epoch: 1152/1500\n",
+ "fps step: 10325 fps step and policy inference: 8127 fps total: 6868 epoch: 1153/1500\n",
+ "fps step: 8768 fps step and policy inference: 6854 fps total: 5937 epoch: 1154/1500\n",
+ "fps step: 8404 fps step and policy inference: 6649 fps total: 5782 epoch: 1155/1500\n",
+ "fps step: 9767 fps step and policy inference: 7744 fps total: 6595 epoch: 1156/1500\n",
+ "fps step: 10587 fps step and policy inference: 8366 fps total: 7040 epoch: 1157/1500\n",
+ "fps step: 10413 fps step and policy inference: 8187 fps total: 6914 epoch: 1158/1500\n",
+ "fps step: 11282 fps step and policy inference: 8964 fps total: 7467 epoch: 1159/1500\n",
+ "fps step: 8518 fps step and policy inference: 6694 fps total: 5817 epoch: 1160/1500\n",
+ "fps step: 9901 fps step and policy inference: 7828 fps total: 6656 epoch: 1161/1500\n",
+ "fps step: 8234 fps step and policy inference: 6544 fps total: 5705 epoch: 1162/1500\n",
+ "fps step: 10351 fps step and policy inference: 8171 fps total: 6900 epoch: 1163/1500\n",
+ "fps step: 10338 fps step and policy inference: 8258 fps total: 6978 epoch: 1164/1500\n",
+ "fps step: 8144 fps step and policy inference: 6288 fps total: 5513 epoch: 1165/1500\n",
+ "fps step: 10132 fps step and policy inference: 7999 fps total: 6778 epoch: 1166/1500\n",
+ "fps step: 10315 fps step and policy inference: 8164 fps total: 6896 epoch: 1167/1500\n",
+ "fps step: 9280 fps step and policy inference: 7306 fps total: 6276 epoch: 1168/1500\n",
+ "fps step: 10325 fps step and policy inference: 8107 fps total: 6855 epoch: 1169/1500\n",
+ "fps step: 10758 fps step and policy inference: 8590 fps total: 7198 epoch: 1170/1500\n",
+ "fps step: 9188 fps step and policy inference: 7329 fps total: 6299 epoch: 1171/1500\n",
+ "fps step: 8970 fps step and policy inference: 7001 fps total: 6048 epoch: 1172/1500\n",
+ "fps step: 8052 fps step and policy inference: 6273 fps total: 5499 epoch: 1173/1500\n",
+ "fps step: 9240 fps step and policy inference: 7300 fps total: 6276 epoch: 1174/1500\n",
+ "fps step: 9157 fps step and policy inference: 7277 fps total: 6253 epoch: 1175/1500\n",
+ "fps step: 9627 fps step and policy inference: 7507 fps total: 6418 epoch: 1176/1500\n",
+ "fps step: 9956 fps step and policy inference: 7960 fps total: 6750 epoch: 1177/1500\n",
+ "fps step: 9337 fps step and policy inference: 7318 fps total: 6268 epoch: 1178/1500\n",
+ "fps step: 9050 fps step and policy inference: 7138 fps total: 6150 epoch: 1179/1500\n",
+ "fps step: 9857 fps step and policy inference: 7735 fps total: 6588 epoch: 1180/1500\n",
+ "fps step: 9723 fps step and policy inference: 7618 fps total: 6501 epoch: 1181/1500\n",
+ "fps step: 10551 fps step and policy inference: 8456 fps total: 7109 epoch: 1182/1500\n",
+ "fps step: 9514 fps step and policy inference: 7500 fps total: 6416 epoch: 1183/1500\n",
+ "fps step: 10049 fps step and policy inference: 7889 fps total: 6699 epoch: 1184/1500\n",
+ "fps step: 9830 fps step and policy inference: 7770 fps total: 6615 epoch: 1185/1500\n",
+ "fps step: 9892 fps step and policy inference: 7787 fps total: 6627 epoch: 1186/1500\n",
+ "fps step: 10172 fps step and policy inference: 8083 fps total: 6837 epoch: 1187/1500\n",
+ "fps step: 10344 fps step and policy inference: 8199 fps total: 6928 epoch: 1188/1500\n",
+ "fps step: 10723 fps step and policy inference: 8482 fps total: 7122 epoch: 1189/1500\n",
+ "fps step: 9931 fps step and policy inference: 7840 fps total: 6666 epoch: 1190/1500\n",
+ "fps step: 9017 fps step and policy inference: 7116 fps total: 6136 epoch: 1191/1500\n",
+ "fps step: 10255 fps step and policy inference: 8155 fps total: 6890 epoch: 1192/1500\n",
+ "fps step: 10663 fps step and policy inference: 8495 fps total: 7131 epoch: 1193/1500\n",
+ "fps step: 10276 fps step and policy inference: 8058 fps total: 6822 epoch: 1194/1500\n",
+ "fps step: 9545 fps step and policy inference: 7647 fps total: 6524 epoch: 1195/1500\n",
+ "fps step: 9716 fps step and policy inference: 7681 fps total: 6549 epoch: 1196/1500\n",
+ "fps step: 10236 fps step and policy inference: 8152 fps total: 6889 epoch: 1197/1500\n",
+ "fps step: 11097 fps step and policy inference: 8706 fps total: 7267 epoch: 1198/1500\n",
+ "fps step: 8658 fps step and policy inference: 6794 fps total: 5900 epoch: 1199/1500\n",
+ "fps step: 9650 fps step and policy inference: 7627 fps total: 6509 epoch: 1200/1500\n",
+ "fps step: 9538 fps step and policy inference: 7499 fps total: 6417 epoch: 1201/1500\n",
+ "fps step: 9938 fps step and policy inference: 7867 fps total: 6663 epoch: 1202/1500\n",
+ "fps step: 8403 fps step and policy inference: 6552 fps total: 5709 epoch: 1203/1500\n",
+ "fps step: 9594 fps step and policy inference: 7511 fps total: 6424 epoch: 1204/1500\n",
+ "fps step: 11182 fps step and policy inference: 8965 fps total: 7433 epoch: 1205/1500\n",
+ "fps step: 8698 fps step and policy inference: 6818 fps total: 5913 epoch: 1206/1500\n",
+ "fps step: 9324 fps step and policy inference: 7379 fps total: 6327 epoch: 1207/1500\n",
+ "fps step: 10032 fps step and policy inference: 8000 fps total: 6780 epoch: 1208/1500\n",
+ "fps step: 8479 fps step and policy inference: 6712 fps total: 5830 epoch: 1209/1500\n",
+ "fps step: 9089 fps step and policy inference: 7144 fps total: 6156 epoch: 1210/1500\n",
+ "fps step: 10001 fps step and policy inference: 7930 fps total: 6726 epoch: 1211/1500\n",
+ "fps step: 10264 fps step and policy inference: 8140 fps total: 6877 epoch: 1212/1500\n",
+ "fps step: 10926 fps step and policy inference: 8697 fps total: 7268 epoch: 1213/1500\n",
+ "fps step: 10678 fps step and policy inference: 8441 fps total: 7093 epoch: 1214/1500\n",
+ "fps step: 9185 fps step and policy inference: 7210 fps total: 6194 epoch: 1215/1500\n",
+ "fps step: 10087 fps step and policy inference: 8026 fps total: 6798 epoch: 1216/1500\n",
+ "fps step: 10284 fps step and policy inference: 8145 fps total: 6884 epoch: 1217/1500\n",
+ "fps step: 9313 fps step and policy inference: 7498 fps total: 6416 epoch: 1218/1500\n",
+ "fps step: 9105 fps step and policy inference: 7278 fps total: 6253 epoch: 1219/1500\n",
+ "fps step: 8425 fps step and policy inference: 6683 fps total: 5807 epoch: 1220/1500\n",
+ "fps step: 9366 fps step and policy inference: 7508 fps total: 6423 epoch: 1221/1500\n",
+ "fps step: 9086 fps step and policy inference: 7103 fps total: 6124 epoch: 1222/1500\n",
+ "fps step: 10796 fps step and policy inference: 8577 fps total: 7188 epoch: 1223/1500\n",
+ "fps step: 8987 fps step and policy inference: 7118 fps total: 6133 epoch: 1224/1500\n",
+ "fps step: 8119 fps step and policy inference: 6392 fps total: 5590 epoch: 1225/1500\n",
+ "fps step: 9350 fps step and policy inference: 7682 fps total: 6563 epoch: 1226/1500\n",
+ "fps step: 8755 fps step and policy inference: 6927 fps total: 5993 epoch: 1227/1500\n",
+ "fps step: 9787 fps step and policy inference: 7784 fps total: 6623 epoch: 1228/1500\n",
+ "fps step: 10179 fps step and policy inference: 8122 fps total: 6824 epoch: 1229/1500\n",
+ "fps step: 9720 fps step and policy inference: 7721 fps total: 6586 epoch: 1230/1500\n",
+ "fps step: 8085 fps step and policy inference: 6405 fps total: 5597 epoch: 1231/1500\n",
+ "fps step: 10510 fps step and policy inference: 8398 fps total: 7067 epoch: 1232/1500\n",
+ "fps step: 9480 fps step and policy inference: 7497 fps total: 6417 epoch: 1233/1500\n",
+ "fps step: 9539 fps step and policy inference: 7582 fps total: 6479 epoch: 1234/1500\n",
+ "fps step: 10523 fps step and policy inference: 8322 fps total: 7020 epoch: 1235/1500\n",
+ "fps step: 9164 fps step and policy inference: 7280 fps total: 6262 epoch: 1236/1500\n",
+ "fps step: 8675 fps step and policy inference: 6890 fps total: 5965 epoch: 1237/1500\n",
+ "fps step: 8750 fps step and policy inference: 7018 fps total: 6066 epoch: 1238/1500\n",
+ "fps step: 10270 fps step and policy inference: 8023 fps total: 6797 epoch: 1239/1500\n",
+ "fps step: 9178 fps step and policy inference: 7300 fps total: 6270 epoch: 1240/1500\n",
+ "fps step: 8600 fps step and policy inference: 6895 fps total: 5970 epoch: 1241/1500\n",
+ "fps step: 10562 fps step and policy inference: 8432 fps total: 7090 epoch: 1242/1500\n",
+ "fps step: 8849 fps step and policy inference: 7055 fps total: 6046 epoch: 1243/1500\n",
+ "fps step: 8779 fps step and policy inference: 6958 fps total: 6015 epoch: 1244/1500\n",
+ "fps step: 9704 fps step and policy inference: 7822 fps total: 6648 epoch: 1245/1500\n",
+ "fps step: 10002 fps step and policy inference: 7998 fps total: 6771 epoch: 1246/1500\n",
+ "fps step: 8798 fps step and policy inference: 6978 fps total: 6030 epoch: 1247/1500\n",
+ "fps step: 8994 fps step and policy inference: 7146 fps total: 6135 epoch: 1248/1500\n",
+ "fps step: 9571 fps step and policy inference: 7626 fps total: 6485 epoch: 1249/1500\n",
+ "fps step: 10156 fps step and policy inference: 8143 fps total: 6883 epoch: 1250/1500\n",
+ "fps step: 8555 fps step and policy inference: 6740 fps total: 5844 epoch: 1251/1500\n",
+ "fps step: 9082 fps step and policy inference: 7326 fps total: 6288 epoch: 1252/1500\n",
+ "fps step: 8397 fps step and policy inference: 6657 fps total: 5794 epoch: 1253/1500\n",
+ "fps step: 9842 fps step and policy inference: 7823 fps total: 6651 epoch: 1254/1500\n",
+ "fps step: 9709 fps step and policy inference: 7780 fps total: 6629 epoch: 1255/1500\n",
+ "fps step: 9124 fps step and policy inference: 7333 fps total: 6302 epoch: 1256/1500\n",
+ "fps step: 7466 fps step and policy inference: 5815 fps total: 5141 epoch: 1257/1500\n",
+ "fps step: 8198 fps step and policy inference: 6478 fps total: 5652 epoch: 1258/1500\n",
+ "fps step: 10525 fps step and policy inference: 8434 fps total: 7092 epoch: 1259/1500\n",
+ "fps step: 9044 fps step and policy inference: 7254 fps total: 6206 epoch: 1260/1500\n",
+ "fps step: 11139 fps step and policy inference: 8824 fps total: 7373 epoch: 1261/1500\n",
+ "fps step: 9884 fps step and policy inference: 7943 fps total: 6752 epoch: 1262/1500\n",
+ "fps step: 7708 fps step and policy inference: 6117 fps total: 5378 epoch: 1263/1500\n",
+ "fps step: 9372 fps step and policy inference: 7418 fps total: 6357 epoch: 1264/1500\n",
+ "fps step: 9518 fps step and policy inference: 7559 fps total: 6461 epoch: 1265/1500\n",
+ "fps step: 10243 fps step and policy inference: 8224 fps total: 6941 epoch: 1266/1500\n",
+ "fps step: 9414 fps step and policy inference: 7463 fps total: 6386 epoch: 1267/1500\n",
+ "fps step: 8728 fps step and policy inference: 6857 fps total: 5943 epoch: 1268/1500\n",
+ "fps step: 8136 fps step and policy inference: 6370 fps total: 5572 epoch: 1269/1500\n",
+ "fps step: 8398 fps step and policy inference: 6464 fps total: 5634 epoch: 1270/1500\n",
+ "fps step: 11085 fps step and policy inference: 8830 fps total: 7367 epoch: 1271/1500\n",
+ "fps step: 9697 fps step and policy inference: 7653 fps total: 6530 epoch: 1272/1500\n",
+ "fps step: 8961 fps step and policy inference: 7009 fps total: 6056 epoch: 1273/1500\n",
+ "fps step: 9288 fps step and policy inference: 7334 fps total: 6295 epoch: 1274/1500\n",
+ "fps step: 9301 fps step and policy inference: 7294 fps total: 6267 epoch: 1275/1500\n",
+ "fps step: 11132 fps step and policy inference: 8844 fps total: 7384 epoch: 1276/1500\n",
+ "fps step: 9958 fps step and policy inference: 7975 fps total: 6766 epoch: 1277/1500\n",
+ "fps step: 8294 fps step and policy inference: 6390 fps total: 5590 epoch: 1278/1500\n",
+ "fps step: 9425 fps step and policy inference: 7400 fps total: 6352 epoch: 1279/1500\n",
+ "fps step: 9297 fps step and policy inference: 7430 fps total: 6372 epoch: 1280/1500\n",
+ "fps step: 9640 fps step and policy inference: 7595 fps total: 6485 epoch: 1281/1500\n",
+ "fps step: 9960 fps step and policy inference: 7909 fps total: 6714 epoch: 1282/1500\n",
+ "fps step: 10946 fps step and policy inference: 8703 fps total: 7265 epoch: 1283/1500\n",
+ "fps step: 8842 fps step and policy inference: 7029 fps total: 6074 epoch: 1284/1500\n",
+ "fps step: 10187 fps step and policy inference: 8084 fps total: 6846 epoch: 1285/1500\n",
+ "fps step: 10204 fps step and policy inference: 8182 fps total: 6890 epoch: 1286/1500\n",
+ "fps step: 9694 fps step and policy inference: 7752 fps total: 6602 epoch: 1287/1500\n",
+ "fps step: 9911 fps step and policy inference: 7884 fps total: 6697 epoch: 1288/1500\n",
+ "fps step: 9049 fps step and policy inference: 7234 fps total: 6222 epoch: 1289/1500\n",
+ "fps step: 10098 fps step and policy inference: 8084 fps total: 6842 epoch: 1290/1500\n",
+ "fps step: 9172 fps step and policy inference: 7296 fps total: 6264 epoch: 1291/1500\n",
+ "fps step: 8809 fps step and policy inference: 7028 fps total: 6067 epoch: 1292/1500\n",
+ "fps step: 10637 fps step and policy inference: 8314 fps total: 6982 epoch: 1293/1500\n",
+ "fps step: 9074 fps step and policy inference: 7131 fps total: 6146 epoch: 1294/1500\n",
+ "fps step: 10338 fps step and policy inference: 8125 fps total: 6842 epoch: 1295/1500\n",
+ "fps step: 9558 fps step and policy inference: 7451 fps total: 6379 epoch: 1296/1500\n",
+ "fps step: 10053 fps step and policy inference: 7902 fps total: 6712 epoch: 1297/1500\n",
+ "fps step: 9171 fps step and policy inference: 7234 fps total: 6220 epoch: 1298/1500\n",
+ "fps step: 9512 fps step and policy inference: 7611 fps total: 6498 epoch: 1299/1500\n",
+ "fps step: 10103 fps step and policy inference: 8059 fps total: 6823 epoch: 1300/1500\n",
+ "fps step: 8922 fps step and policy inference: 6992 fps total: 6040 epoch: 1301/1500\n",
+ "fps step: 9920 fps step and policy inference: 7829 fps total: 6639 epoch: 1302/1500\n",
+ "fps step: 8928 fps step and policy inference: 6942 fps total: 6005 epoch: 1303/1500\n",
+ "fps step: 10702 fps step and policy inference: 8446 fps total: 7098 epoch: 1304/1500\n",
+ "fps step: 11239 fps step and policy inference: 8833 fps total: 7369 epoch: 1305/1500\n",
+ "fps step: 10395 fps step and policy inference: 8203 fps total: 6914 epoch: 1306/1500\n",
+ "fps step: 8452 fps step and policy inference: 6684 fps total: 5810 epoch: 1307/1500\n",
+ "fps step: 9915 fps step and policy inference: 7929 fps total: 6730 epoch: 1308/1500\n",
+ "fps step: 9285 fps step and policy inference: 7401 fps total: 6343 epoch: 1309/1500\n",
+ "fps step: 9652 fps step and policy inference: 7606 fps total: 6493 epoch: 1310/1500\n",
+ "fps step: 10251 fps step and policy inference: 8240 fps total: 6951 epoch: 1311/1500\n",
+ "fps step: 9079 fps step and policy inference: 7187 fps total: 6195 epoch: 1312/1500\n",
+ "fps step: 9674 fps step and policy inference: 7541 fps total: 6447 epoch: 1313/1500\n",
+ "fps step: 9662 fps step and policy inference: 7670 fps total: 6541 epoch: 1314/1500\n",
+ "fps step: 9133 fps step and policy inference: 7177 fps total: 6185 epoch: 1315/1500\n",
+ "fps step: 9734 fps step and policy inference: 7652 fps total: 6526 epoch: 1316/1500\n",
+ "fps step: 10349 fps step and policy inference: 8284 fps total: 6983 epoch: 1317/1500\n",
+ "fps step: 8847 fps step and policy inference: 6918 fps total: 5987 epoch: 1318/1500\n",
+ "fps step: 9869 fps step and policy inference: 7843 fps total: 6644 epoch: 1319/1500\n",
+ "fps step: 9547 fps step and policy inference: 7488 fps total: 6414 epoch: 1320/1500\n",
+ "fps step: 9683 fps step and policy inference: 7556 fps total: 6458 epoch: 1321/1500\n",
+ "fps step: 8682 fps step and policy inference: 6744 fps total: 5856 epoch: 1322/1500\n",
+ "fps step: 8761 fps step and policy inference: 6809 fps total: 5889 epoch: 1323/1500\n",
+ "fps step: 9972 fps step and policy inference: 7907 fps total: 6727 epoch: 1324/1500\n",
+ "fps step: 9092 fps step and policy inference: 7298 fps total: 6269 epoch: 1325/1500\n",
+ "fps step: 9460 fps step and policy inference: 7389 fps total: 6341 epoch: 1326/1500\n",
+ "fps step: 10777 fps step and policy inference: 8550 fps total: 7176 epoch: 1327/1500\n",
+ "fps step: 9708 fps step and policy inference: 7768 fps total: 6614 epoch: 1328/1500\n",
+ "fps step: 8862 fps step and policy inference: 6881 fps total: 5963 epoch: 1329/1500\n",
+ "fps step: 9157 fps step and policy inference: 7199 fps total: 6197 epoch: 1330/1500\n",
+ "fps step: 9555 fps step and policy inference: 7577 fps total: 6476 epoch: 1331/1500\n",
+ "fps step: 10039 fps step and policy inference: 7865 fps total: 6681 epoch: 1332/1500\n",
+ "fps step: 8975 fps step and policy inference: 7044 fps total: 6077 epoch: 1333/1500\n",
+ "fps step: 9790 fps step and policy inference: 7811 fps total: 6648 epoch: 1334/1500\n",
+ "fps step: 10558 fps step and policy inference: 8305 fps total: 7001 epoch: 1335/1500\n",
+ "fps step: 9680 fps step and policy inference: 7632 fps total: 6522 epoch: 1336/1500\n",
+ "fps step: 10539 fps step and policy inference: 8213 fps total: 6933 epoch: 1337/1500\n",
+ "fps step: 9603 fps step and policy inference: 7592 fps total: 6492 epoch: 1338/1500\n",
+ "fps step: 10512 fps step and policy inference: 8289 fps total: 6954 epoch: 1339/1500\n",
+ "fps step: 10099 fps step and policy inference: 7911 fps total: 6715 epoch: 1340/1500\n",
+ "fps step: 10330 fps step and policy inference: 8025 fps total: 6797 epoch: 1341/1500\n",
+ "fps step: 9605 fps step and policy inference: 7663 fps total: 6528 epoch: 1342/1500\n",
+ "fps step: 8403 fps step and policy inference: 6655 fps total: 5792 epoch: 1343/1500\n",
+ "fps step: 9530 fps step and policy inference: 7666 fps total: 6541 epoch: 1344/1500\n",
+ "fps step: 9362 fps step and policy inference: 7352 fps total: 6314 epoch: 1345/1500\n",
+ "fps step: 9624 fps step and policy inference: 7728 fps total: 6587 epoch: 1346/1500\n",
+ "fps step: 9853 fps step and policy inference: 7817 fps total: 6638 epoch: 1347/1500\n",
+ "fps step: 10151 fps step and policy inference: 7962 fps total: 6753 epoch: 1348/1500\n",
+ "fps step: 10124 fps step and policy inference: 7994 fps total: 6766 epoch: 1349/1500\n",
+ "fps step: 9774 fps step and policy inference: 7859 fps total: 6679 epoch: 1350/1500\n",
+ "fps step: 9437 fps step and policy inference: 7461 fps total: 6388 epoch: 1351/1500\n",
+ "fps step: 10064 fps step and policy inference: 7919 fps total: 6711 epoch: 1352/1500\n",
+ "fps step: 9908 fps step and policy inference: 7727 fps total: 6571 epoch: 1353/1500\n",
+ "fps step: 8788 fps step and policy inference: 6803 fps total: 5903 epoch: 1354/1500\n",
+ "fps step: 11135 fps step and policy inference: 8751 fps total: 7313 epoch: 1355/1500\n",
+ "fps step: 9600 fps step and policy inference: 7572 fps total: 6470 epoch: 1356/1500\n",
+ "fps step: 10715 fps step and policy inference: 8558 fps total: 7178 epoch: 1357/1500\n",
+ "fps step: 9642 fps step and policy inference: 7678 fps total: 6549 epoch: 1358/1500\n",
+ "fps step: 11075 fps step and policy inference: 8817 fps total: 7363 epoch: 1359/1500\n",
+ "fps step: 9176 fps step and policy inference: 7158 fps total: 6166 epoch: 1360/1500\n",
+ "fps step: 9310 fps step and policy inference: 7352 fps total: 6302 epoch: 1361/1500\n",
+ "fps step: 9255 fps step and policy inference: 7230 fps total: 6220 epoch: 1362/1500\n",
+ "fps step: 9753 fps step and policy inference: 7708 fps total: 6570 epoch: 1363/1500\n",
+ "fps step: 9615 fps step and policy inference: 7529 fps total: 6438 epoch: 1364/1500\n",
+ "fps step: 10302 fps step and policy inference: 8144 fps total: 6884 epoch: 1365/1500\n",
+ "fps step: 9827 fps step and policy inference: 7857 fps total: 6651 epoch: 1366/1500\n",
+ "fps step: 10240 fps step and policy inference: 8028 fps total: 6799 epoch: 1367/1500\n",
+ "fps step: 9543 fps step and policy inference: 7560 fps total: 6462 epoch: 1368/1500\n",
+ "fps step: 9227 fps step and policy inference: 7244 fps total: 6227 epoch: 1369/1500\n",
+ "fps step: 10012 fps step and policy inference: 7949 fps total: 6741 epoch: 1370/1500\n",
+ "fps step: 8941 fps step and policy inference: 6994 fps total: 6046 epoch: 1371/1500\n",
+ "fps step: 9693 fps step and policy inference: 7653 fps total: 6534 epoch: 1372/1500\n",
+ "fps step: 9076 fps step and policy inference: 7323 fps total: 6285 epoch: 1373/1500\n",
+ "fps step: 10021 fps step and policy inference: 8061 fps total: 6812 epoch: 1374/1500\n",
+ "fps step: 9922 fps step and policy inference: 7812 fps total: 6633 epoch: 1375/1500\n",
+ "fps step: 8476 fps step and policy inference: 6599 fps total: 5750 epoch: 1376/1500\n",
+ "fps step: 10252 fps step and policy inference: 8091 fps total: 6844 epoch: 1377/1500\n",
+ "fps step: 10727 fps step and policy inference: 8561 fps total: 7180 epoch: 1378/1500\n",
+ "fps step: 7882 fps step and policy inference: 6093 fps total: 5348 epoch: 1379/1500\n",
+ "fps step: 8933 fps step and policy inference: 6970 fps total: 6025 epoch: 1380/1500\n",
+ "fps step: 10517 fps step and policy inference: 8244 fps total: 6953 epoch: 1381/1500\n",
+ "fps step: 9634 fps step and policy inference: 7557 fps total: 6451 epoch: 1382/1500\n",
+ "fps step: 11736 fps step and policy inference: 9341 fps total: 7696 epoch: 1383/1500\n",
+ "fps step: 9801 fps step and policy inference: 7948 fps total: 6749 epoch: 1384/1500\n",
+ "fps step: 9914 fps step and policy inference: 7798 fps total: 6622 epoch: 1385/1500\n",
+ "fps step: 9993 fps step and policy inference: 7910 fps total: 6716 epoch: 1386/1500\n",
+ "fps step: 9625 fps step and policy inference: 7589 fps total: 6481 epoch: 1387/1500\n",
+ "fps step: 10526 fps step and policy inference: 8382 fps total: 7052 epoch: 1388/1500\n",
+ "fps step: 10519 fps step and policy inference: 8315 fps total: 7006 epoch: 1389/1500\n",
+ "fps step: 8891 fps step and policy inference: 7042 fps total: 6065 epoch: 1390/1500\n",
+ "fps step: 10608 fps step and policy inference: 8421 fps total: 7083 epoch: 1391/1500\n",
+ "fps step: 9295 fps step and policy inference: 7250 fps total: 6234 epoch: 1392/1500\n",
+ "fps step: 8993 fps step and policy inference: 7008 fps total: 6035 epoch: 1393/1500\n",
+ "fps step: 9619 fps step and policy inference: 7518 fps total: 6430 epoch: 1394/1500\n",
+ "fps step: 9738 fps step and policy inference: 7626 fps total: 6491 epoch: 1395/1500\n",
+ "fps step: 8274 fps step and policy inference: 6466 fps total: 5647 epoch: 1396/1500\n",
+ "fps step: 9097 fps step and policy inference: 7045 fps total: 6082 epoch: 1397/1500\n",
+ "fps step: 10321 fps step and policy inference: 8220 fps total: 6938 epoch: 1398/1500\n",
+ "fps step: 9485 fps step and policy inference: 7485 fps total: 6412 epoch: 1399/1500\n",
+ "fps step: 11534 fps step and policy inference: 9162 fps total: 7596 epoch: 1400/1500\n",
+ "fps step: 11105 fps step and policy inference: 8792 fps total: 7341 epoch: 1401/1500\n",
+ "fps step: 8409 fps step and policy inference: 6472 fps total: 5629 epoch: 1402/1500\n",
+ "fps step: 9876 fps step and policy inference: 7870 fps total: 6687 epoch: 1403/1500\n",
+ "fps step: 9532 fps step and policy inference: 7511 fps total: 6426 epoch: 1404/1500\n",
+ "fps step: 9792 fps step and policy inference: 7816 fps total: 6647 epoch: 1405/1500\n",
+ "fps step: 10924 fps step and policy inference: 8651 fps total: 7243 epoch: 1406/1500\n",
+ "fps step: 9970 fps step and policy inference: 7987 fps total: 6772 epoch: 1407/1500\n",
+ "fps step: 10609 fps step and policy inference: 8388 fps total: 7059 epoch: 1408/1500\n",
+ "fps step: 9882 fps step and policy inference: 7847 fps total: 6672 epoch: 1409/1500\n",
+ "fps step: 9289 fps step and policy inference: 7171 fps total: 6174 epoch: 1410/1500\n",
+ "fps step: 9990 fps step and policy inference: 7984 fps total: 6768 epoch: 1411/1500\n",
+ "fps step: 11173 fps step and policy inference: 8803 fps total: 7349 epoch: 1412/1500\n",
+ "fps step: 9709 fps step and policy inference: 7656 fps total: 6533 epoch: 1413/1500\n",
+ "fps step: 8603 fps step and policy inference: 6777 fps total: 5882 epoch: 1414/1500\n",
+ "fps step: 9168 fps step and policy inference: 7195 fps total: 6175 epoch: 1415/1500\n",
+ "fps step: 9854 fps step and policy inference: 7848 fps total: 6670 epoch: 1416/1500\n",
+ "fps step: 8090 fps step and policy inference: 6347 fps total: 5554 epoch: 1417/1500\n",
+ "fps step: 10163 fps step and policy inference: 7965 fps total: 6758 epoch: 1418/1500\n",
+ "fps step: 10415 fps step and policy inference: 8180 fps total: 6913 epoch: 1419/1500\n",
+ "fps step: 10105 fps step and policy inference: 7940 fps total: 6739 epoch: 1420/1500\n",
+ "fps step: 10777 fps step and policy inference: 8619 fps total: 7224 epoch: 1421/1500\n",
+ "fps step: 8949 fps step and policy inference: 7006 fps total: 6049 epoch: 1422/1500\n",
+ "fps step: 10163 fps step and policy inference: 7941 fps total: 6737 epoch: 1423/1500\n",
+ "fps step: 9110 fps step and policy inference: 7186 fps total: 6186 epoch: 1424/1500\n",
+ "fps step: 9823 fps step and policy inference: 7730 fps total: 6587 epoch: 1425/1500\n",
+ "fps step: 9785 fps step and policy inference: 7683 fps total: 6551 epoch: 1426/1500\n",
+ "fps step: 9999 fps step and policy inference: 7878 fps total: 6695 epoch: 1427/1500\n",
+ "fps step: 9790 fps step and policy inference: 7803 fps total: 6637 epoch: 1428/1500\n",
+ "fps step: 9986 fps step and policy inference: 7872 fps total: 6688 epoch: 1429/1500\n",
+ "fps step: 8969 fps step and policy inference: 7094 fps total: 6119 epoch: 1430/1500\n",
+ "fps step: 10210 fps step and policy inference: 8200 fps total: 6914 epoch: 1431/1500\n",
+ "fps step: 9213 fps step and policy inference: 7078 fps total: 6098 epoch: 1432/1500\n",
+ "fps step: 9545 fps step and policy inference: 7473 fps total: 6386 epoch: 1433/1500\n",
+ "fps step: 9519 fps step and policy inference: 7605 fps total: 6503 epoch: 1434/1500\n",
+ "fps step: 10029 fps step and policy inference: 8005 fps total: 6783 epoch: 1435/1500\n",
+ "fps step: 9224 fps step and policy inference: 7298 fps total: 6269 epoch: 1436/1500\n",
+ "fps step: 10343 fps step and policy inference: 8174 fps total: 6885 epoch: 1437/1500\n",
+ "fps step: 10041 fps step and policy inference: 7863 fps total: 6681 epoch: 1438/1500\n",
+ "fps step: 10178 fps step and policy inference: 8061 fps total: 6827 epoch: 1439/1500\n",
+ "fps step: 8837 fps step and policy inference: 6918 fps total: 5972 epoch: 1440/1500\n",
+ "fps step: 9305 fps step and policy inference: 7332 fps total: 6292 epoch: 1441/1500\n",
+ "fps step: 9481 fps step and policy inference: 7335 fps total: 6301 epoch: 1442/1500\n",
+ "fps step: 10414 fps step and policy inference: 8296 fps total: 6991 epoch: 1443/1500\n",
+ "fps step: 9989 fps step and policy inference: 7980 fps total: 6767 epoch: 1444/1500\n",
+ "fps step: 10269 fps step and policy inference: 8093 fps total: 6847 epoch: 1445/1500\n",
+ "fps step: 10481 fps step and policy inference: 8365 fps total: 7042 epoch: 1446/1500\n",
+ "fps step: 9861 fps step and policy inference: 7822 fps total: 6649 epoch: 1447/1500\n",
+ "fps step: 10607 fps step and policy inference: 8350 fps total: 7024 epoch: 1448/1500\n",
+ "fps step: 10750 fps step and policy inference: 8371 fps total: 7043 epoch: 1449/1500\n",
+ "fps step: 10458 fps step and policy inference: 8259 fps total: 6967 epoch: 1450/1500\n",
+ "fps step: 11213 fps step and policy inference: 8849 fps total: 7377 epoch: 1451/1500\n",
+ "fps step: 9717 fps step and policy inference: 7562 fps total: 6468 epoch: 1452/1500\n",
+ "fps step: 11493 fps step and policy inference: 9068 fps total: 7532 epoch: 1453/1500\n",
+ "fps step: 10553 fps step and policy inference: 8381 fps total: 7044 epoch: 1454/1500\n",
+ "fps step: 10594 fps step and policy inference: 8317 fps total: 7019 epoch: 1455/1500\n",
+ "fps step: 10039 fps step and policy inference: 7889 fps total: 6700 epoch: 1456/1500\n",
+ "fps step: 11001 fps step and policy inference: 8772 fps total: 7327 epoch: 1457/1500\n",
+ "fps step: 9552 fps step and policy inference: 7344 fps total: 6302 epoch: 1458/1500\n",
+ "fps step: 10889 fps step and policy inference: 8478 fps total: 7116 epoch: 1459/1500\n",
+ "fps step: 10383 fps step and policy inference: 8324 fps total: 7014 epoch: 1460/1500\n",
+ "fps step: 9435 fps step and policy inference: 7418 fps total: 6357 epoch: 1461/1500\n",
+ "fps step: 10711 fps step and policy inference: 8472 fps total: 7119 epoch: 1462/1500\n",
+ "fps step: 9338 fps step and policy inference: 7231 fps total: 6197 epoch: 1463/1500\n",
+ "fps step: 9975 fps step and policy inference: 7819 fps total: 6630 epoch: 1464/1500\n",
+ "fps step: 10224 fps step and policy inference: 8180 fps total: 6916 epoch: 1465/1500\n",
+ "fps step: 9808 fps step and policy inference: 7723 fps total: 6581 epoch: 1466/1500\n",
+ "fps step: 10425 fps step and policy inference: 8263 fps total: 6969 epoch: 1467/1500\n",
+ "fps step: 10491 fps step and policy inference: 8360 fps total: 7037 epoch: 1468/1500\n",
+ "fps step: 10278 fps step and policy inference: 8093 fps total: 6845 epoch: 1469/1500\n",
+ "fps step: 10932 fps step and policy inference: 8673 fps total: 7261 epoch: 1470/1500\n",
+ "fps step: 9288 fps step and policy inference: 7351 fps total: 6314 epoch: 1471/1500\n",
+ "fps step: 10106 fps step and policy inference: 8029 fps total: 6805 epoch: 1472/1500\n",
+ "fps step: 10236 fps step and policy inference: 8015 fps total: 6796 epoch: 1473/1500\n",
+ "fps step: 10492 fps step and policy inference: 8223 fps total: 6940 epoch: 1474/1500\n",
+ "fps step: 9656 fps step and policy inference: 7590 fps total: 6484 epoch: 1475/1500\n",
+ "fps step: 10833 fps step and policy inference: 8513 fps total: 7140 epoch: 1476/1500\n",
+ "fps step: 9279 fps step and policy inference: 7231 fps total: 6227 epoch: 1477/1500\n",
+ "saving next best rewards: [101.35467]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 10157 fps step and policy inference: 7973 fps total: 6762 epoch: 1478/1500\n",
+ "fps step: 9341 fps step and policy inference: 7423 fps total: 6364 epoch: 1479/1500\n",
+ "fps step: 9767 fps step and policy inference: 7731 fps total: 6586 epoch: 1480/1500\n",
+ "fps step: 10088 fps step and policy inference: 7949 fps total: 6742 epoch: 1481/1500\n",
+ "fps step: 8824 fps step and policy inference: 7014 fps total: 6041 epoch: 1482/1500\n",
+ "fps step: 10540 fps step and policy inference: 8485 fps total: 7125 epoch: 1483/1500\n",
+ "fps step: 9838 fps step and policy inference: 7659 fps total: 6535 epoch: 1484/1500\n",
+ "fps step: 11216 fps step and policy inference: 8842 fps total: 7387 epoch: 1485/1500\n",
+ "fps step: 10265 fps step and policy inference: 8172 fps total: 6877 epoch: 1486/1500\n",
+ "fps step: 10528 fps step and policy inference: 8202 fps total: 6926 epoch: 1487/1500\n",
+ "fps step: 10432 fps step and policy inference: 8248 fps total: 6967 epoch: 1488/1500\n",
+ "fps step: 9970 fps step and policy inference: 7903 fps total: 6717 epoch: 1489/1500\n",
+ "fps step: 11418 fps step and policy inference: 9016 fps total: 7495 epoch: 1490/1500\n",
+ "fps step: 9970 fps step and policy inference: 7881 fps total: 6695 epoch: 1491/1500\n",
+ "fps step: 9412 fps step and policy inference: 7361 fps total: 6313 epoch: 1492/1500\n",
+ "fps step: 10358 fps step and policy inference: 8238 fps total: 6949 epoch: 1493/1500\n",
+ "fps step: 10282 fps step and policy inference: 8062 fps total: 6826 epoch: 1494/1500\n",
+ "fps step: 10922 fps step and policy inference: 8583 fps total: 7196 epoch: 1495/1500\n",
+ "fps step: 9082 fps step and policy inference: 7072 fps total: 6101 epoch: 1496/1500\n",
+ "fps step: 9564 fps step and policy inference: 7629 fps total: 6512 epoch: 1497/1500\n",
+ "fps step: 9908 fps step and policy inference: 7884 fps total: 6702 epoch: 1498/1500\n",
+ "saving next best rewards: [103.26936]\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/Breakout_envpool.pth'\n",
+ "fps step: 9820 fps step and policy inference: 7647 fps total: 6525 epoch: 1499/1500\n",
+ "fps step: 10937 fps step and policy inference: 8670 fps total: 7253 epoch: 1500/1500\n",
+ "=> saving checkpoint 'runs/Breakout_envpool_23-10-01-59/nn/last_Breakout_envpoolep1500rew[96.06702].pth'\n",
+ "MAX EPOCHS NUM!\n"
+ ]
+ }
+ ]
+ }
+ ]
+}
\ No newline at end of file
diff --git a/demo/envpool_demo_halfcheetah.ipynb b/demo/envpool_demo_halfcheetah.ipynb
new file mode 100644
index 00000000..93ca1b1c
--- /dev/null
+++ b/demo/envpool_demo_halfcheetah.ipynb
@@ -0,0 +1,1375 @@
+{
+ "nbformat": 4,
+ "nbformat_minor": 0,
+ "metadata": {
+ "colab": {
+ "provenance": [],
+ "collapsed_sections": []
+ },
+ "kernelspec": {
+ "name": "python3",
+ "display_name": "Python 3"
+ },
+ "language_info": {
+ "name": "python"
+ }
+ },
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "source": [
+ "[](https://colab.research.google.com/github/sail-sg/envpool/blob/main/demo/envpool-demo-halfcheetah.ipynb)"
+ ],
+ "metadata": {
+ "id": "sWsMLda2UqXs"
+ }
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "id": "E1e0nhtA2Cix"
+ },
+ "outputs": [],
+ "source": [
+ "# OS: Ubuntu 18.04.4 LTS x86_64\n",
+ "# Kernel: 4.18.0-15-generic\n",
+ "# CPU: Intel(R) Core(TM) i9-10920X CPU (24) @ 3.50GHz\n",
+ "# GPU: NVIDIA GeForce RTX 2080 Ti\n",
+ "!apt-get install -y \\\n",
+ " libgl1-mesa-dev \\\n",
+ " libgl1-mesa-glx \\\n",
+ " libglew-dev \\\n",
+ " libosmesa6-dev \\\n",
+ " software-properties-common\n",
+ "\n",
+ "!apt-get install -y patchelf\n",
+ "!pip install git+https://github.com/Denys88/rl_games\n",
+ "!pip install envpool\n",
+ "!pip install gym\n",
+ "\n",
+ "!pip install free-mujoco-py\n",
+ "!apt-get install -y xvfb python-opengl ffmpeg > /dev/null 2>&1\n",
+ "!pip install imageio==2.4.1\n",
+ "!pip install -U colabgymrender"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "from rl_games.torch_runner import Runner"
+ ],
+ "metadata": {
+ "id": "Yx5DAo1t3WmH"
+ },
+ "execution_count": null,
+ "outputs": []
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "## breakout_envpool config:\n",
+ "halfcheetah_config = {'params': {'seed': 5, \n",
+ " 'algo': {'name': 'a2c_continuous'}, \n",
+ " 'model': {'name': 'continuous_a2c_logstd'}, \n",
+ " 'network': {'name': 'actor_critic', 'separate': False, \n",
+ " 'space': {'continuous': {'mu_activation': 'None', \n",
+ " 'sigma_activation': 'None', \n",
+ " 'mu_init': {'name': 'default'}, \n",
+ " 'sigma_init': {'name': 'const_initializer', 'val': 0}, \n",
+ " 'fixed_sigma': True}}, \n",
+ " 'mlp': {'units': [128, 64, 32], 'activation': 'elu', 'initializer': {'name': 'variance_scaling_initializer', 'scale': 2.0}}}, \n",
+ " 'config': {'name': 'HalfCheetah-v4_envpool', \n",
+ " 'env_name': 'envpool', \n",
+ " 'score_to_win': 20000, \n",
+ " 'normalize_input': True, \n",
+ " 'normalize_value': True, \n",
+ " 'value_bootstrap': True, \n",
+ " 'reward_shaper': {'scale_value': 0.1}, \n",
+ " 'normalize_advantage': True, \n",
+ " 'use_smooth_clamp': True, \n",
+ " 'gamma': 0.99, \n",
+ " 'tau': 0.95, \n",
+ " 'learning_rate': '5e-4', \n",
+ " 'lr_schedule': 'adaptive', \n",
+ " 'kl_threshold': 0.008, \n",
+ " 'grad_norm': 1.0, \n",
+ " 'entropy_coef': 0.0, \n",
+ " 'truncate_grads': True, \n",
+ " 'e_clip': 0.2, \n",
+ " 'clip_value': False, \n",
+ " 'num_actors': 64, \n",
+ " 'horizon_length': 256, \n",
+ " 'minibatch_size': 2048, \n",
+ " 'mini_epochs': 5, \n",
+ " 'critic_coef': 4, \n",
+ " 'bounds_loss_coef': 0.0, \n",
+ " 'max_epochs': 1000, \n",
+ " 'env_config': {'env_name': 'HalfCheetah-v4', 'seed': 5}, \n",
+ " 'player': {'render': True, 'deterministic': True, 'games_num': 100}}}}"
+ ],
+ "metadata": {
+ "id": "MS2z_yYs3a7b"
+ },
+ "execution_count": null,
+ "outputs": []
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "runner = Runner()\n",
+ "runner.load(halfcheetah_config)\n",
+ "runner.run({\n",
+ " 'train': True,\n",
+ "})"
+ ],
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "id": "qrFeIFHo4IY9",
+ "outputId": "87f182e6-3c2c-4e09-c317-733f1147a457"
+ },
+ "execution_count": null,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "self.seed = 5\n",
+ "Started to train\n",
+ "current training device: cuda:0\n",
+ "build mlp: 17\n",
+ "17 2.0\n",
+ "128 2.0\n",
+ "64 2.0\n",
+ "32 2.0\n",
+ "32 2.0\n",
+ "RunningMeanStd: (1,)\n",
+ "RunningMeanStd: (17,)\n",
+ "fps step: 62489 fps step and policy inference: 28838 fps total: 18177 epoch: 1/1000\n",
+ "fps step: 59257 fps step and policy inference: 25729 fps total: 17079 epoch: 2/1000\n",
+ "fps step: 60795 fps step and policy inference: 27544 fps total: 18056 epoch: 3/1000\n",
+ "fps step: 58677 fps step and policy inference: 26044 fps total: 17284 epoch: 4/1000\n",
+ "fps step: 63526 fps step and policy inference: 28307 fps total: 18434 epoch: 5/1000\n",
+ "fps step: 59637 fps step and policy inference: 25673 fps total: 17049 epoch: 6/1000\n",
+ "fps step: 62897 fps step and policy inference: 27898 fps total: 17977 epoch: 7/1000\n",
+ "fps step: 58517 fps step and policy inference: 25473 fps total: 16640 epoch: 8/1000\n",
+ "fps step: 60929 fps step and policy inference: 26612 fps total: 17576 epoch: 9/1000\n",
+ "fps step: 63709 fps step and policy inference: 29117 fps total: 18629 epoch: 10/1000\n",
+ "fps step: 62857 fps step and policy inference: 28868 fps total: 18389 epoch: 11/1000\n",
+ "fps step: 61236 fps step and policy inference: 27666 fps total: 17975 epoch: 12/1000\n",
+ "fps step: 57221 fps step and policy inference: 24114 fps total: 16467 epoch: 13/1000\n",
+ "fps step: 58832 fps step and policy inference: 25503 fps total: 16985 epoch: 14/1000\n",
+ "fps step: 55099 fps step and policy inference: 23083 fps total: 15914 epoch: 15/1000\n",
+ "fps step: 61074 fps step and policy inference: 27910 fps total: 18081 epoch: 16/1000\n",
+ "fps step: 63668 fps step and policy inference: 29079 fps total: 18638 epoch: 17/1000\n",
+ "fps step: 60069 fps step and policy inference: 25792 fps total: 17030 epoch: 18/1000\n",
+ "fps step: 57573 fps step and policy inference: 24878 fps total: 16948 epoch: 19/1000\n",
+ "fps step: 56115 fps step and policy inference: 23933 fps total: 16315 epoch: 20/1000\n",
+ "fps step: 57745 fps step and policy inference: 24578 fps total: 16783 epoch: 21/1000\n",
+ "fps step: 53966 fps step and policy inference: 22445 fps total: 15664 epoch: 22/1000\n",
+ "fps step: 61524 fps step and policy inference: 27537 fps total: 17900 epoch: 23/1000\n",
+ "fps step: 58779 fps step and policy inference: 25934 fps total: 17381 epoch: 24/1000\n",
+ "fps step: 52672 fps step and policy inference: 21218 fps total: 15020 epoch: 25/1000\n",
+ "fps step: 59777 fps step and policy inference: 26382 fps total: 17494 epoch: 26/1000\n",
+ "fps step: 62714 fps step and policy inference: 28913 fps total: 18568 epoch: 27/1000\n",
+ "fps step: 57198 fps step and policy inference: 24978 fps total: 16413 epoch: 28/1000\n",
+ "fps step: 61468 fps step and policy inference: 27164 fps total: 17860 epoch: 29/1000\n",
+ "fps step: 58239 fps step and policy inference: 25519 fps total: 16988 epoch: 30/1000\n",
+ "fps step: 62896 fps step and policy inference: 28207 fps total: 18397 epoch: 31/1000\n",
+ "fps step: 59044 fps step and policy inference: 26482 fps total: 17332 epoch: 32/1000\n",
+ "fps step: 61429 fps step and policy inference: 27303 fps total: 17842 epoch: 33/1000\n",
+ "fps step: 63161 fps step and policy inference: 28998 fps total: 18845 epoch: 34/1000\n",
+ "fps step: 63017 fps step and policy inference: 28755 fps total: 18589 epoch: 35/1000\n",
+ "fps step: 62251 fps step and policy inference: 28705 fps total: 18404 epoch: 36/1000\n",
+ "fps step: 62700 fps step and policy inference: 28795 fps total: 18492 epoch: 37/1000\n",
+ "fps step: 62210 fps step and policy inference: 27264 fps total: 17575 epoch: 38/1000\n",
+ "fps step: 61301 fps step and policy inference: 27138 fps total: 17849 epoch: 39/1000\n",
+ "fps step: 56082 fps step and policy inference: 24102 fps total: 16441 epoch: 40/1000\n",
+ "fps step: 60447 fps step and policy inference: 25617 fps total: 17025 epoch: 41/1000\n",
+ "fps step: 62412 fps step and policy inference: 28181 fps total: 18306 epoch: 42/1000\n",
+ "fps step: 63187 fps step and policy inference: 28796 fps total: 18530 epoch: 43/1000\n",
+ "fps step: 54260 fps step and policy inference: 23093 fps total: 15820 epoch: 44/1000\n",
+ "fps step: 60733 fps step and policy inference: 27127 fps total: 17810 epoch: 45/1000\n",
+ "fps step: 62970 fps step and policy inference: 28388 fps total: 18371 epoch: 46/1000\n",
+ "fps step: 59272 fps step and policy inference: 26318 fps total: 17545 epoch: 47/1000\n",
+ "fps step: 59943 fps step and policy inference: 26412 fps total: 17492 epoch: 48/1000\n",
+ "fps step: 63175 fps step and policy inference: 28592 fps total: 18418 epoch: 49/1000\n",
+ "fps step: 59465 fps step and policy inference: 25713 fps total: 17272 epoch: 50/1000\n",
+ "fps step: 61699 fps step and policy inference: 27925 fps total: 17755 epoch: 51/1000\n",
+ "fps step: 58724 fps step and policy inference: 25306 fps total: 17061 epoch: 52/1000\n",
+ "fps step: 64858 fps step and policy inference: 29352 fps total: 18826 epoch: 53/1000\n",
+ "fps step: 62531 fps step and policy inference: 28374 fps total: 18407 epoch: 54/1000\n",
+ "fps step: 57367 fps step and policy inference: 25014 fps total: 16783 epoch: 55/1000\n",
+ "fps step: 63817 fps step and policy inference: 29242 fps total: 18656 epoch: 56/1000\n",
+ "fps step: 60278 fps step and policy inference: 26452 fps total: 17480 epoch: 57/1000\n",
+ "fps step: 59292 fps step and policy inference: 25667 fps total: 16972 epoch: 58/1000\n",
+ "fps step: 60863 fps step and policy inference: 27519 fps total: 17867 epoch: 59/1000\n",
+ "fps step: 56723 fps step and policy inference: 23950 fps total: 16163 epoch: 60/1000\n",
+ "fps step: 62854 fps step and policy inference: 28723 fps total: 17996 epoch: 61/1000\n",
+ "fps step: 59979 fps step and policy inference: 25690 fps total: 16983 epoch: 62/1000\n",
+ "fps step: 57546 fps step and policy inference: 24947 fps total: 16949 epoch: 63/1000\n",
+ "fps step: 62993 fps step and policy inference: 28805 fps total: 18482 epoch: 64/1000\n",
+ "fps step: 63619 fps step and policy inference: 29099 fps total: 18780 epoch: 65/1000\n",
+ "fps step: 63280 fps step and policy inference: 28868 fps total: 18518 epoch: 66/1000\n",
+ "fps step: 60695 fps step and policy inference: 27255 fps total: 17742 epoch: 67/1000\n",
+ "fps step: 59488 fps step and policy inference: 26040 fps total: 17204 epoch: 68/1000\n",
+ "fps step: 60187 fps step and policy inference: 26535 fps total: 17579 epoch: 69/1000\n",
+ "fps step: 61765 fps step and policy inference: 27640 fps total: 18121 epoch: 70/1000\n",
+ "fps step: 60686 fps step and policy inference: 27727 fps total: 18083 epoch: 71/1000\n",
+ "fps step: 58671 fps step and policy inference: 25200 fps total: 16908 epoch: 72/1000\n",
+ "fps step: 63463 fps step and policy inference: 29134 fps total: 18733 epoch: 73/1000\n",
+ "fps step: 62365 fps step and policy inference: 27860 fps total: 18087 epoch: 74/1000\n",
+ "fps step: 61111 fps step and policy inference: 27739 fps total: 17827 epoch: 75/1000\n",
+ "fps step: 59068 fps step and policy inference: 25653 fps total: 17196 epoch: 76/1000\n",
+ "fps step: 61347 fps step and policy inference: 27244 fps total: 18006 epoch: 77/1000\n",
+ "fps step: 62337 fps step and policy inference: 28172 fps total: 18044 epoch: 78/1000\n",
+ "fps step: 59452 fps step and policy inference: 26281 fps total: 17512 epoch: 79/1000\n",
+ "fps step: 56739 fps step and policy inference: 23746 fps total: 16087 epoch: 80/1000\n",
+ "fps step: 63251 fps step and policy inference: 28940 fps total: 18718 epoch: 81/1000\n",
+ "fps step: 59842 fps step and policy inference: 26611 fps total: 17657 epoch: 82/1000\n",
+ "fps step: 58949 fps step and policy inference: 26520 fps total: 17081 epoch: 83/1000\n",
+ "fps step: 63041 fps step and policy inference: 28887 fps total: 18348 epoch: 84/1000\n",
+ "fps step: 60835 fps step and policy inference: 27249 fps total: 17750 epoch: 85/1000\n",
+ "fps step: 60378 fps step and policy inference: 26812 fps total: 17567 epoch: 86/1000\n",
+ "fps step: 60610 fps step and policy inference: 27698 fps total: 18131 epoch: 87/1000\n",
+ "fps step: 61544 fps step and policy inference: 27280 fps total: 17831 epoch: 88/1000\n",
+ "fps step: 63088 fps step and policy inference: 29082 fps total: 18664 epoch: 89/1000\n",
+ "fps step: 61092 fps step and policy inference: 27005 fps total: 17373 epoch: 90/1000\n",
+ "fps step: 62580 fps step and policy inference: 28682 fps total: 18358 epoch: 91/1000\n",
+ "fps step: 58603 fps step and policy inference: 25218 fps total: 16843 epoch: 92/1000\n",
+ "fps step: 62884 fps step and policy inference: 28326 fps total: 18169 epoch: 93/1000\n",
+ "fps step: 57822 fps step and policy inference: 25110 fps total: 16955 epoch: 94/1000\n",
+ "fps step: 62838 fps step and policy inference: 28925 fps total: 18683 epoch: 95/1000\n",
+ "fps step: 62404 fps step and policy inference: 27836 fps total: 18142 epoch: 96/1000\n",
+ "fps step: 58138 fps step and policy inference: 25055 fps total: 16945 epoch: 97/1000\n",
+ "fps step: 61548 fps step and policy inference: 28088 fps total: 18224 epoch: 98/1000\n",
+ "fps step: 59236 fps step and policy inference: 25742 fps total: 17194 epoch: 99/1000\n",
+ "fps step: 56503 fps step and policy inference: 23412 fps total: 16142 epoch: 100/1000\n",
+ "saving next best rewards: [818.1738]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 55977 fps step and policy inference: 23308 fps total: 16112 epoch: 101/1000\n",
+ "fps step: 57966 fps step and policy inference: 24683 fps total: 16552 epoch: 102/1000\n",
+ "saving next best rewards: [854.3523]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 58796 fps step and policy inference: 25647 fps total: 17072 epoch: 103/1000\n",
+ "fps step: 60225 fps step and policy inference: 26564 fps total: 17628 epoch: 104/1000\n",
+ "fps step: 59508 fps step and policy inference: 26049 fps total: 17276 epoch: 105/1000\n",
+ "fps step: 59092 fps step and policy inference: 26138 fps total: 17406 epoch: 106/1000\n",
+ "fps step: 61510 fps step and policy inference: 27648 fps total: 17889 epoch: 107/1000\n",
+ "fps step: 61161 fps step and policy inference: 27304 fps total: 17956 epoch: 108/1000\n",
+ "fps step: 58138 fps step and policy inference: 24792 fps total: 16786 epoch: 109/1000\n",
+ "fps step: 54678 fps step and policy inference: 22304 fps total: 15245 epoch: 110/1000\n",
+ "fps step: 59510 fps step and policy inference: 26058 fps total: 17111 epoch: 111/1000\n",
+ "fps step: 57409 fps step and policy inference: 24470 fps total: 16425 epoch: 112/1000\n",
+ "fps step: 59472 fps step and policy inference: 25751 fps total: 17012 epoch: 113/1000\n",
+ "fps step: 62189 fps step and policy inference: 28613 fps total: 18491 epoch: 114/1000\n",
+ "saving next best rewards: [870.3095]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 62183 fps step and policy inference: 27982 fps total: 17981 epoch: 115/1000\n",
+ "fps step: 61433 fps step and policy inference: 27499 fps total: 17935 epoch: 116/1000\n",
+ "fps step: 58587 fps step and policy inference: 25146 fps total: 16842 epoch: 117/1000\n",
+ "fps step: 60632 fps step and policy inference: 27442 fps total: 17844 epoch: 118/1000\n",
+ "saving next best rewards: [918.2434]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 58136 fps step and policy inference: 24578 fps total: 16536 epoch: 119/1000\n",
+ "fps step: 62610 fps step and policy inference: 28009 fps total: 18148 epoch: 120/1000\n",
+ "fps step: 63344 fps step and policy inference: 29001 fps total: 18502 epoch: 121/1000\n",
+ "fps step: 61870 fps step and policy inference: 28372 fps total: 18252 epoch: 122/1000\n",
+ "saving next best rewards: [937.3634]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 54965 fps step and policy inference: 22899 fps total: 15810 epoch: 123/1000\n",
+ "fps step: 60728 fps step and policy inference: 26725 fps total: 17678 epoch: 124/1000\n",
+ "fps step: 63091 fps step and policy inference: 28688 fps total: 18479 epoch: 125/1000\n",
+ "fps step: 59160 fps step and policy inference: 26359 fps total: 17601 epoch: 126/1000\n",
+ "saving next best rewards: [955.939]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 55060 fps step and policy inference: 22733 fps total: 15774 epoch: 127/1000\n",
+ "fps step: 61413 fps step and policy inference: 27554 fps total: 17989 epoch: 128/1000\n",
+ "fps step: 58584 fps step and policy inference: 25364 fps total: 17144 epoch: 129/1000\n",
+ "fps step: 51518 fps step and policy inference: 20816 fps total: 14764 epoch: 130/1000\n",
+ "saving next best rewards: [966.7492]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 60184 fps step and policy inference: 26641 fps total: 17556 epoch: 131/1000\n",
+ "fps step: 62209 fps step and policy inference: 27211 fps total: 17760 epoch: 132/1000\n",
+ "fps step: 52783 fps step and policy inference: 21497 fps total: 15082 epoch: 133/1000\n",
+ "saving next best rewards: [979.53186]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 56714 fps step and policy inference: 24234 fps total: 16508 epoch: 134/1000\n",
+ "fps step: 63343 fps step and policy inference: 29084 fps total: 18303 epoch: 135/1000\n",
+ "fps step: 59452 fps step and policy inference: 25928 fps total: 17301 epoch: 136/1000\n",
+ "fps step: 58352 fps step and policy inference: 25362 fps total: 16745 epoch: 137/1000\n",
+ "saving next best rewards: [992.0665]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 60725 fps step and policy inference: 27126 fps total: 17775 epoch: 138/1000\n",
+ "fps step: 63116 fps step and policy inference: 28799 fps total: 18523 epoch: 139/1000\n",
+ "fps step: 63430 fps step and policy inference: 28814 fps total: 17929 epoch: 140/1000\n",
+ "fps step: 60021 fps step and policy inference: 26799 fps total: 17671 epoch: 141/1000\n",
+ "saving next best rewards: [1004.3942]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 58098 fps step and policy inference: 24623 fps total: 16788 epoch: 142/1000\n",
+ "fps step: 62993 fps step and policy inference: 28547 fps total: 18448 epoch: 143/1000\n",
+ "fps step: 63490 fps step and policy inference: 29112 fps total: 18648 epoch: 144/1000\n",
+ "fps step: 62382 fps step and policy inference: 28568 fps total: 18513 epoch: 145/1000\n",
+ "saving next best rewards: [1014.3695]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 59623 fps step and policy inference: 25775 fps total: 17114 epoch: 146/1000\n",
+ "fps step: 63340 fps step and policy inference: 28704 fps total: 18478 epoch: 147/1000\n",
+ "fps step: 59438 fps step and policy inference: 25644 fps total: 17110 epoch: 148/1000\n",
+ "fps step: 58213 fps step and policy inference: 25355 fps total: 17023 epoch: 149/1000\n",
+ "saving next best rewards: [1018.8403]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 55967 fps step and policy inference: 22887 fps total: 15900 epoch: 150/1000\n",
+ "fps step: 60166 fps step and policy inference: 26158 fps total: 17115 epoch: 151/1000\n",
+ "fps step: 63266 fps step and policy inference: 28796 fps total: 18210 epoch: 152/1000\n",
+ "fps step: 59524 fps step and policy inference: 26499 fps total: 17495 epoch: 153/1000\n",
+ "saving next best rewards: [1021.1794]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 61424 fps step and policy inference: 27457 fps total: 17864 epoch: 154/1000\n",
+ "fps step: 63307 fps step and policy inference: 28658 fps total: 18524 epoch: 155/1000\n",
+ "fps step: 64071 fps step and policy inference: 29170 fps total: 18449 epoch: 156/1000\n",
+ "fps step: 53743 fps step and policy inference: 22470 fps total: 15766 epoch: 157/1000\n",
+ "saving next best rewards: [1031.0359]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 61759 fps step and policy inference: 27685 fps total: 18167 epoch: 158/1000\n",
+ "fps step: 57292 fps step and policy inference: 23898 fps total: 16332 epoch: 159/1000\n",
+ "fps step: 62145 fps step and policy inference: 27699 fps total: 18048 epoch: 160/1000\n",
+ "fps step: 59662 fps step and policy inference: 26677 fps total: 17645 epoch: 161/1000\n",
+ "saving next best rewards: [1036.5492]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 63847 fps step and policy inference: 28908 fps total: 18564 epoch: 162/1000\n",
+ "fps step: 62794 fps step and policy inference: 28387 fps total: 18298 epoch: 163/1000\n",
+ "fps step: 57818 fps step and policy inference: 24093 fps total: 16215 epoch: 164/1000\n",
+ "fps step: 61826 fps step and policy inference: 28602 fps total: 18307 epoch: 165/1000\n",
+ "saving next best rewards: [1044.2512]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 53912 fps step and policy inference: 22274 fps total: 15653 epoch: 166/1000\n",
+ "fps step: 60681 fps step and policy inference: 26400 fps total: 17509 epoch: 167/1000\n",
+ "fps step: 61480 fps step and policy inference: 27276 fps total: 17862 epoch: 168/1000\n",
+ "fps step: 61681 fps step and policy inference: 28315 fps total: 18214 epoch: 169/1000\n",
+ "saving next best rewards: [1045.2772]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 61809 fps step and policy inference: 27350 fps total: 17755 epoch: 170/1000\n",
+ "fps step: 60529 fps step and policy inference: 26745 fps total: 17480 epoch: 171/1000\n",
+ "fps step: 58244 fps step and policy inference: 24935 fps total: 16855 epoch: 172/1000\n",
+ "fps step: 59116 fps step and policy inference: 26386 fps total: 17524 epoch: 173/1000\n",
+ "saving next best rewards: [1055.2021]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 58656 fps step and policy inference: 25185 fps total: 16747 epoch: 174/1000\n",
+ "fps step: 63285 fps step and policy inference: 28757 fps total: 18469 epoch: 175/1000\n",
+ "fps step: 60012 fps step and policy inference: 26713 fps total: 17587 epoch: 176/1000\n",
+ "saving next best rewards: [1064.368]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 52495 fps step and policy inference: 21303 fps total: 15147 epoch: 177/1000\n",
+ "fps step: 57237 fps step and policy inference: 24322 fps total: 16470 epoch: 178/1000\n",
+ "fps step: 59387 fps step and policy inference: 26095 fps total: 17463 epoch: 179/1000\n",
+ "fps step: 55453 fps step and policy inference: 23355 fps total: 15778 epoch: 180/1000\n",
+ "saving next best rewards: [1079.767]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 63381 fps step and policy inference: 29169 fps total: 18674 epoch: 181/1000\n",
+ "fps step: 62007 fps step and policy inference: 27923 fps total: 18057 epoch: 182/1000\n",
+ "fps step: 63320 fps step and policy inference: 28486 fps total: 17819 epoch: 183/1000\n",
+ "fps step: 54846 fps step and policy inference: 23235 fps total: 16086 epoch: 184/1000\n",
+ "saving next best rewards: [1082.8523]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 60588 fps step and policy inference: 26242 fps total: 17424 epoch: 185/1000\n",
+ "fps step: 62491 fps step and policy inference: 27826 fps total: 18014 epoch: 186/1000\n",
+ "fps step: 61947 fps step and policy inference: 27844 fps total: 18145 epoch: 187/1000\n",
+ "fps step: 56150 fps step and policy inference: 23668 fps total: 16109 epoch: 188/1000\n",
+ "saving next best rewards: [1088.9779]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 60702 fps step and policy inference: 26620 fps total: 17593 epoch: 189/1000\n",
+ "fps step: 62702 fps step and policy inference: 28239 fps total: 18331 epoch: 190/1000\n",
+ "fps step: 57254 fps step and policy inference: 23870 fps total: 16310 epoch: 191/1000\n",
+ "fps step: 58445 fps step and policy inference: 25717 fps total: 17220 epoch: 192/1000\n",
+ "saving next best rewards: [1089.729]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 60018 fps step and policy inference: 26193 fps total: 17355 epoch: 193/1000\n",
+ "fps step: 60899 fps step and policy inference: 26656 fps total: 17582 epoch: 194/1000\n",
+ "fps step: 63322 fps step and policy inference: 29001 fps total: 18603 epoch: 195/1000\n",
+ "fps step: 59671 fps step and policy inference: 26670 fps total: 17552 epoch: 196/1000\n",
+ "saving next best rewards: [1097.6427]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 58770 fps step and policy inference: 25414 fps total: 17109 epoch: 197/1000\n",
+ "fps step: 63941 fps step and policy inference: 29231 fps total: 18511 epoch: 198/1000\n",
+ "fps step: 64049 fps step and policy inference: 29096 fps total: 18569 epoch: 199/1000\n",
+ "fps step: 57649 fps step and policy inference: 24789 fps total: 16575 epoch: 200/1000\n",
+ "saving next best rewards: [1109.6387]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 61038 fps step and policy inference: 27217 fps total: 17889 epoch: 201/1000\n",
+ "fps step: 58032 fps step and policy inference: 24637 fps total: 16666 epoch: 202/1000\n",
+ "fps step: 62128 fps step and policy inference: 27806 fps total: 18146 epoch: 203/1000\n",
+ "fps step: 60136 fps step and policy inference: 26360 fps total: 17321 epoch: 204/1000\n",
+ "saving next best rewards: [1115.0956]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 61262 fps step and policy inference: 27108 fps total: 17925 epoch: 205/1000\n",
+ "fps step: 60909 fps step and policy inference: 26745 fps total: 17664 epoch: 206/1000\n",
+ "fps step: 63554 fps step and policy inference: 29099 fps total: 18578 epoch: 207/1000\n",
+ "fps step: 60651 fps step and policy inference: 26687 fps total: 17523 epoch: 208/1000\n",
+ "fps step: 61632 fps step and policy inference: 27116 fps total: 17880 epoch: 209/1000\n",
+ "fps step: 63597 fps step and policy inference: 29071 fps total: 18725 epoch: 210/1000\n",
+ "fps step: 58683 fps step and policy inference: 25137 fps total: 16889 epoch: 211/1000\n",
+ "fps step: 56336 fps step and policy inference: 23755 fps total: 16185 epoch: 212/1000\n",
+ "fps step: 63272 fps step and policy inference: 28807 fps total: 18539 epoch: 213/1000\n",
+ "fps step: 56177 fps step and policy inference: 23539 fps total: 16185 epoch: 214/1000\n",
+ "fps step: 62573 fps step and policy inference: 28339 fps total: 18443 epoch: 215/1000\n",
+ "fps step: 58686 fps step and policy inference: 26035 fps total: 17020 epoch: 216/1000\n",
+ "saving next best rewards: [1117.6224]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 60725 fps step and policy inference: 26845 fps total: 17643 epoch: 217/1000\n",
+ "fps step: 57516 fps step and policy inference: 24437 fps total: 16690 epoch: 218/1000\n",
+ "fps step: 62228 fps step and policy inference: 28445 fps total: 18209 epoch: 219/1000\n",
+ "fps step: 61373 fps step and policy inference: 27334 fps total: 17934 epoch: 220/1000\n",
+ "fps step: 60654 fps step and policy inference: 27032 fps total: 17752 epoch: 221/1000\n",
+ "fps step: 60400 fps step and policy inference: 26190 fps total: 17470 epoch: 222/1000\n",
+ "fps step: 59608 fps step and policy inference: 26097 fps total: 17148 epoch: 223/1000\n",
+ "saving next best rewards: [1124.0813]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 57027 fps step and policy inference: 23687 fps total: 16088 epoch: 224/1000\n",
+ "fps step: 57619 fps step and policy inference: 24306 fps total: 16539 epoch: 225/1000\n",
+ "fps step: 63007 fps step and policy inference: 28694 fps total: 18484 epoch: 226/1000\n",
+ "fps step: 61284 fps step and policy inference: 27926 fps total: 18001 epoch: 227/1000\n",
+ "fps step: 57291 fps step and policy inference: 24168 fps total: 16510 epoch: 228/1000\n",
+ "fps step: 61256 fps step and policy inference: 26897 fps total: 17726 epoch: 229/1000\n",
+ "fps step: 63996 fps step and policy inference: 29199 fps total: 18730 epoch: 230/1000\n",
+ "fps step: 56697 fps step and policy inference: 24222 fps total: 16512 epoch: 231/1000\n",
+ "fps step: 59111 fps step and policy inference: 25209 fps total: 16910 epoch: 232/1000\n",
+ "fps step: 64035 fps step and policy inference: 29163 fps total: 18757 epoch: 233/1000\n",
+ "fps step: 64258 fps step and policy inference: 29323 fps total: 18830 epoch: 234/1000\n",
+ "fps step: 60071 fps step and policy inference: 26160 fps total: 17275 epoch: 235/1000\n",
+ "saving next best rewards: [1130.4342]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 63187 fps step and policy inference: 28685 fps total: 18483 epoch: 236/1000\n",
+ "fps step: 63720 fps step and policy inference: 29157 fps total: 18567 epoch: 237/1000\n",
+ "fps step: 64498 fps step and policy inference: 29313 fps total: 18751 epoch: 238/1000\n",
+ "fps step: 59118 fps step and policy inference: 25956 fps total: 17245 epoch: 239/1000\n",
+ "saving next best rewards: [1137.3552]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 63212 fps step and policy inference: 28423 fps total: 18353 epoch: 240/1000\n",
+ "fps step: 63883 fps step and policy inference: 29064 fps total: 18705 epoch: 241/1000\n",
+ "fps step: 62502 fps step and policy inference: 27885 fps total: 18149 epoch: 242/1000\n",
+ "fps step: 61104 fps step and policy inference: 27336 fps total: 17981 epoch: 243/1000\n",
+ "saving next best rewards: [1137.7528]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 56261 fps step and policy inference: 23781 fps total: 16281 epoch: 244/1000\n",
+ "fps step: 58167 fps step and policy inference: 24650 fps total: 16352 epoch: 245/1000\n",
+ "fps step: 61624 fps step and policy inference: 27250 fps total: 17930 epoch: 246/1000\n",
+ "fps step: 62123 fps step and policy inference: 28746 fps total: 18554 epoch: 247/1000\n",
+ "fps step: 62070 fps step and policy inference: 27384 fps total: 18024 epoch: 248/1000\n",
+ "fps step: 63197 fps step and policy inference: 28269 fps total: 18265 epoch: 249/1000\n",
+ "fps step: 59432 fps step and policy inference: 25177 fps total: 16953 epoch: 250/1000\n",
+ "fps step: 62073 fps step and policy inference: 28336 fps total: 18316 epoch: 251/1000\n",
+ "saving next best rewards: [1143.3489]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 58832 fps step and policy inference: 25478 fps total: 17068 epoch: 252/1000\n",
+ "fps step: 63878 fps step and policy inference: 29061 fps total: 18816 epoch: 253/1000\n",
+ "fps step: 64047 fps step and policy inference: 29297 fps total: 18872 epoch: 254/1000\n",
+ "fps step: 60815 fps step and policy inference: 27235 fps total: 17837 epoch: 255/1000\n",
+ "saving next best rewards: [1149.7886]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 62451 fps step and policy inference: 27115 fps total: 17892 epoch: 256/1000\n",
+ "fps step: 63899 fps step and policy inference: 29187 fps total: 18743 epoch: 257/1000\n",
+ "fps step: 56585 fps step and policy inference: 23267 fps total: 16044 epoch: 258/1000\n",
+ "fps step: 61139 fps step and policy inference: 27192 fps total: 17565 epoch: 259/1000\n",
+ "saving next best rewards: [1157.3303]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 54157 fps step and policy inference: 22416 fps total: 15586 epoch: 260/1000\n",
+ "fps step: 61792 fps step and policy inference: 27005 fps total: 17796 epoch: 261/1000\n",
+ "fps step: 58866 fps step and policy inference: 25580 fps total: 17148 epoch: 262/1000\n",
+ "fps step: 64016 fps step and policy inference: 29257 fps total: 18738 epoch: 263/1000\n",
+ "fps step: 62813 fps step and policy inference: 28483 fps total: 18380 epoch: 264/1000\n",
+ "fps step: 63618 fps step and policy inference: 28759 fps total: 18269 epoch: 265/1000\n",
+ "fps step: 61337 fps step and policy inference: 27135 fps total: 17945 epoch: 266/1000\n",
+ "saving next best rewards: [1160.3453]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 57231 fps step and policy inference: 24086 fps total: 16513 epoch: 267/1000\n",
+ "fps step: 62961 fps step and policy inference: 28303 fps total: 18374 epoch: 268/1000\n",
+ "fps step: 64282 fps step and policy inference: 29331 fps total: 18869 epoch: 269/1000\n",
+ "fps step: 59730 fps step and policy inference: 26390 fps total: 17527 epoch: 270/1000\n",
+ "fps step: 59302 fps step and policy inference: 25509 fps total: 16933 epoch: 271/1000\n",
+ "fps step: 62656 fps step and policy inference: 27934 fps total: 18195 epoch: 272/1000\n",
+ "fps step: 63277 fps step and policy inference: 28701 fps total: 18426 epoch: 273/1000\n",
+ "fps step: 58232 fps step and policy inference: 24897 fps total: 16755 epoch: 274/1000\n",
+ "saving next best rewards: [1162.6523]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 59110 fps step and policy inference: 25363 fps total: 16920 epoch: 275/1000\n",
+ "fps step: 64067 fps step and policy inference: 28988 fps total: 18629 epoch: 276/1000\n",
+ "fps step: 61367 fps step and policy inference: 26994 fps total: 17495 epoch: 277/1000\n",
+ "fps step: 57936 fps step and policy inference: 25385 fps total: 16943 epoch: 278/1000\n",
+ "fps step: 64025 fps step and policy inference: 28893 fps total: 18667 epoch: 279/1000\n",
+ "fps step: 64409 fps step and policy inference: 29293 fps total: 18799 epoch: 280/1000\n",
+ "fps step: 62562 fps step and policy inference: 27920 fps total: 17770 epoch: 281/1000\n",
+ "fps step: 53359 fps step and policy inference: 22069 fps total: 15446 epoch: 282/1000\n",
+ "saving next best rewards: [1164.9218]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 58106 fps step and policy inference: 24867 fps total: 16867 epoch: 283/1000\n",
+ "fps step: 64851 fps step and policy inference: 29332 fps total: 18721 epoch: 284/1000\n",
+ "fps step: 59382 fps step and policy inference: 25522 fps total: 17002 epoch: 285/1000\n",
+ "fps step: 53972 fps step and policy inference: 22373 fps total: 15686 epoch: 286/1000\n",
+ "saving next best rewards: [1172.8223]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 61222 fps step and policy inference: 26537 fps total: 17583 epoch: 287/1000\n",
+ "fps step: 59395 fps step and policy inference: 25745 fps total: 17061 epoch: 288/1000\n",
+ "fps step: 56450 fps step and policy inference: 23282 fps total: 16132 epoch: 289/1000\n",
+ "fps step: 60346 fps step and policy inference: 27116 fps total: 17859 epoch: 290/1000\n",
+ "saving next best rewards: [1174.0287]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 63443 fps step and policy inference: 28706 fps total: 18478 epoch: 291/1000\n",
+ "fps step: 59199 fps step and policy inference: 25645 fps total: 17173 epoch: 292/1000\n",
+ "fps step: 57056 fps step and policy inference: 23844 fps total: 16242 epoch: 293/1000\n",
+ "fps step: 62399 fps step and policy inference: 28775 fps total: 18367 epoch: 294/1000\n",
+ "saving next best rewards: [1180.4784]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 62428 fps step and policy inference: 28219 fps total: 18132 epoch: 295/1000\n",
+ "fps step: 61060 fps step and policy inference: 26285 fps total: 17218 epoch: 296/1000\n",
+ "fps step: 62570 fps step and policy inference: 28138 fps total: 18062 epoch: 297/1000\n",
+ "fps step: 57146 fps step and policy inference: 24483 fps total: 16452 epoch: 298/1000\n",
+ "saving next best rewards: [1184.8309]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 59406 fps step and policy inference: 25481 fps total: 17035 epoch: 299/1000\n",
+ "fps step: 63539 fps step and policy inference: 28570 fps total: 18094 epoch: 300/1000\n",
+ "fps step: 63589 fps step and policy inference: 28810 fps total: 18588 epoch: 301/1000\n",
+ "fps step: 52385 fps step and policy inference: 21419 fps total: 15208 epoch: 302/1000\n",
+ "saving next best rewards: [1196.1543]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 63451 fps step and policy inference: 28645 fps total: 18415 epoch: 303/1000\n",
+ "fps step: 60355 fps step and policy inference: 26325 fps total: 17503 epoch: 304/1000\n",
+ "fps step: 61408 fps step and policy inference: 26984 fps total: 17700 epoch: 305/1000\n",
+ "saving next best rewards: [1201.6825]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 62779 fps step and policy inference: 28466 fps total: 18248 epoch: 306/1000\n",
+ "fps step: 61646 fps step and policy inference: 27128 fps total: 17780 epoch: 307/1000\n",
+ "fps step: 61342 fps step and policy inference: 27032 fps total: 17751 epoch: 308/1000\n",
+ "fps step: 61445 fps step and policy inference: 27625 fps total: 17987 epoch: 309/1000\n",
+ "saving next best rewards: [1204.7086]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 56338 fps step and policy inference: 23749 fps total: 16278 epoch: 310/1000\n",
+ "fps step: 64389 fps step and policy inference: 29254 fps total: 18712 epoch: 311/1000\n",
+ "fps step: 59849 fps step and policy inference: 25990 fps total: 17169 epoch: 312/1000\n",
+ "fps step: 61382 fps step and policy inference: 28071 fps total: 18176 epoch: 313/1000\n",
+ "saving next best rewards: [1207.7092]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 64034 fps step and policy inference: 29315 fps total: 18767 epoch: 314/1000\n",
+ "fps step: 61822 fps step and policy inference: 26715 fps total: 17501 epoch: 315/1000\n",
+ "fps step: 63845 fps step and policy inference: 29042 fps total: 18599 epoch: 316/1000\n",
+ "fps step: 60355 fps step and policy inference: 26972 fps total: 17279 epoch: 317/1000\n",
+ "saving next best rewards: [1210.9998]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 55137 fps step and policy inference: 22540 fps total: 15598 epoch: 318/1000\n",
+ "fps step: 62445 fps step and policy inference: 27587 fps total: 17973 epoch: 319/1000\n",
+ "fps step: 64104 fps step and policy inference: 29262 fps total: 18698 epoch: 320/1000\n",
+ "fps step: 60364 fps step and policy inference: 26735 fps total: 17629 epoch: 321/1000\n",
+ "fps step: 59335 fps step and policy inference: 25394 fps total: 16927 epoch: 322/1000\n",
+ "fps step: 59815 fps step and policy inference: 25773 fps total: 17063 epoch: 323/1000\n",
+ "fps step: 58422 fps step and policy inference: 24653 fps total: 16837 epoch: 324/1000\n",
+ "fps step: 60764 fps step and policy inference: 27381 fps total: 17940 epoch: 325/1000\n",
+ "saving next best rewards: [1215.7773]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 61592 fps step and policy inference: 27201 fps total: 17857 epoch: 326/1000\n",
+ "fps step: 59832 fps step and policy inference: 25960 fps total: 17239 epoch: 327/1000\n",
+ "fps step: 55772 fps step and policy inference: 23227 fps total: 16063 epoch: 328/1000\n",
+ "fps step: 58474 fps step and policy inference: 25738 fps total: 17241 epoch: 329/1000\n",
+ "saving next best rewards: [1217.6003]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 57184 fps step and policy inference: 24211 fps total: 16308 epoch: 330/1000\n",
+ "fps step: 58795 fps step and policy inference: 25166 fps total: 16923 epoch: 331/1000\n",
+ "fps step: 62108 fps step and policy inference: 27531 fps total: 17897 epoch: 332/1000\n",
+ "fps step: 61529 fps step and policy inference: 28041 fps total: 18206 epoch: 333/1000\n",
+ "saving next best rewards: [1223.5044]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 62009 fps step and policy inference: 27146 fps total: 17833 epoch: 334/1000\n",
+ "fps step: 64133 fps step and policy inference: 29233 fps total: 18252 epoch: 335/1000\n",
+ "fps step: 61345 fps step and policy inference: 26742 fps total: 17692 epoch: 336/1000\n",
+ "fps step: 57977 fps step and policy inference: 24615 fps total: 16749 epoch: 337/1000\n",
+ "saving next best rewards: [1227.4424]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 59286 fps step and policy inference: 25991 fps total: 17052 epoch: 338/1000\n",
+ "fps step: 59283 fps step and policy inference: 25324 fps total: 17125 epoch: 339/1000\n",
+ "fps step: 62111 fps step and policy inference: 27240 fps total: 17526 epoch: 340/1000\n",
+ "fps step: 59684 fps step and policy inference: 26502 fps total: 17476 epoch: 341/1000\n",
+ "saving next best rewards: [1231.1469]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 60334 fps step and policy inference: 26344 fps total: 17587 epoch: 342/1000\n",
+ "fps step: 64175 fps step and policy inference: 29215 fps total: 18404 epoch: 343/1000\n",
+ "fps step: 62995 fps step and policy inference: 28092 fps total: 18195 epoch: 344/1000\n",
+ "fps step: 54933 fps step and policy inference: 23115 fps total: 16014 epoch: 345/1000\n",
+ "saving next best rewards: [1234.4147]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 63558 fps step and policy inference: 28652 fps total: 18277 epoch: 346/1000\n",
+ "fps step: 56428 fps step and policy inference: 23439 fps total: 16081 epoch: 347/1000\n",
+ "fps step: 63663 fps step and policy inference: 28493 fps total: 18444 epoch: 348/1000\n",
+ "fps step: 56483 fps step and policy inference: 24396 fps total: 16591 epoch: 349/1000\n",
+ "fps step: 63196 fps step and policy inference: 28191 fps total: 18350 epoch: 350/1000\n",
+ "fps step: 64097 fps step and policy inference: 29270 fps total: 18778 epoch: 351/1000\n",
+ "fps step: 59494 fps step and policy inference: 25389 fps total: 16942 epoch: 352/1000\n",
+ "saving next best rewards: [1236.9548]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 59189 fps step and policy inference: 25580 fps total: 17182 epoch: 353/1000\n",
+ "fps step: 62819 fps step and policy inference: 27056 fps total: 17385 epoch: 354/1000\n",
+ "fps step: 63210 fps step and policy inference: 28272 fps total: 18326 epoch: 355/1000\n",
+ "fps step: 59075 fps step and policy inference: 25842 fps total: 17183 epoch: 356/1000\n",
+ "saving next best rewards: [1238.1478]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 64531 fps step and policy inference: 29314 fps total: 18755 epoch: 357/1000\n",
+ "fps step: 63732 fps step and policy inference: 28607 fps total: 18304 epoch: 358/1000\n",
+ "fps step: 62327 fps step and policy inference: 27566 fps total: 18104 epoch: 359/1000\n",
+ "fps step: 61949 fps step and policy inference: 27936 fps total: 18090 epoch: 360/1000\n",
+ "saving next best rewards: [1248.426]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 64254 fps step and policy inference: 29384 fps total: 18759 epoch: 361/1000\n",
+ "fps step: 61956 fps step and policy inference: 27287 fps total: 17877 epoch: 362/1000\n",
+ "fps step: 63478 fps step and policy inference: 28371 fps total: 18212 epoch: 363/1000\n",
+ "fps step: 56509 fps step and policy inference: 24072 fps total: 16367 epoch: 364/1000\n",
+ "saving next best rewards: [1251.962]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 63911 fps step and policy inference: 28946 fps total: 18647 epoch: 365/1000\n",
+ "fps step: 61149 fps step and policy inference: 26554 fps total: 17464 epoch: 366/1000\n",
+ "fps step: 63568 fps step and policy inference: 28464 fps total: 18453 epoch: 367/1000\n",
+ "fps step: 60921 fps step and policy inference: 27292 fps total: 17939 epoch: 368/1000\n",
+ "saving next best rewards: [1252.8398]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 58503 fps step and policy inference: 24972 fps total: 16864 epoch: 369/1000\n",
+ "fps step: 62531 fps step and policy inference: 27682 fps total: 17976 epoch: 370/1000\n",
+ "fps step: 55777 fps step and policy inference: 23014 fps total: 16044 epoch: 371/1000\n",
+ "fps step: 57071 fps step and policy inference: 24325 fps total: 16614 epoch: 372/1000\n",
+ "saving next best rewards: [1253.3486]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 55737 fps step and policy inference: 22748 fps total: 15934 epoch: 373/1000\n",
+ "fps step: 59492 fps step and policy inference: 25497 fps total: 17148 epoch: 374/1000\n",
+ "fps step: 62714 fps step and policy inference: 27836 fps total: 18038 epoch: 375/1000\n",
+ "fps step: 61360 fps step and policy inference: 27563 fps total: 18097 epoch: 376/1000\n",
+ "saving next best rewards: [1259.0176]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 64248 fps step and policy inference: 28943 fps total: 18400 epoch: 377/1000\n",
+ "fps step: 63138 fps step and policy inference: 27692 fps total: 17791 epoch: 378/1000\n",
+ "fps step: 62216 fps step and policy inference: 27512 fps total: 18044 epoch: 379/1000\n",
+ "fps step: 60661 fps step and policy inference: 27162 fps total: 17897 epoch: 380/1000\n",
+ "saving next best rewards: [1260.0424]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 63198 fps step and policy inference: 28152 fps total: 17863 epoch: 381/1000\n",
+ "fps step: 56380 fps step and policy inference: 23586 fps total: 16411 epoch: 382/1000\n",
+ "fps step: 59104 fps step and policy inference: 24738 fps total: 16798 epoch: 383/1000\n",
+ "fps step: 60809 fps step and policy inference: 27392 fps total: 17979 epoch: 384/1000\n",
+ "fps step: 65002 fps step and policy inference: 29453 fps total: 18876 epoch: 385/1000\n",
+ "fps step: 54039 fps step and policy inference: 21837 fps total: 15381 epoch: 386/1000\n",
+ "fps step: 63805 fps step and policy inference: 28536 fps total: 18384 epoch: 387/1000\n",
+ "fps step: 61994 fps step and policy inference: 28644 fps total: 18407 epoch: 388/1000\n",
+ "fps step: 61852 fps step and policy inference: 26945 fps total: 17553 epoch: 389/1000\n",
+ "fps step: 61277 fps step and policy inference: 26748 fps total: 17713 epoch: 390/1000\n",
+ "fps step: 62279 fps step and policy inference: 27544 fps total: 17840 epoch: 391/1000\n",
+ "fps step: 59057 fps step and policy inference: 25705 fps total: 16941 epoch: 392/1000\n",
+ "fps step: 60737 fps step and policy inference: 26298 fps total: 17526 epoch: 393/1000\n",
+ "fps step: 63529 fps step and policy inference: 28750 fps total: 18509 epoch: 394/1000\n",
+ "fps step: 61691 fps step and policy inference: 27073 fps total: 17896 epoch: 395/1000\n",
+ "saving next best rewards: [1265.2601]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 63241 fps step and policy inference: 28679 fps total: 18243 epoch: 396/1000\n",
+ "fps step: 60918 fps step and policy inference: 26651 fps total: 17738 epoch: 397/1000\n",
+ "fps step: 64495 fps step and policy inference: 29347 fps total: 18543 epoch: 398/1000\n",
+ "fps step: 60749 fps step and policy inference: 26633 fps total: 17424 epoch: 399/1000\n",
+ "saving next best rewards: [1276.3889]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 59523 fps step and policy inference: 25423 fps total: 16892 epoch: 400/1000\n",
+ "fps step: 62212 fps step and policy inference: 27468 fps total: 17945 epoch: 401/1000\n",
+ "fps step: 64040 fps step and policy inference: 28583 fps total: 18517 epoch: 402/1000\n",
+ "fps step: 61809 fps step and policy inference: 28009 fps total: 18081 epoch: 403/1000\n",
+ "saving next best rewards: [1281.0923]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 60562 fps step and policy inference: 26078 fps total: 17308 epoch: 404/1000\n",
+ "fps step: 63884 fps step and policy inference: 28616 fps total: 18718 epoch: 405/1000\n",
+ "fps step: 63472 fps step and policy inference: 28564 fps total: 18467 epoch: 406/1000\n",
+ "fps step: 58573 fps step and policy inference: 25507 fps total: 17052 epoch: 407/1000\n",
+ "saving next best rewards: [1282.646]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 63892 fps step and policy inference: 28262 fps total: 18249 epoch: 408/1000\n",
+ "fps step: 64376 fps step and policy inference: 28991 fps total: 18589 epoch: 409/1000\n",
+ "fps step: 61011 fps step and policy inference: 26096 fps total: 17220 epoch: 410/1000\n",
+ "fps step: 59430 fps step and policy inference: 26010 fps total: 17141 epoch: 411/1000\n",
+ "saving next best rewards: [1286.6921]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 54222 fps step and policy inference: 21933 fps total: 15357 epoch: 412/1000\n",
+ "fps step: 63262 fps step and policy inference: 28120 fps total: 18322 epoch: 413/1000\n",
+ "fps step: 57760 fps step and policy inference: 24498 fps total: 16640 epoch: 414/1000\n",
+ "fps step: 58362 fps step and policy inference: 25124 fps total: 17006 epoch: 415/1000\n",
+ "saving next best rewards: [1288.418]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 64772 fps step and policy inference: 29352 fps total: 18802 epoch: 416/1000\n",
+ "fps step: 61402 fps step and policy inference: 26723 fps total: 17664 epoch: 417/1000\n",
+ "fps step: 64656 fps step and policy inference: 29362 fps total: 18748 epoch: 418/1000\n",
+ "fps step: 59562 fps step and policy inference: 26291 fps total: 17484 epoch: 419/1000\n",
+ "saving next best rewards: [1293.5834]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 64542 fps step and policy inference: 29109 fps total: 18670 epoch: 420/1000\n",
+ "fps step: 59579 fps step and policy inference: 25265 fps total: 17037 epoch: 421/1000\n",
+ "fps step: 62767 fps step and policy inference: 27801 fps total: 17956 epoch: 422/1000\n",
+ "fps step: 62791 fps step and policy inference: 28866 fps total: 18645 epoch: 423/1000\n",
+ "fps step: 62293 fps step and policy inference: 27369 fps total: 17976 epoch: 424/1000\n",
+ "fps step: 62304 fps step and policy inference: 27197 fps total: 17794 epoch: 425/1000\n",
+ "fps step: 64087 fps step and policy inference: 28848 fps total: 18353 epoch: 426/1000\n",
+ "fps step: 62300 fps step and policy inference: 28258 fps total: 18326 epoch: 427/1000\n",
+ "fps step: 63152 fps step and policy inference: 27881 fps total: 18224 epoch: 428/1000\n",
+ "fps step: 64429 fps step and policy inference: 29254 fps total: 18639 epoch: 429/1000\n",
+ "fps step: 64348 fps step and policy inference: 29187 fps total: 18399 epoch: 430/1000\n",
+ "fps step: 61494 fps step and policy inference: 27639 fps total: 18098 epoch: 431/1000\n",
+ "fps step: 65164 fps step and policy inference: 29389 fps total: 18731 epoch: 432/1000\n",
+ "fps step: 62329 fps step and policy inference: 27434 fps total: 17992 epoch: 433/1000\n",
+ "fps step: 61393 fps step and policy inference: 26868 fps total: 18101 epoch: 434/1000\n",
+ "fps step: 64386 fps step and policy inference: 29459 fps total: 18865 epoch: 435/1000\n",
+ "fps step: 63183 fps step and policy inference: 28118 fps total: 18286 epoch: 436/1000\n",
+ "fps step: 65129 fps step and policy inference: 29364 fps total: 18689 epoch: 437/1000\n",
+ "fps step: 59747 fps step and policy inference: 25920 fps total: 17314 epoch: 438/1000\n",
+ "saving next best rewards: [1298.6871]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 57127 fps step and policy inference: 23983 fps total: 16435 epoch: 439/1000\n",
+ "fps step: 63933 fps step and policy inference: 28543 fps total: 18509 epoch: 440/1000\n",
+ "fps step: 63179 fps step and policy inference: 27978 fps total: 18156 epoch: 441/1000\n",
+ "fps step: 61102 fps step and policy inference: 27257 fps total: 17922 epoch: 442/1000\n",
+ "saving next best rewards: [1300.9241]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 61971 fps step and policy inference: 27372 fps total: 17993 epoch: 443/1000\n",
+ "fps step: 63607 fps step and policy inference: 28503 fps total: 18351 epoch: 444/1000\n",
+ "fps step: 60126 fps step and policy inference: 25784 fps total: 17152 epoch: 445/1000\n",
+ "fps step: 60449 fps step and policy inference: 26801 fps total: 17694 epoch: 446/1000\n",
+ "saving next best rewards: [1304.0745]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 65066 fps step and policy inference: 29376 fps total: 18805 epoch: 447/1000\n",
+ "fps step: 61442 fps step and policy inference: 26110 fps total: 17228 epoch: 448/1000\n",
+ "fps step: 57930 fps step and policy inference: 24525 fps total: 16726 epoch: 449/1000\n",
+ "fps step: 62530 fps step and policy inference: 28470 fps total: 18433 epoch: 450/1000\n",
+ "saving next best rewards: [1307.29]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 64421 fps step and policy inference: 29405 fps total: 18629 epoch: 451/1000\n",
+ "fps step: 60282 fps step and policy inference: 25864 fps total: 17101 epoch: 452/1000\n",
+ "fps step: 59823 fps step and policy inference: 25848 fps total: 17006 epoch: 453/1000\n",
+ "fps step: 58763 fps step and policy inference: 25913 fps total: 17334 epoch: 454/1000\n",
+ "saving next best rewards: [1313.0443]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 58773 fps step and policy inference: 24575 fps total: 16680 epoch: 455/1000\n",
+ "fps step: 64315 fps step and policy inference: 29117 fps total: 18755 epoch: 456/1000\n",
+ "fps step: 64261 fps step and policy inference: 29260 fps total: 18659 epoch: 457/1000\n",
+ "fps step: 56827 fps step and policy inference: 24129 fps total: 16404 epoch: 458/1000\n",
+ "saving next best rewards: [1314.1877]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 62575 fps step and policy inference: 27883 fps total: 17974 epoch: 459/1000\n",
+ "fps step: 63822 fps step and policy inference: 28063 fps total: 18158 epoch: 460/1000\n",
+ "fps step: 55705 fps step and policy inference: 22776 fps total: 15694 epoch: 461/1000\n",
+ "fps step: 57542 fps step and policy inference: 24791 fps total: 16736 epoch: 462/1000\n",
+ "fps step: 64792 fps step and policy inference: 29194 fps total: 18352 epoch: 463/1000\n",
+ "fps step: 63058 fps step and policy inference: 28016 fps total: 18151 epoch: 464/1000\n",
+ "fps step: 64502 fps step and policy inference: 29363 fps total: 18640 epoch: 465/1000\n",
+ "fps step: 62307 fps step and policy inference: 28165 fps total: 18209 epoch: 466/1000\n",
+ "saving next best rewards: [1319.8751]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 58391 fps step and policy inference: 25089 fps total: 16733 epoch: 467/1000\n",
+ "fps step: 61267 fps step and policy inference: 26286 fps total: 17453 epoch: 468/1000\n",
+ "fps step: 64473 fps step and policy inference: 29260 fps total: 18781 epoch: 469/1000\n",
+ "fps step: 62085 fps step and policy inference: 28557 fps total: 18349 epoch: 470/1000\n",
+ "saving next best rewards: [1323.2725]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 64408 fps step and policy inference: 29162 fps total: 18785 epoch: 471/1000\n",
+ "fps step: 64582 fps step and policy inference: 29305 fps total: 18794 epoch: 472/1000\n",
+ "fps step: 63739 fps step and policy inference: 28464 fps total: 18245 epoch: 473/1000\n",
+ "fps step: 62670 fps step and policy inference: 28708 fps total: 18468 epoch: 474/1000\n",
+ "fps step: 64741 fps step and policy inference: 29388 fps total: 18783 epoch: 475/1000\n",
+ "fps step: 63896 fps step and policy inference: 28858 fps total: 18572 epoch: 476/1000\n",
+ "fps step: 63306 fps step and policy inference: 28251 fps total: 18017 epoch: 477/1000\n",
+ "fps step: 61894 fps step and policy inference: 28032 fps total: 18146 epoch: 478/1000\n",
+ "fps step: 63768 fps step and policy inference: 28632 fps total: 18453 epoch: 479/1000\n",
+ "fps step: 64544 fps step and policy inference: 29180 fps total: 18594 epoch: 480/1000\n",
+ "fps step: 61799 fps step and policy inference: 27732 fps total: 18158 epoch: 481/1000\n",
+ "fps step: 64330 fps step and policy inference: 29355 fps total: 18782 epoch: 482/1000\n",
+ "fps step: 64853 fps step and policy inference: 29342 fps total: 18756 epoch: 483/1000\n",
+ "fps step: 61074 fps step and policy inference: 26348 fps total: 17556 epoch: 484/1000\n",
+ "fps step: 61380 fps step and policy inference: 27279 fps total: 17824 epoch: 485/1000\n",
+ "saving next best rewards: [1324.4282]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 58174 fps step and policy inference: 24670 fps total: 16822 epoch: 486/1000\n",
+ "fps step: 63431 fps step and policy inference: 28296 fps total: 18363 epoch: 487/1000\n",
+ "fps step: 62792 fps step and policy inference: 27741 fps total: 18070 epoch: 488/1000\n",
+ "fps step: 59334 fps step and policy inference: 25839 fps total: 17254 epoch: 489/1000\n",
+ "fps step: 62062 fps step and policy inference: 27277 fps total: 17795 epoch: 490/1000\n",
+ "fps step: 63245 fps step and policy inference: 28089 fps total: 18318 epoch: 491/1000\n",
+ "fps step: 59990 fps step and policy inference: 25803 fps total: 17279 epoch: 492/1000\n",
+ "fps step: 56716 fps step and policy inference: 23884 fps total: 16258 epoch: 493/1000\n",
+ "fps step: 56751 fps step and policy inference: 23924 fps total: 16474 epoch: 494/1000\n",
+ "fps step: 60642 fps step and policy inference: 26325 fps total: 17391 epoch: 495/1000\n",
+ "fps step: 60537 fps step and policy inference: 26305 fps total: 17444 epoch: 496/1000\n",
+ "fps step: 61699 fps step and policy inference: 27953 fps total: 18293 epoch: 497/1000\n",
+ "saving next best rewards: [1327.4231]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 60508 fps step and policy inference: 26045 fps total: 17370 epoch: 498/1000\n",
+ "fps step: 64956 fps step and policy inference: 29330 fps total: 18333 epoch: 499/1000\n",
+ "fps step: 62938 fps step and policy inference: 28021 fps total: 18213 epoch: 500/1000\n",
+ "fps step: 57778 fps step and policy inference: 24370 fps total: 16361 epoch: 501/1000\n",
+ "saving next best rewards: [1333.0974]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 59584 fps step and policy inference: 24485 fps total: 16260 epoch: 502/1000\n",
+ "fps step: 61799 fps step and policy inference: 27031 fps total: 17665 epoch: 503/1000\n",
+ "fps step: 58676 fps step and policy inference: 24584 fps total: 16349 epoch: 504/1000\n",
+ "fps step: 57629 fps step and policy inference: 24659 fps total: 16767 epoch: 505/1000\n",
+ "fps step: 60617 fps step and policy inference: 26104 fps total: 17241 epoch: 506/1000\n",
+ "fps step: 57168 fps step and policy inference: 23902 fps total: 16777 epoch: 507/1000\n",
+ "fps step: 63485 fps step and policy inference: 28143 fps total: 18313 epoch: 508/1000\n",
+ "fps step: 60925 fps step and policy inference: 27090 fps total: 17686 epoch: 509/1000\n",
+ "saving next best rewards: [1337.2572]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 63260 fps step and policy inference: 28304 fps total: 18041 epoch: 510/1000\n",
+ "fps step: 60772 fps step and policy inference: 26313 fps total: 17413 epoch: 511/1000\n",
+ "fps step: 64293 fps step and policy inference: 29102 fps total: 18524 epoch: 512/1000\n",
+ "fps step: 58775 fps step and policy inference: 25718 fps total: 17212 epoch: 513/1000\n",
+ "saving next best rewards: [1337.2776]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 61159 fps step and policy inference: 26561 fps total: 17503 epoch: 514/1000\n",
+ "fps step: 54913 fps step and policy inference: 22451 fps total: 15771 epoch: 515/1000\n",
+ "fps step: 62039 fps step and policy inference: 27122 fps total: 17825 epoch: 516/1000\n",
+ "fps step: 62142 fps step and policy inference: 28240 fps total: 18361 epoch: 517/1000\n",
+ "saving next best rewards: [1340.7289]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 62469 fps step and policy inference: 26657 fps total: 17213 epoch: 518/1000\n",
+ "fps step: 61594 fps step and policy inference: 26728 fps total: 17681 epoch: 519/1000\n",
+ "fps step: 62627 fps step and policy inference: 27584 fps total: 17908 epoch: 520/1000\n",
+ "fps step: 59999 fps step and policy inference: 27334 fps total: 17773 epoch: 521/1000\n",
+ "fps step: 64815 fps step and policy inference: 29117 fps total: 18603 epoch: 522/1000\n",
+ "fps step: 64687 fps step and policy inference: 29346 fps total: 18779 epoch: 523/1000\n",
+ "fps step: 62694 fps step and policy inference: 28020 fps total: 18019 epoch: 524/1000\n",
+ "saving next best rewards: [1344.4573]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 62923 fps step and policy inference: 28441 fps total: 18214 epoch: 525/1000\n",
+ "fps step: 63125 fps step and policy inference: 27899 fps total: 18099 epoch: 526/1000\n",
+ "fps step: 59261 fps step and policy inference: 25169 fps total: 16936 epoch: 527/1000\n",
+ "fps step: 59076 fps step and policy inference: 25701 fps total: 17083 epoch: 528/1000\n",
+ "fps step: 62134 fps step and policy inference: 27260 fps total: 17986 epoch: 529/1000\n",
+ "fps step: 64425 fps step and policy inference: 29135 fps total: 18739 epoch: 530/1000\n",
+ "fps step: 64626 fps step and policy inference: 29321 fps total: 18753 epoch: 531/1000\n",
+ "fps step: 59394 fps step and policy inference: 26082 fps total: 17159 epoch: 532/1000\n",
+ "fps step: 55865 fps step and policy inference: 22776 fps total: 15919 epoch: 533/1000\n",
+ "fps step: 64345 fps step and policy inference: 29286 fps total: 18735 epoch: 534/1000\n",
+ "fps step: 63550 fps step and policy inference: 28393 fps total: 18384 epoch: 535/1000\n",
+ "fps step: 60922 fps step and policy inference: 27390 fps total: 17952 epoch: 536/1000\n",
+ "fps step: 62430 fps step and policy inference: 27299 fps total: 17663 epoch: 537/1000\n",
+ "fps step: 63434 fps step and policy inference: 28154 fps total: 18300 epoch: 538/1000\n",
+ "fps step: 63185 fps step and policy inference: 27848 fps total: 18127 epoch: 539/1000\n",
+ "fps step: 53495 fps step and policy inference: 22101 fps total: 15445 epoch: 540/1000\n",
+ "fps step: 65143 fps step and policy inference: 29432 fps total: 18703 epoch: 541/1000\n",
+ "fps step: 64698 fps step and policy inference: 29055 fps total: 18575 epoch: 542/1000\n",
+ "fps step: 62091 fps step and policy inference: 27436 fps total: 17968 epoch: 543/1000\n",
+ "fps step: 61928 fps step and policy inference: 28434 fps total: 18405 epoch: 544/1000\n",
+ "fps step: 64844 fps step and policy inference: 28876 fps total: 18496 epoch: 545/1000\n",
+ "fps step: 65351 fps step and policy inference: 29634 fps total: 18695 epoch: 546/1000\n",
+ "fps step: 63792 fps step and policy inference: 28375 fps total: 18365 epoch: 547/1000\n",
+ "fps step: 55845 fps step and policy inference: 24157 fps total: 16442 epoch: 548/1000\n",
+ "fps step: 63959 fps step and policy inference: 28509 fps total: 18599 epoch: 549/1000\n",
+ "fps step: 64577 fps step and policy inference: 28737 fps total: 18529 epoch: 550/1000\n",
+ "fps step: 59867 fps step and policy inference: 25482 fps total: 16890 epoch: 551/1000\n",
+ "fps step: 56425 fps step and policy inference: 24126 fps total: 16475 epoch: 552/1000\n",
+ "fps step: 62834 fps step and policy inference: 27954 fps total: 18153 epoch: 553/1000\n",
+ "fps step: 63191 fps step and policy inference: 27836 fps total: 18137 epoch: 554/1000\n",
+ "fps step: 65204 fps step and policy inference: 29334 fps total: 18794 epoch: 555/1000\n",
+ "fps step: 61663 fps step and policy inference: 28205 fps total: 18167 epoch: 556/1000\n",
+ "fps step: 63854 fps step and policy inference: 28445 fps total: 18038 epoch: 557/1000\n",
+ "fps step: 61837 fps step and policy inference: 26383 fps total: 17335 epoch: 558/1000\n",
+ "fps step: 61831 fps step and policy inference: 27145 fps total: 17570 epoch: 559/1000\n",
+ "fps step: 59321 fps step and policy inference: 26148 fps total: 17370 epoch: 560/1000\n",
+ "fps step: 61711 fps step and policy inference: 26632 fps total: 17543 epoch: 561/1000\n",
+ "fps step: 63681 fps step and policy inference: 28123 fps total: 18237 epoch: 562/1000\n",
+ "fps step: 62340 fps step and policy inference: 27595 fps total: 17869 epoch: 563/1000\n",
+ "fps step: 58926 fps step and policy inference: 25145 fps total: 16958 epoch: 564/1000\n",
+ "fps step: 62122 fps step and policy inference: 27079 fps total: 17808 epoch: 565/1000\n",
+ "fps step: 63565 fps step and policy inference: 28182 fps total: 18280 epoch: 566/1000\n",
+ "fps step: 55535 fps step and policy inference: 22578 fps total: 15593 epoch: 567/1000\n",
+ "fps step: 60718 fps step and policy inference: 26475 fps total: 17524 epoch: 568/1000\n",
+ "fps step: 62195 fps step and policy inference: 26975 fps total: 17631 epoch: 569/1000\n",
+ "fps step: 65029 fps step and policy inference: 29271 fps total: 18730 epoch: 570/1000\n",
+ "fps step: 59928 fps step and policy inference: 25975 fps total: 17152 epoch: 571/1000\n",
+ "saving next best rewards: [1349.1943]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 62374 fps step and policy inference: 28032 fps total: 17980 epoch: 572/1000\n",
+ "fps step: 60965 fps step and policy inference: 26613 fps total: 17656 epoch: 573/1000\n",
+ "fps step: 61610 fps step and policy inference: 26946 fps total: 17766 epoch: 574/1000\n",
+ "fps step: 59577 fps step and policy inference: 26121 fps total: 17362 epoch: 575/1000\n",
+ "fps step: 62412 fps step and policy inference: 27705 fps total: 18206 epoch: 576/1000\n",
+ "fps step: 63333 fps step and policy inference: 28405 fps total: 18332 epoch: 577/1000\n",
+ "fps step: 55054 fps step and policy inference: 22357 fps total: 15605 epoch: 578/1000\n",
+ "fps step: 58540 fps step and policy inference: 25507 fps total: 17029 epoch: 579/1000\n",
+ "saving next best rewards: [1350.5806]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 60377 fps step and policy inference: 26051 fps total: 17327 epoch: 580/1000\n",
+ "fps step: 64378 fps step and policy inference: 28781 fps total: 18432 epoch: 581/1000\n",
+ "fps step: 64786 fps step and policy inference: 29351 fps total: 18614 epoch: 582/1000\n",
+ "fps step: 61395 fps step and policy inference: 27953 fps total: 18086 epoch: 583/1000\n",
+ "saving next best rewards: [1355.495]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 60380 fps step and policy inference: 25693 fps total: 17121 epoch: 584/1000\n",
+ "fps step: 62853 fps step and policy inference: 28112 fps total: 18011 epoch: 585/1000\n",
+ "fps step: 60409 fps step and policy inference: 25890 fps total: 17295 epoch: 586/1000\n",
+ "fps step: 58774 fps step and policy inference: 26081 fps total: 17313 epoch: 587/1000\n",
+ "fps step: 60482 fps step and policy inference: 26056 fps total: 17360 epoch: 588/1000\n",
+ "fps step: 62268 fps step and policy inference: 27207 fps total: 17900 epoch: 589/1000\n",
+ "fps step: 59464 fps step and policy inference: 25200 fps total: 17017 epoch: 590/1000\n",
+ "fps step: 56163 fps step and policy inference: 23900 fps total: 16277 epoch: 591/1000\n",
+ "fps step: 64955 fps step and policy inference: 29407 fps total: 18800 epoch: 592/1000\n",
+ "fps step: 54908 fps step and policy inference: 21910 fps total: 15353 epoch: 593/1000\n",
+ "fps step: 64980 fps step and policy inference: 29361 fps total: 18744 epoch: 594/1000\n",
+ "fps step: 60848 fps step and policy inference: 27571 fps total: 17970 epoch: 595/1000\n",
+ "fps step: 64297 fps step and policy inference: 28859 fps total: 18435 epoch: 596/1000\n",
+ "fps step: 64159 fps step and policy inference: 28793 fps total: 18490 epoch: 597/1000\n",
+ "fps step: 64309 fps step and policy inference: 29118 fps total: 18730 epoch: 598/1000\n",
+ "fps step: 58701 fps step and policy inference: 25477 fps total: 17030 epoch: 599/1000\n",
+ "fps step: 64971 fps step and policy inference: 29433 fps total: 18849 epoch: 600/1000\n",
+ "fps step: 64978 fps step and policy inference: 29240 fps total: 18607 epoch: 601/1000\n",
+ "fps step: 61151 fps step and policy inference: 26557 fps total: 17645 epoch: 602/1000\n",
+ "fps step: 59915 fps step and policy inference: 26934 fps total: 17677 epoch: 603/1000\n",
+ "fps step: 58196 fps step and policy inference: 24447 fps total: 16633 epoch: 604/1000\n",
+ "fps step: 61973 fps step and policy inference: 26974 fps total: 17588 epoch: 605/1000\n",
+ "fps step: 61235 fps step and policy inference: 26292 fps total: 17406 epoch: 606/1000\n",
+ "fps step: 60012 fps step and policy inference: 26970 fps total: 17725 epoch: 607/1000\n",
+ "fps step: 56853 fps step and policy inference: 23720 fps total: 16290 epoch: 608/1000\n",
+ "fps step: 65031 fps step and policy inference: 29435 fps total: 18377 epoch: 609/1000\n",
+ "fps step: 63468 fps step and policy inference: 27974 fps total: 17936 epoch: 610/1000\n",
+ "fps step: 62423 fps step and policy inference: 28285 fps total: 18264 epoch: 611/1000\n",
+ "fps step: 64983 fps step and policy inference: 29345 fps total: 18882 epoch: 612/1000\n",
+ "fps step: 63526 fps step and policy inference: 28382 fps total: 18412 epoch: 613/1000\n",
+ "fps step: 59973 fps step and policy inference: 26061 fps total: 17332 epoch: 614/1000\n",
+ "saving next best rewards: [1358.167]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 59749 fps step and policy inference: 25684 fps total: 17264 epoch: 615/1000\n",
+ "fps step: 63686 fps step and policy inference: 27605 fps total: 17984 epoch: 616/1000\n",
+ "fps step: 62340 fps step and policy inference: 27106 fps total: 17452 epoch: 617/1000\n",
+ "fps step: 58807 fps step and policy inference: 25627 fps total: 17076 epoch: 618/1000\n",
+ "fps step: 64613 fps step and policy inference: 29085 fps total: 18179 epoch: 619/1000\n",
+ "fps step: 64906 fps step and policy inference: 29345 fps total: 18859 epoch: 620/1000\n",
+ "fps step: 65045 fps step and policy inference: 29383 fps total: 18704 epoch: 621/1000\n",
+ "fps step: 57850 fps step and policy inference: 25281 fps total: 16976 epoch: 622/1000\n",
+ "saving next best rewards: [1361.5159]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 60914 fps step and policy inference: 26750 fps total: 17776 epoch: 623/1000\n",
+ "fps step: 64743 fps step and policy inference: 29021 fps total: 18284 epoch: 624/1000\n",
+ "fps step: 60889 fps step and policy inference: 26448 fps total: 17464 epoch: 625/1000\n",
+ "fps step: 61546 fps step and policy inference: 27818 fps total: 17957 epoch: 626/1000\n",
+ "fps step: 62530 fps step and policy inference: 26990 fps total: 17791 epoch: 627/1000\n",
+ "fps step: 65887 fps step and policy inference: 29702 fps total: 18841 epoch: 628/1000\n",
+ "fps step: 63142 fps step and policy inference: 27824 fps total: 17979 epoch: 629/1000\n",
+ "fps step: 57624 fps step and policy inference: 24977 fps total: 16943 epoch: 630/1000\n",
+ "fps step: 64623 fps step and policy inference: 29304 fps total: 18682 epoch: 631/1000\n",
+ "fps step: 64273 fps step and policy inference: 28595 fps total: 18409 epoch: 632/1000\n",
+ "fps step: 59627 fps step and policy inference: 25484 fps total: 17055 epoch: 633/1000\n",
+ "fps step: 59710 fps step and policy inference: 26232 fps total: 17318 epoch: 634/1000\n",
+ "fps step: 64718 fps step and policy inference: 29414 fps total: 18806 epoch: 635/1000\n",
+ "fps step: 64508 fps step and policy inference: 29285 fps total: 18780 epoch: 636/1000\n",
+ "fps step: 64314 fps step and policy inference: 28544 fps total: 18395 epoch: 637/1000\n",
+ "fps step: 59218 fps step and policy inference: 25922 fps total: 17262 epoch: 638/1000\n",
+ "saving next best rewards: [1365.7087]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 60581 fps step and policy inference: 26165 fps total: 17442 epoch: 639/1000\n",
+ "fps step: 65443 fps step and policy inference: 29473 fps total: 18656 epoch: 640/1000\n",
+ "fps step: 61782 fps step and policy inference: 26919 fps total: 17589 epoch: 641/1000\n",
+ "fps step: 55197 fps step and policy inference: 23003 fps total: 15980 epoch: 642/1000\n",
+ "fps step: 60397 fps step and policy inference: 25779 fps total: 17284 epoch: 643/1000\n",
+ "fps step: 64741 fps step and policy inference: 29235 fps total: 18802 epoch: 644/1000\n",
+ "fps step: 59791 fps step and policy inference: 25154 fps total: 16946 epoch: 645/1000\n",
+ "fps step: 61059 fps step and policy inference: 27882 fps total: 18217 epoch: 646/1000\n",
+ "fps step: 63908 fps step and policy inference: 28422 fps total: 18076 epoch: 647/1000\n",
+ "fps step: 64861 fps step and policy inference: 29335 fps total: 18641 epoch: 648/1000\n",
+ "fps step: 63530 fps step and policy inference: 28224 fps total: 17873 epoch: 649/1000\n",
+ "fps step: 52488 fps step and policy inference: 21800 fps total: 15372 epoch: 650/1000\n",
+ "fps step: 62219 fps step and policy inference: 27122 fps total: 17863 epoch: 651/1000\n",
+ "fps step: 57596 fps step and policy inference: 24142 fps total: 16213 epoch: 652/1000\n",
+ "fps step: 59926 fps step and policy inference: 25933 fps total: 17390 epoch: 653/1000\n",
+ "fps step: 61421 fps step and policy inference: 27718 fps total: 18057 epoch: 654/1000\n",
+ "fps step: 62676 fps step and policy inference: 27994 fps total: 18198 epoch: 655/1000\n",
+ "fps step: 65161 fps step and policy inference: 29352 fps total: 18695 epoch: 656/1000\n",
+ "fps step: 60651 fps step and policy inference: 26922 fps total: 17677 epoch: 657/1000\n",
+ "fps step: 60800 fps step and policy inference: 27051 fps total: 17676 epoch: 658/1000\n",
+ "fps step: 63664 fps step and policy inference: 28278 fps total: 18398 epoch: 659/1000\n",
+ "fps step: 62708 fps step and policy inference: 27410 fps total: 17909 epoch: 660/1000\n",
+ "fps step: 62163 fps step and policy inference: 28462 fps total: 18389 epoch: 661/1000\n",
+ "fps step: 62659 fps step and policy inference: 27352 fps total: 17929 epoch: 662/1000\n",
+ "fps step: 59424 fps step and policy inference: 25035 fps total: 16706 epoch: 663/1000\n",
+ "fps step: 64779 fps step and policy inference: 29286 fps total: 18788 epoch: 664/1000\n",
+ "fps step: 60673 fps step and policy inference: 27197 fps total: 17868 epoch: 665/1000\n",
+ "fps step: 63765 fps step and policy inference: 28516 fps total: 18437 epoch: 666/1000\n",
+ "fps step: 62249 fps step and policy inference: 27172 fps total: 17847 epoch: 667/1000\n",
+ "fps step: 64026 fps step and policy inference: 29161 fps total: 18673 epoch: 668/1000\n",
+ "fps step: 61238 fps step and policy inference: 27208 fps total: 17721 epoch: 669/1000\n",
+ "fps step: 55969 fps step and policy inference: 23017 fps total: 16050 epoch: 670/1000\n",
+ "fps step: 65209 fps step and policy inference: 29434 fps total: 18942 epoch: 671/1000\n",
+ "fps step: 60623 fps step and policy inference: 25987 fps total: 17303 epoch: 672/1000\n",
+ "fps step: 60156 fps step and policy inference: 27379 fps total: 17910 epoch: 673/1000\n",
+ "fps step: 64821 fps step and policy inference: 29340 fps total: 19026 epoch: 674/1000\n",
+ "fps step: 65174 fps step and policy inference: 29516 fps total: 18696 epoch: 675/1000\n",
+ "fps step: 61261 fps step and policy inference: 26601 fps total: 17675 epoch: 676/1000\n",
+ "fps step: 57354 fps step and policy inference: 25013 fps total: 16846 epoch: 677/1000\n",
+ "fps step: 64492 fps step and policy inference: 29261 fps total: 18728 epoch: 678/1000\n",
+ "fps step: 64827 fps step and policy inference: 29490 fps total: 18561 epoch: 679/1000\n",
+ "fps step: 64333 fps step and policy inference: 29285 fps total: 18638 epoch: 680/1000\n",
+ "fps step: 58684 fps step and policy inference: 25832 fps total: 17368 epoch: 681/1000\n",
+ "fps step: 64698 fps step and policy inference: 28924 fps total: 18526 epoch: 682/1000\n",
+ "fps step: 58979 fps step and policy inference: 24549 fps total: 16677 epoch: 683/1000\n",
+ "fps step: 65011 fps step and policy inference: 29389 fps total: 18705 epoch: 684/1000\n",
+ "fps step: 56996 fps step and policy inference: 24538 fps total: 16550 epoch: 685/1000\n",
+ "fps step: 63560 fps step and policy inference: 28211 fps total: 18069 epoch: 686/1000\n",
+ "fps step: 62088 fps step and policy inference: 27069 fps total: 17876 epoch: 687/1000\n",
+ "fps step: 54544 fps step and policy inference: 21818 fps total: 15244 epoch: 688/1000\n",
+ "fps step: 56320 fps step and policy inference: 24355 fps total: 16499 epoch: 689/1000\n",
+ "fps step: 63029 fps step and policy inference: 28110 fps total: 18037 epoch: 690/1000\n",
+ "fps step: 63746 fps step and policy inference: 28517 fps total: 18448 epoch: 691/1000\n",
+ "fps step: 65095 fps step and policy inference: 29029 fps total: 18609 epoch: 692/1000\n",
+ "fps step: 57262 fps step and policy inference: 25133 fps total: 17020 epoch: 693/1000\n",
+ "fps step: 65220 fps step and policy inference: 29490 fps total: 18877 epoch: 694/1000\n",
+ "fps step: 63604 fps step and policy inference: 28058 fps total: 18269 epoch: 695/1000\n",
+ "fps step: 57935 fps step and policy inference: 24212 fps total: 16509 epoch: 696/1000\n",
+ "fps step: 61432 fps step and policy inference: 28530 fps total: 18424 epoch: 697/1000\n",
+ "fps step: 57988 fps step and policy inference: 23716 fps total: 16287 epoch: 698/1000\n",
+ "fps step: 65021 fps step and policy inference: 29429 fps total: 18855 epoch: 699/1000\n",
+ "fps step: 63084 fps step and policy inference: 28875 fps total: 18625 epoch: 700/1000\n",
+ "fps step: 64304 fps step and policy inference: 29079 fps total: 18605 epoch: 701/1000\n",
+ "fps step: 61282 fps step and policy inference: 26517 fps total: 17652 epoch: 702/1000\n",
+ "fps step: 64949 fps step and policy inference: 29376 fps total: 18780 epoch: 703/1000\n",
+ "fps step: 60342 fps step and policy inference: 26893 fps total: 17811 epoch: 704/1000\n",
+ "saving next best rewards: [1367.465]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 58607 fps step and policy inference: 24855 fps total: 16827 epoch: 705/1000\n",
+ "fps step: 59376 fps step and policy inference: 25029 fps total: 16919 epoch: 706/1000\n",
+ "fps step: 61451 fps step and policy inference: 26481 fps total: 17532 epoch: 707/1000\n",
+ "fps step: 62659 fps step and policy inference: 28629 fps total: 18499 epoch: 708/1000\n",
+ "fps step: 61540 fps step and policy inference: 26633 fps total: 17325 epoch: 709/1000\n",
+ "fps step: 64161 fps step and policy inference: 28495 fps total: 18377 epoch: 710/1000\n",
+ "fps step: 63487 fps step and policy inference: 28199 fps total: 18185 epoch: 711/1000\n",
+ "fps step: 61774 fps step and policy inference: 28267 fps total: 18358 epoch: 712/1000\n",
+ "saving next best rewards: [1370.4733]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 59487 fps step and policy inference: 25634 fps total: 17280 epoch: 713/1000\n",
+ "fps step: 62549 fps step and policy inference: 27309 fps total: 17739 epoch: 714/1000\n",
+ "fps step: 63312 fps step and policy inference: 28123 fps total: 18219 epoch: 715/1000\n",
+ "fps step: 56991 fps step and policy inference: 24342 fps total: 16653 epoch: 716/1000\n",
+ "fps step: 65137 fps step and policy inference: 29506 fps total: 18893 epoch: 717/1000\n",
+ "fps step: 65284 fps step and policy inference: 29624 fps total: 18937 epoch: 718/1000\n",
+ "fps step: 64804 fps step and policy inference: 29434 fps total: 18665 epoch: 719/1000\n",
+ "fps step: 58450 fps step and policy inference: 25764 fps total: 17036 epoch: 720/1000\n",
+ "saving next best rewards: [1373.7653]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 62768 fps step and policy inference: 27397 fps total: 17913 epoch: 721/1000\n",
+ "fps step: 64691 fps step and policy inference: 29326 fps total: 18799 epoch: 722/1000\n",
+ "fps step: 60022 fps step and policy inference: 25624 fps total: 17116 epoch: 723/1000\n",
+ "fps step: 58832 fps step and policy inference: 25678 fps total: 17270 epoch: 724/1000\n",
+ "saving next best rewards: [1376.0829]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 63925 fps step and policy inference: 28446 fps total: 18446 epoch: 725/1000\n",
+ "fps step: 65018 fps step and policy inference: 29226 fps total: 18726 epoch: 726/1000\n",
+ "fps step: 64779 fps step and policy inference: 29151 fps total: 18579 epoch: 727/1000\n",
+ "fps step: 62436 fps step and policy inference: 28296 fps total: 18302 epoch: 728/1000\n",
+ "fps step: 62953 fps step and policy inference: 27888 fps total: 18131 epoch: 729/1000\n",
+ "fps step: 64556 fps step and policy inference: 29228 fps total: 18672 epoch: 730/1000\n",
+ "fps step: 65159 fps step and policy inference: 29405 fps total: 18817 epoch: 731/1000\n",
+ "fps step: 60009 fps step and policy inference: 26660 fps total: 17739 epoch: 732/1000\n",
+ "saving next best rewards: [1380.6776]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 60384 fps step and policy inference: 25853 fps total: 17280 epoch: 733/1000\n",
+ "fps step: 64415 fps step and policy inference: 28914 fps total: 18606 epoch: 734/1000\n",
+ "fps step: 64252 fps step and policy inference: 28990 fps total: 18612 epoch: 735/1000\n",
+ "fps step: 59520 fps step and policy inference: 26190 fps total: 17499 epoch: 736/1000\n",
+ "saving next best rewards: [1387.0175]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 63797 fps step and policy inference: 28424 fps total: 18473 epoch: 737/1000\n",
+ "fps step: 57971 fps step and policy inference: 24085 fps total: 16330 epoch: 738/1000\n",
+ "fps step: 61557 fps step and policy inference: 26236 fps total: 17471 epoch: 739/1000\n",
+ "fps step: 56160 fps step and policy inference: 24439 fps total: 16687 epoch: 740/1000\n",
+ "saving next best rewards: [1387.8055]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 52389 fps step and policy inference: 20566 fps total: 14575 epoch: 741/1000\n",
+ "fps step: 58802 fps step and policy inference: 24658 fps total: 16759 epoch: 742/1000\n",
+ "fps step: 58130 fps step and policy inference: 24847 fps total: 16799 epoch: 743/1000\n",
+ "saving next best rewards: [1394.1024]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 60735 fps step and policy inference: 26802 fps total: 17698 epoch: 744/1000\n",
+ "fps step: 64893 fps step and policy inference: 29414 fps total: 18795 epoch: 745/1000\n",
+ "fps step: 64687 fps step and policy inference: 29430 fps total: 18732 epoch: 746/1000\n",
+ "fps step: 58438 fps step and policy inference: 25438 fps total: 16910 epoch: 747/1000\n",
+ "saving next best rewards: [1399.6426]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 60846 fps step and policy inference: 26661 fps total: 17706 epoch: 748/1000\n",
+ "fps step: 65074 fps step and policy inference: 29438 fps total: 18591 epoch: 749/1000\n",
+ "fps step: 64931 fps step and policy inference: 29368 fps total: 18902 epoch: 750/1000\n",
+ "fps step: 55713 fps step and policy inference: 23426 fps total: 16038 epoch: 751/1000\n",
+ "fps step: 57269 fps step and policy inference: 23025 fps total: 15797 epoch: 752/1000\n",
+ "fps step: 64701 fps step and policy inference: 29322 fps total: 18625 epoch: 753/1000\n",
+ "fps step: 63478 fps step and policy inference: 27677 fps total: 18091 epoch: 754/1000\n",
+ "fps step: 60536 fps step and policy inference: 27224 fps total: 17840 epoch: 755/1000\n",
+ "fps step: 62573 fps step and policy inference: 27949 fps total: 18328 epoch: 756/1000\n",
+ "fps step: 64997 fps step and policy inference: 29179 fps total: 18742 epoch: 757/1000\n",
+ "fps step: 64765 fps step and policy inference: 29330 fps total: 18846 epoch: 758/1000\n",
+ "fps step: 54384 fps step and policy inference: 22655 fps total: 15751 epoch: 759/1000\n",
+ "fps step: 63392 fps step and policy inference: 27882 fps total: 18293 epoch: 760/1000\n",
+ "fps step: 65422 fps step and policy inference: 29650 fps total: 18883 epoch: 761/1000\n",
+ "fps step: 63501 fps step and policy inference: 27136 fps total: 17856 epoch: 762/1000\n",
+ "fps step: 60204 fps step and policy inference: 26891 fps total: 17782 epoch: 763/1000\n",
+ "fps step: 64059 fps step and policy inference: 28091 fps total: 18209 epoch: 764/1000\n",
+ "fps step: 62769 fps step and policy inference: 27775 fps total: 17888 epoch: 765/1000\n",
+ "fps step: 62257 fps step and policy inference: 27272 fps total: 17930 epoch: 766/1000\n",
+ "fps step: 60899 fps step and policy inference: 27406 fps total: 17921 epoch: 767/1000\n",
+ "fps step: 65092 fps step and policy inference: 29398 fps total: 18776 epoch: 768/1000\n",
+ "fps step: 65059 fps step and policy inference: 29444 fps total: 18608 epoch: 769/1000\n",
+ "fps step: 61641 fps step and policy inference: 26676 fps total: 17405 epoch: 770/1000\n",
+ "fps step: 60601 fps step and policy inference: 27555 fps total: 18012 epoch: 771/1000\n",
+ "fps step: 64738 fps step and policy inference: 28909 fps total: 18563 epoch: 772/1000\n",
+ "fps step: 62984 fps step and policy inference: 27596 fps total: 18108 epoch: 773/1000\n",
+ "fps step: 63896 fps step and policy inference: 28230 fps total: 18393 epoch: 774/1000\n",
+ "fps step: 59087 fps step and policy inference: 26251 fps total: 17388 epoch: 775/1000\n",
+ "fps step: 63428 fps step and policy inference: 28227 fps total: 18212 epoch: 776/1000\n",
+ "fps step: 62503 fps step and policy inference: 27376 fps total: 17930 epoch: 777/1000\n",
+ "fps step: 65040 fps step and policy inference: 29430 fps total: 18794 epoch: 778/1000\n",
+ "fps step: 61212 fps step and policy inference: 27660 fps total: 18048 epoch: 779/1000\n",
+ "fps step: 64340 fps step and policy inference: 28970 fps total: 18539 epoch: 780/1000\n",
+ "fps step: 63720 fps step and policy inference: 28526 fps total: 18462 epoch: 781/1000\n",
+ "fps step: 64291 fps step and policy inference: 29193 fps total: 18644 epoch: 782/1000\n",
+ "fps step: 59294 fps step and policy inference: 26298 fps total: 17346 epoch: 783/1000\n",
+ "fps step: 61740 fps step and policy inference: 27232 fps total: 18010 epoch: 784/1000\n",
+ "fps step: 61569 fps step and policy inference: 26847 fps total: 17783 epoch: 785/1000\n",
+ "fps step: 63587 fps step and policy inference: 29031 fps total: 18536 epoch: 786/1000\n",
+ "saving next best rewards: [1405.6737]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 62186 fps step and policy inference: 28023 fps total: 18177 epoch: 787/1000\n",
+ "fps step: 57388 fps step and policy inference: 24317 fps total: 16501 epoch: 788/1000\n",
+ "fps step: 54435 fps step and policy inference: 21933 fps total: 15350 epoch: 789/1000\n",
+ "fps step: 58481 fps step and policy inference: 25166 fps total: 16947 epoch: 790/1000\n",
+ "saving next best rewards: [1414.5619]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 64567 fps step and policy inference: 29366 fps total: 18695 epoch: 791/1000\n",
+ "fps step: 65338 fps step and policy inference: 29440 fps total: 18755 epoch: 792/1000\n",
+ "fps step: 60359 fps step and policy inference: 25753 fps total: 17174 epoch: 793/1000\n",
+ "fps step: 61519 fps step and policy inference: 27843 fps total: 17880 epoch: 794/1000\n",
+ "saving next best rewards: [1424.5096]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 57198 fps step and policy inference: 23734 fps total: 16219 epoch: 795/1000\n",
+ "fps step: 64958 fps step and policy inference: 29344 fps total: 18786 epoch: 796/1000\n",
+ "fps step: 56453 fps step and policy inference: 23349 fps total: 16074 epoch: 797/1000\n",
+ "fps step: 57300 fps step and policy inference: 24662 fps total: 16662 epoch: 798/1000\n",
+ "saving next best rewards: [1425.7277]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 61912 fps step and policy inference: 27063 fps total: 17556 epoch: 799/1000\n",
+ "fps step: 60417 fps step and policy inference: 26121 fps total: 17060 epoch: 800/1000\n",
+ "fps step: 63766 fps step and policy inference: 28430 fps total: 18036 epoch: 801/1000\n",
+ "fps step: 57450 fps step and policy inference: 24434 fps total: 16663 epoch: 802/1000\n",
+ "fps step: 63170 fps step and policy inference: 27481 fps total: 17869 epoch: 803/1000\n",
+ "fps step: 65788 fps step and policy inference: 29547 fps total: 18721 epoch: 804/1000\n",
+ "fps step: 63652 fps step and policy inference: 27876 fps total: 17850 epoch: 805/1000\n",
+ "fps step: 51227 fps step and policy inference: 20606 fps total: 14768 epoch: 806/1000\n",
+ "fps step: 60015 fps step and policy inference: 25597 fps total: 17053 epoch: 807/1000\n",
+ "fps step: 60902 fps step and policy inference: 26354 fps total: 17391 epoch: 808/1000\n",
+ "fps step: 61033 fps step and policy inference: 26106 fps total: 17212 epoch: 809/1000\n",
+ "fps step: 61706 fps step and policy inference: 28034 fps total: 18154 epoch: 810/1000\n",
+ "fps step: 63632 fps step and policy inference: 28189 fps total: 18322 epoch: 811/1000\n",
+ "fps step: 65128 fps step and policy inference: 29361 fps total: 18554 epoch: 812/1000\n",
+ "fps step: 63577 fps step and policy inference: 28355 fps total: 18179 epoch: 813/1000\n",
+ "fps step: 61753 fps step and policy inference: 27956 fps total: 18220 epoch: 814/1000\n",
+ "fps step: 64780 fps step and policy inference: 29343 fps total: 18816 epoch: 815/1000\n",
+ "fps step: 61977 fps step and policy inference: 26909 fps total: 17707 epoch: 816/1000\n",
+ "fps step: 64488 fps step and policy inference: 29156 fps total: 18525 epoch: 817/1000\n",
+ "fps step: 56748 fps step and policy inference: 24412 fps total: 16702 epoch: 818/1000\n",
+ "fps step: 64196 fps step and policy inference: 28710 fps total: 18340 epoch: 819/1000\n",
+ "fps step: 62832 fps step and policy inference: 27970 fps total: 18220 epoch: 820/1000\n",
+ "fps step: 63312 fps step and policy inference: 27595 fps total: 18016 epoch: 821/1000\n",
+ "fps step: 58759 fps step and policy inference: 25688 fps total: 17241 epoch: 822/1000\n",
+ "fps step: 64296 fps step and policy inference: 28758 fps total: 18494 epoch: 823/1000\n",
+ "fps step: 64635 fps step and policy inference: 29320 fps total: 18859 epoch: 824/1000\n",
+ "fps step: 62600 fps step and policy inference: 27888 fps total: 18097 epoch: 825/1000\n",
+ "fps step: 54799 fps step and policy inference: 23771 fps total: 16180 epoch: 826/1000\n",
+ "fps step: 61593 fps step and policy inference: 27847 fps total: 18046 epoch: 827/1000\n",
+ "fps step: 59549 fps step and policy inference: 25363 fps total: 17005 epoch: 828/1000\n",
+ "fps step: 54408 fps step and policy inference: 23118 fps total: 15875 epoch: 829/1000\n",
+ "fps step: 58900 fps step and policy inference: 25857 fps total: 17233 epoch: 830/1000\n",
+ "fps step: 62115 fps step and policy inference: 27866 fps total: 17690 epoch: 831/1000\n",
+ "fps step: 63260 fps step and policy inference: 28584 fps total: 18539 epoch: 832/1000\n",
+ "fps step: 61573 fps step and policy inference: 28393 fps total: 18322 epoch: 833/1000\n",
+ "fps step: 61700 fps step and policy inference: 27220 fps total: 17646 epoch: 834/1000\n",
+ "fps step: 60305 fps step and policy inference: 26627 fps total: 17606 epoch: 835/1000\n",
+ "fps step: 61926 fps step and policy inference: 28216 fps total: 18280 epoch: 836/1000\n",
+ "fps step: 52283 fps step and policy inference: 21530 fps total: 15216 epoch: 837/1000\n",
+ "fps step: 58770 fps step and policy inference: 25534 fps total: 16871 epoch: 838/1000\n",
+ "fps step: 61112 fps step and policy inference: 26890 fps total: 17651 epoch: 839/1000\n",
+ "fps step: 59952 fps step and policy inference: 25966 fps total: 17330 epoch: 840/1000\n",
+ "fps step: 59896 fps step and policy inference: 27350 fps total: 17898 epoch: 841/1000\n",
+ "fps step: 56240 fps step and policy inference: 23252 fps total: 15922 epoch: 842/1000\n",
+ "fps step: 61868 fps step and policy inference: 27635 fps total: 18110 epoch: 843/1000\n",
+ "fps step: 61611 fps step and policy inference: 26999 fps total: 17853 epoch: 844/1000\n",
+ "fps step: 58313 fps step and policy inference: 25726 fps total: 16947 epoch: 845/1000\n",
+ "fps step: 60287 fps step and policy inference: 25070 fps total: 16797 epoch: 846/1000\n",
+ "fps step: 64883 fps step and policy inference: 29618 fps total: 18886 epoch: 847/1000\n",
+ "fps step: 61223 fps step and policy inference: 26645 fps total: 17493 epoch: 848/1000\n",
+ "fps step: 60807 fps step and policy inference: 27627 fps total: 18112 epoch: 849/1000\n",
+ "fps step: 60501 fps step and policy inference: 26161 fps total: 17321 epoch: 850/1000\n",
+ "fps step: 62006 fps step and policy inference: 27136 fps total: 17843 epoch: 851/1000\n",
+ "fps step: 60146 fps step and policy inference: 26099 fps total: 17348 epoch: 852/1000\n",
+ "fps step: 58225 fps step and policy inference: 25365 fps total: 16942 epoch: 853/1000\n",
+ "fps step: 61986 fps step and policy inference: 27425 fps total: 18056 epoch: 854/1000\n",
+ "fps step: 63673 fps step and policy inference: 28654 fps total: 18318 epoch: 855/1000\n",
+ "fps step: 64309 fps step and policy inference: 29305 fps total: 18686 epoch: 856/1000\n",
+ "fps step: 54689 fps step and policy inference: 23082 fps total: 15860 epoch: 857/1000\n",
+ "fps step: 61882 fps step and policy inference: 27191 fps total: 17580 epoch: 858/1000\n",
+ "fps step: 57184 fps step and policy inference: 24479 fps total: 16672 epoch: 859/1000\n",
+ "fps step: 56405 fps step and policy inference: 23531 fps total: 16114 epoch: 860/1000\n",
+ "fps step: 59907 fps step and policy inference: 26959 fps total: 17529 epoch: 861/1000\n",
+ "fps step: 63627 fps step and policy inference: 29161 fps total: 18680 epoch: 862/1000\n",
+ "fps step: 64293 fps step and policy inference: 29479 fps total: 18906 epoch: 863/1000\n",
+ "fps step: 63906 fps step and policy inference: 29379 fps total: 18730 epoch: 864/1000\n",
+ "fps step: 60450 fps step and policy inference: 26820 fps total: 17764 epoch: 865/1000\n",
+ "fps step: 62607 fps step and policy inference: 27735 fps total: 17945 epoch: 866/1000\n",
+ "fps step: 57956 fps step and policy inference: 24606 fps total: 16696 epoch: 867/1000\n",
+ "fps step: 62950 fps step and policy inference: 28313 fps total: 18313 epoch: 868/1000\n",
+ "fps step: 58636 fps step and policy inference: 25917 fps total: 17234 epoch: 869/1000\n",
+ "fps step: 65238 fps step and policy inference: 29480 fps total: 18798 epoch: 870/1000\n",
+ "fps step: 63384 fps step and policy inference: 28233 fps total: 18331 epoch: 871/1000\n",
+ "fps step: 63144 fps step and policy inference: 28443 fps total: 18118 epoch: 872/1000\n",
+ "fps step: 60874 fps step and policy inference: 26507 fps total: 17595 epoch: 873/1000\n",
+ "fps step: 61946 fps step and policy inference: 26893 fps total: 17742 epoch: 874/1000\n",
+ "fps step: 57903 fps step and policy inference: 24140 fps total: 16162 epoch: 875/1000\n",
+ "fps step: 62133 fps step and policy inference: 28264 fps total: 18269 epoch: 876/1000\n",
+ "fps step: 63808 fps step and policy inference: 28704 fps total: 18538 epoch: 877/1000\n",
+ "fps step: 65556 fps step and policy inference: 29554 fps total: 18886 epoch: 878/1000\n",
+ "fps step: 64753 fps step and policy inference: 28869 fps total: 18601 epoch: 879/1000\n",
+ "fps step: 60283 fps step and policy inference: 26866 fps total: 17566 epoch: 880/1000\n",
+ "fps step: 62708 fps step and policy inference: 27536 fps total: 17991 epoch: 881/1000\n",
+ "fps step: 63292 fps step and policy inference: 27783 fps total: 17916 epoch: 882/1000\n",
+ "fps step: 65521 fps step and policy inference: 29520 fps total: 18884 epoch: 883/1000\n",
+ "fps step: 60582 fps step and policy inference: 26893 fps total: 17294 epoch: 884/1000\n",
+ "fps step: 65190 fps step and policy inference: 29197 fps total: 18593 epoch: 885/1000\n",
+ "fps step: 64429 fps step and policy inference: 28679 fps total: 18492 epoch: 886/1000\n",
+ "fps step: 64731 fps step and policy inference: 29078 fps total: 18657 epoch: 887/1000\n",
+ "fps step: 62848 fps step and policy inference: 28861 fps total: 18337 epoch: 888/1000\n",
+ "fps step: 59944 fps step and policy inference: 25704 fps total: 17163 epoch: 889/1000\n",
+ "fps step: 61445 fps step and policy inference: 26500 fps total: 17503 epoch: 890/1000\n",
+ "fps step: 64624 fps step and policy inference: 28473 fps total: 18213 epoch: 891/1000\n",
+ "fps step: 59181 fps step and policy inference: 26082 fps total: 17266 epoch: 892/1000\n",
+ "fps step: 59881 fps step and policy inference: 25856 fps total: 17011 epoch: 893/1000\n",
+ "fps step: 64583 fps step and policy inference: 29297 fps total: 18747 epoch: 894/1000\n",
+ "fps step: 55539 fps step and policy inference: 22799 fps total: 15750 epoch: 895/1000\n",
+ "fps step: 55499 fps step and policy inference: 23022 fps total: 15869 epoch: 896/1000\n",
+ "fps step: 64203 fps step and policy inference: 28725 fps total: 18562 epoch: 897/1000\n",
+ "fps step: 59273 fps step and policy inference: 25259 fps total: 17028 epoch: 898/1000\n",
+ "fps step: 61519 fps step and policy inference: 27025 fps total: 17684 epoch: 899/1000\n",
+ "fps step: 55772 fps step and policy inference: 23569 fps total: 16032 epoch: 900/1000\n",
+ "fps step: 61585 fps step and policy inference: 27075 fps total: 17666 epoch: 901/1000\n",
+ "fps step: 62623 fps step and policy inference: 27492 fps total: 17830 epoch: 902/1000\n",
+ "fps step: 65341 fps step and policy inference: 29400 fps total: 18832 epoch: 903/1000\n",
+ "fps step: 60967 fps step and policy inference: 27247 fps total: 17724 epoch: 904/1000\n",
+ "fps step: 60512 fps step and policy inference: 26248 fps total: 17460 epoch: 905/1000\n",
+ "fps step: 63532 fps step and policy inference: 28140 fps total: 18235 epoch: 906/1000\n",
+ "fps step: 60367 fps step and policy inference: 25690 fps total: 17184 epoch: 907/1000\n",
+ "fps step: 62895 fps step and policy inference: 28370 fps total: 18157 epoch: 908/1000\n",
+ "fps step: 65008 fps step and policy inference: 29457 fps total: 18711 epoch: 909/1000\n",
+ "fps step: 63763 fps step and policy inference: 28398 fps total: 18301 epoch: 910/1000\n",
+ "fps step: 60465 fps step and policy inference: 25820 fps total: 17195 epoch: 911/1000\n",
+ "fps step: 60869 fps step and policy inference: 27174 fps total: 17869 epoch: 912/1000\n",
+ "fps step: 61110 fps step and policy inference: 26172 fps total: 17282 epoch: 913/1000\n",
+ "fps step: 62112 fps step and policy inference: 26934 fps total: 17636 epoch: 914/1000\n",
+ "fps step: 60295 fps step and policy inference: 26295 fps total: 17401 epoch: 915/1000\n",
+ "fps step: 61925 fps step and policy inference: 26987 fps total: 17701 epoch: 916/1000\n",
+ "fps step: 62743 fps step and policy inference: 26705 fps total: 17628 epoch: 917/1000\n",
+ "fps step: 65241 fps step and policy inference: 29366 fps total: 18695 epoch: 918/1000\n",
+ "fps step: 60543 fps step and policy inference: 26661 fps total: 17718 epoch: 919/1000\n",
+ "fps step: 62774 fps step and policy inference: 27645 fps total: 18072 epoch: 920/1000\n",
+ "fps step: 65305 fps step and policy inference: 29377 fps total: 18705 epoch: 921/1000\n",
+ "fps step: 62846 fps step and policy inference: 27371 fps total: 17998 epoch: 922/1000\n",
+ "fps step: 53102 fps step and policy inference: 21580 fps total: 15166 epoch: 923/1000\n",
+ "fps step: 64953 fps step and policy inference: 29279 fps total: 18479 epoch: 924/1000\n",
+ "fps step: 62093 fps step and policy inference: 26942 fps total: 17667 epoch: 925/1000\n",
+ "fps step: 63232 fps step and policy inference: 27618 fps total: 17926 epoch: 926/1000\n",
+ "fps step: 62544 fps step and policy inference: 28440 fps total: 18377 epoch: 927/1000\n",
+ "saving next best rewards: [1426.002]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 62722 fps step and policy inference: 26824 fps total: 17650 epoch: 928/1000\n",
+ "fps step: 64052 fps step and policy inference: 28520 fps total: 18406 epoch: 929/1000\n",
+ "fps step: 62612 fps step and policy inference: 27692 fps total: 17941 epoch: 930/1000\n",
+ "fps step: 60558 fps step and policy inference: 26850 fps total: 17653 epoch: 931/1000\n",
+ "saving next best rewards: [1434.0319]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 57064 fps step and policy inference: 23084 fps total: 15666 epoch: 932/1000\n",
+ "fps step: 55983 fps step and policy inference: 23025 fps total: 15866 epoch: 933/1000\n",
+ "fps step: 60273 fps step and policy inference: 25338 fps total: 16972 epoch: 934/1000\n",
+ "fps step: 60194 fps step and policy inference: 26258 fps total: 17405 epoch: 935/1000\n",
+ "saving next best rewards: [1438.5753]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 63351 fps step and policy inference: 28180 fps total: 18227 epoch: 936/1000\n",
+ "fps step: 57293 fps step and policy inference: 23834 fps total: 16221 epoch: 937/1000\n",
+ "fps step: 65128 fps step and policy inference: 29562 fps total: 18689 epoch: 938/1000\n",
+ "fps step: 61508 fps step and policy inference: 27599 fps total: 17988 epoch: 939/1000\n",
+ "saving next best rewards: [1439.3237]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 60866 fps step and policy inference: 26166 fps total: 17288 epoch: 940/1000\n",
+ "fps step: 59155 fps step and policy inference: 25144 fps total: 16955 epoch: 941/1000\n",
+ "fps step: 65483 fps step and policy inference: 29693 fps total: 18971 epoch: 942/1000\n",
+ "fps step: 60957 fps step and policy inference: 27041 fps total: 17460 epoch: 943/1000\n",
+ "fps step: 60740 fps step and policy inference: 26196 fps total: 17329 epoch: 944/1000\n",
+ "fps step: 65710 fps step and policy inference: 29449 fps total: 18782 epoch: 945/1000\n",
+ "fps step: 65489 fps step and policy inference: 29613 fps total: 18869 epoch: 946/1000\n",
+ "fps step: 63046 fps step and policy inference: 28949 fps total: 18520 epoch: 947/1000\n",
+ "fps step: 65135 fps step and policy inference: 29486 fps total: 18710 epoch: 948/1000\n",
+ "fps step: 58646 fps step and policy inference: 24418 fps total: 16667 epoch: 949/1000\n",
+ "fps step: 61280 fps step and policy inference: 26106 fps total: 17377 epoch: 950/1000\n",
+ "fps step: 61203 fps step and policy inference: 27328 fps total: 17836 epoch: 951/1000\n",
+ "fps step: 61239 fps step and policy inference: 26207 fps total: 17312 epoch: 952/1000\n",
+ "fps step: 65237 fps step and policy inference: 29542 fps total: 18838 epoch: 953/1000\n",
+ "fps step: 61777 fps step and policy inference: 26985 fps total: 17755 epoch: 954/1000\n",
+ "fps step: 59421 fps step and policy inference: 25727 fps total: 17117 epoch: 955/1000\n",
+ "fps step: 63533 fps step and policy inference: 28199 fps total: 18371 epoch: 956/1000\n",
+ "fps step: 64278 fps step and policy inference: 28596 fps total: 18450 epoch: 957/1000\n",
+ "fps step: 63753 fps step and policy inference: 28884 fps total: 18541 epoch: 958/1000\n",
+ "fps step: 62489 fps step and policy inference: 27790 fps total: 18152 epoch: 959/1000\n",
+ "fps step: 62830 fps step and policy inference: 27596 fps total: 17557 epoch: 960/1000\n",
+ "fps step: 62475 fps step and policy inference: 27113 fps total: 17595 epoch: 961/1000\n",
+ "fps step: 58453 fps step and policy inference: 25104 fps total: 16847 epoch: 962/1000\n",
+ "fps step: 65070 fps step and policy inference: 29474 fps total: 18828 epoch: 963/1000\n",
+ "fps step: 65784 fps step and policy inference: 29738 fps total: 18519 epoch: 964/1000\n",
+ "fps step: 63191 fps step and policy inference: 27438 fps total: 17955 epoch: 965/1000\n",
+ "fps step: 60883 fps step and policy inference: 26868 fps total: 17727 epoch: 966/1000\n",
+ "fps step: 64428 fps step and policy inference: 28584 fps total: 18251 epoch: 967/1000\n",
+ "fps step: 62989 fps step and policy inference: 27762 fps total: 18138 epoch: 968/1000\n",
+ "fps step: 60044 fps step and policy inference: 25651 fps total: 17034 epoch: 969/1000\n",
+ "fps step: 58854 fps step and policy inference: 24867 fps total: 16842 epoch: 970/1000\n",
+ "saving next best rewards: [1442.3019]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 60630 fps step and policy inference: 25996 fps total: 17234 epoch: 971/1000\n",
+ "fps step: 62706 fps step and policy inference: 27450 fps total: 17966 epoch: 972/1000\n",
+ "fps step: 64820 fps step and policy inference: 28364 fps total: 18385 epoch: 973/1000\n",
+ "fps step: 61977 fps step and policy inference: 28187 fps total: 18310 epoch: 974/1000\n",
+ "saving next best rewards: [1445.5693]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 63988 fps step and policy inference: 28586 fps total: 18453 epoch: 975/1000\n",
+ "fps step: 64620 fps step and policy inference: 29022 fps total: 18581 epoch: 976/1000\n",
+ "fps step: 60260 fps step and policy inference: 25663 fps total: 17272 epoch: 977/1000\n",
+ "fps step: 55298 fps step and policy inference: 23010 fps total: 15857 epoch: 978/1000\n",
+ "saving next best rewards: [1449.3958]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 51801 fps step and policy inference: 20481 fps total: 14690 epoch: 979/1000\n",
+ "fps step: 64067 fps step and policy inference: 28542 fps total: 18243 epoch: 980/1000\n",
+ "fps step: 58064 fps step and policy inference: 23918 fps total: 16344 epoch: 981/1000\n",
+ "fps step: 60421 fps step and policy inference: 26684 fps total: 17631 epoch: 982/1000\n",
+ "fps step: 64310 fps step and policy inference: 29016 fps total: 18673 epoch: 983/1000\n",
+ "fps step: 64103 fps step and policy inference: 28415 fps total: 18441 epoch: 984/1000\n",
+ "fps step: 65351 fps step and policy inference: 29699 fps total: 18798 epoch: 985/1000\n",
+ "fps step: 63338 fps step and policy inference: 29063 fps total: 18599 epoch: 986/1000\n",
+ "fps step: 64836 fps step and policy inference: 29492 fps total: 18882 epoch: 987/1000\n",
+ "fps step: 65358 fps step and policy inference: 29301 fps total: 18663 epoch: 988/1000\n",
+ "fps step: 65038 fps step and policy inference: 29440 fps total: 18740 epoch: 989/1000\n",
+ "fps step: 60684 fps step and policy inference: 26817 fps total: 17562 epoch: 990/1000\n",
+ "saving next best rewards: [1450.2551]\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/HalfCheetah-v4_envpool.pth'\n",
+ "fps step: 65414 fps step and policy inference: 29733 fps total: 18953 epoch: 991/1000\n",
+ "fps step: 65352 fps step and policy inference: 29688 fps total: 18923 epoch: 992/1000\n",
+ "fps step: 65032 fps step and policy inference: 28509 fps total: 18128 epoch: 993/1000\n",
+ "fps step: 52134 fps step and policy inference: 20856 fps total: 14800 epoch: 994/1000\n",
+ "fps step: 62367 fps step and policy inference: 26950 fps total: 17729 epoch: 995/1000\n",
+ "fps step: 65196 fps step and policy inference: 29371 fps total: 18735 epoch: 996/1000\n",
+ "fps step: 65511 fps step and policy inference: 29571 fps total: 18722 epoch: 997/1000\n",
+ "fps step: 62206 fps step and policy inference: 27874 fps total: 18037 epoch: 998/1000\n",
+ "fps step: 63518 fps step and policy inference: 28140 fps total: 18177 epoch: 999/1000\n",
+ "fps step: 65475 fps step and policy inference: 29327 fps total: 18755 epoch: 1000/1000\n",
+ "=> saving checkpoint 'runs/HalfCheetah-v4_envpool_23-10-41-37/nn/last_HalfCheetah-v4_envpoolep1000rew[1448.0536].pth'\n",
+ "MAX EPOCHS NUM!\n"
+ ]
+ }
+ ]
+ }
+ ]
+}
\ No newline at end of file
diff --git a/demo/envpool_demo_pong.ipynb b/demo/envpool_demo_pong.ipynb
new file mode 100644
index 00000000..2e21945b
--- /dev/null
+++ b/demo/envpool_demo_pong.ipynb
@@ -0,0 +1,895 @@
+{
+ "nbformat": 4,
+ "nbformat_minor": 0,
+ "metadata": {
+ "colab": {
+ "provenance": [],
+ "collapsed_sections": []
+ },
+ "kernelspec": {
+ "name": "python3",
+ "display_name": "Python 3"
+ },
+ "language_info": {
+ "name": "python"
+ },
+ "accelerator": "GPU",
+ "gpuClass": "standard"
+ },
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "source": [
+ "[](https://colab.research.google.com/github/sail-sg/envpool/blob/main/demo/envpool-demo-pong.ipynb)"
+ ],
+ "metadata": {
+ "id": "wSww-5lcUxqN"
+ }
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "id": "YMhfVuVIQQd9"
+ },
+ "outputs": [],
+ "source": [
+ "# OS: Ubuntu 18.04.4 LTS x86_64\n",
+ "# Kernel: 4.18.0-15-generic\n",
+ "# CPU: Intel(R) Core(TM) i9-10920X CPU (24) @ 3.50GHz\n",
+ "# GPU: NVIDIA GeForce RTX 2080 Ti\n",
+ "!apt-get install -y \\\n",
+ " libgl1-mesa-dev \\\n",
+ " libgl1-mesa-glx \\\n",
+ " libglew-dev \\\n",
+ " libosmesa6-dev \\\n",
+ " software-properties-common\n",
+ "\n",
+ "!apt-get install -y patchelf\n",
+ "!pip install git+https://github.com/Denys88/rl_games\n",
+ "!pip install envpool\n",
+ "!pip install gym\n",
+ "\n",
+ "!pip install free-mujoco-py\n",
+ "!apt-get install -y xvfb python-opengl ffmpeg > /dev/null 2>&1\n",
+ "!pip install imageio==2.4.1\n",
+ "!pip install -U colabgymrender"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "from rl_games.torch_runner import Runner"
+ ],
+ "metadata": {
+ "id": "weGrLVCFTKdD"
+ },
+ "execution_count": null,
+ "outputs": []
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "## pong_envpool config:\n",
+ "pong_config = {'params': {'seed': 322, \n",
+ " 'algo': {'name': 'a2c_discrete'}, \n",
+ " 'model': {'name': 'discrete_a2c'}, \n",
+ " 'network': {'name': 'actor_critic', \n",
+ " 'separate': False, \n",
+ " 'space': {'discrete': None}, \n",
+ " 'cnn': {'permute_input': False, \n",
+ " 'type': 'conv2d', \n",
+ " 'activation': 'elu', \n",
+ " 'initializer': {'name': 'default'}, \n",
+ " 'regularizer': {'name': 'None'}, \n",
+ " 'convs': [{'filters': 32, 'kernel_size': 8, 'strides': 4, 'padding': 0}, \n",
+ " {'filters': 64, 'kernel_size': 4, 'strides': 2, 'padding': 0}, \n",
+ " {'filters': 64, 'kernel_size': 3, 'strides': 1, 'padding': 0}]}, \n",
+ " 'mlp': {'units': [512], 'activation': 'elu', 'initializer': {'name': 'orthogonal_initializer', 'gain': 1.41421356237}}}, \n",
+ " 'config': {'name': 'Pong-v5_envpool', \n",
+ " 'env_name': 'envpool', \n",
+ " 'score_to_win': 20.0, \n",
+ " 'normalize_value': True, \n",
+ " 'normalize_input': True, \n",
+ " 'reward_shaper': {'min_val': -1, 'max_val': 1}, \n",
+ " 'normalize_advantage': True, \n",
+ " 'gamma': 0.99, \n",
+ " 'tau': 0.95, \n",
+ " 'learning_rate': '3e-4', \n",
+ " 'lr_schedule': 'adaptive', \n",
+ " 'kl_threshold': 0.01, \n",
+ " 'grad_norm': 1.0, \n",
+ " 'entropy_coef': 0.01, \n",
+ " 'truncate_grads': True, \n",
+ " 'e_clip': 0.2, \n",
+ " 'clip_value': False, \n",
+ " 'num_actors': 64, \n",
+ " 'horizon_length': 128, \n",
+ " 'minibatch_size': 2048, \n",
+ " 'mini_epochs': 4, \n",
+ " 'critic_coef': 2, \n",
+ " 'max_epochs': 500, \n",
+ " 'use_diagnostics': False, \n",
+ " 'env_config': {'env_name': 'Pong-v5', 'has_lives': False}, \n",
+ " 'player': {'render': False, 'games_num': 100, 'n_game_life': 1, 'determenistic': True}}}}"
+ ],
+ "metadata": {
+ "id": "jghnij1BTOiH"
+ },
+ "execution_count": null,
+ "outputs": []
+ },
+ {
+ "cell_type": "code",
+ "source": [
+ "runner = Runner()\n",
+ "runner.load(pong_config)\n",
+ "runner.run({\n",
+ " 'train': True,\n",
+ "})"
+ ],
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "id": "w-DzDWZGUxQW",
+ "outputId": "493b9918-eda9-4a52-c94b-7bedbee1c558"
+ },
+ "execution_count": null,
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "self.seed = 322\n",
+ "Started to train\n",
+ "current training device: cuda:0\n",
+ "conv_name: conv2d\n",
+ "build mlp: 3136\n",
+ "RunningMeanStd: (1,)\n",
+ "RunningMeanStd: (4, 84, 84)\n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stderr",
+ "text": [
+ "/home/liubo/.local/lib/python3.8/site-packages/rl_games/common/experience.py:341: DeprecationWarning: `np.long` is a deprecated alias for `np.compat.long`. To silence this warning, use `np.compat.long` by itself. In the likely event your code does not need to work on Python 2 you can use the builtin `int` for which `np.compat.long` is itself an alias. Doing this will not modify any behaviour and is safe. When replacing `np.long`, you may wish to use e.g. `np.int64` or `np.int32` to specify the precision. If you wish to review your current use, check the release note link for additional information.\n",
+ "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
+ " self.tensor_dict['actions'] = self._create_tensor_from_space(gym.spaces.Box(low=0, high=1,shape=self.actions_shape, dtype=np.long), obs_base_shape)\n"
+ ]
+ },
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "fps step: 11849 fps step and policy inference: 9092 fps total: 5652 epoch: 1/500\n",
+ "fps step: 11318 fps step and policy inference: 8902 fps total: 6340 epoch: 2/500\n",
+ "fps step: 10767 fps step and policy inference: 8437 fps total: 6099 epoch: 3/500\n",
+ "fps step: 10548 fps step and policy inference: 8191 fps total: 5964 epoch: 4/500\n",
+ "fps step: 9426 fps step and policy inference: 7164 fps total: 5403 epoch: 5/500\n",
+ "fps step: 10008 fps step and policy inference: 7780 fps total: 5743 epoch: 6/500\n",
+ "fps step: 7491 fps step and policy inference: 6021 fps total: 4720 epoch: 7/500\n",
+ "fps step: 6301 fps step and policy inference: 5112 fps total: 4146 epoch: 8/500\n",
+ "fps step: 9080 fps step and policy inference: 7118 fps total: 5373 epoch: 9/500\n",
+ "fps step: 11450 fps step and policy inference: 9020 fps total: 6388 epoch: 10/500\n",
+ "fps step: 10282 fps step and policy inference: 7925 fps total: 5822 epoch: 11/500\n",
+ "fps step: 9751 fps step and policy inference: 7514 fps total: 5595 epoch: 12/500\n",
+ "fps step: 9726 fps step and policy inference: 7670 fps total: 5677 epoch: 13/500\n",
+ "fps step: 8260 fps step and policy inference: 6258 fps total: 4860 epoch: 14/500\n",
+ "fps step: 7061 fps step and policy inference: 5648 fps total: 4489 epoch: 15/500\n",
+ "fps step: 7864 fps step and policy inference: 6121 fps total: 4782 epoch: 16/500\n",
+ "fps step: 8639 fps step and policy inference: 7048 fps total: 5329 epoch: 17/500\n",
+ "fps step: 9205 fps step and policy inference: 7044 fps total: 5322 epoch: 18/500\n",
+ "fps step: 9991 fps step and policy inference: 7796 fps total: 5744 epoch: 19/500\n",
+ "fps step: 9746 fps step and policy inference: 7539 fps total: 5602 epoch: 20/500\n",
+ "fps step: 11440 fps step and policy inference: 8887 fps total: 6313 epoch: 21/500\n",
+ "fps step: 10600 fps step and policy inference: 8218 fps total: 5961 epoch: 22/500\n",
+ "fps step: 10379 fps step and policy inference: 8049 fps total: 5880 epoch: 23/500\n",
+ "fps step: 10150 fps step and policy inference: 7868 fps total: 5784 epoch: 24/500\n",
+ "fps step: 10588 fps step and policy inference: 8375 fps total: 6049 epoch: 25/500\n",
+ "fps step: 9702 fps step and policy inference: 7661 fps total: 5661 epoch: 26/500\n",
+ "fps step: 9213 fps step and policy inference: 7453 fps total: 5550 epoch: 27/500\n",
+ "fps step: 8447 fps step and policy inference: 6594 fps total: 5057 epoch: 28/500\n",
+ "fps step: 8010 fps step and policy inference: 6259 fps total: 4853 epoch: 29/500\n",
+ "fps step: 8213 fps step and policy inference: 6496 fps total: 5000 epoch: 30/500\n",
+ "fps step: 8425 fps step and policy inference: 6553 fps total: 5033 epoch: 31/500\n",
+ "fps step: 9779 fps step and policy inference: 7705 fps total: 5685 epoch: 32/500\n",
+ "fps step: 9767 fps step and policy inference: 7593 fps total: 5625 epoch: 33/500\n",
+ "fps step: 9725 fps step and policy inference: 7656 fps total: 5657 epoch: 34/500\n",
+ "fps step: 9525 fps step and policy inference: 7398 fps total: 5516 epoch: 35/500\n",
+ "fps step: 9963 fps step and policy inference: 7631 fps total: 5643 epoch: 36/500\n",
+ "fps step: 11256 fps step and policy inference: 8859 fps total: 6290 epoch: 37/500\n",
+ "fps step: 10843 fps step and policy inference: 8541 fps total: 6131 epoch: 38/500\n",
+ "fps step: 11956 fps step and policy inference: 9404 fps total: 6560 epoch: 39/500\n",
+ "fps step: 10633 fps step and policy inference: 8435 fps total: 6074 epoch: 40/500\n",
+ "fps step: 11297 fps step and policy inference: 8843 fps total: 6283 epoch: 41/500\n",
+ "fps step: 9894 fps step and policy inference: 7692 fps total: 5685 epoch: 42/500\n",
+ "fps step: 11051 fps step and policy inference: 8612 fps total: 6169 epoch: 43/500\n",
+ "fps step: 9465 fps step and policy inference: 7364 fps total: 5495 epoch: 44/500\n",
+ "fps step: 11128 fps step and policy inference: 8773 fps total: 6246 epoch: 45/500\n",
+ "fps step: 10391 fps step and policy inference: 8104 fps total: 5901 epoch: 46/500\n",
+ "fps step: 9384 fps step and policy inference: 7361 fps total: 5493 epoch: 47/500\n",
+ "fps step: 8734 fps step and policy inference: 6835 fps total: 5192 epoch: 48/500\n",
+ "fps step: 8777 fps step and policy inference: 6818 fps total: 5181 epoch: 49/500\n",
+ "fps step: 8144 fps step and policy inference: 6411 fps total: 4944 epoch: 50/500\n",
+ "fps step: 8705 fps step and policy inference: 6878 fps total: 5209 epoch: 51/500\n",
+ "fps step: 9128 fps step and policy inference: 7180 fps total: 5381 epoch: 52/500\n",
+ "fps step: 9166 fps step and policy inference: 7276 fps total: 5435 epoch: 53/500\n",
+ "fps step: 9397 fps step and policy inference: 7263 fps total: 5433 epoch: 54/500\n",
+ "fps step: 8347 fps step and policy inference: 6542 fps total: 5019 epoch: 55/500\n",
+ "fps step: 9542 fps step and policy inference: 7383 fps total: 5503 epoch: 56/500\n",
+ "fps step: 10014 fps step and policy inference: 7747 fps total: 5701 epoch: 57/500\n",
+ "fps step: 10784 fps step and policy inference: 8512 fps total: 6104 epoch: 58/500\n",
+ "fps step: 10798 fps step and policy inference: 8435 fps total: 6062 epoch: 59/500\n",
+ "fps step: 11156 fps step and policy inference: 8602 fps total: 6147 epoch: 60/500\n",
+ "fps step: 11073 fps step and policy inference: 8710 fps total: 6205 epoch: 61/500\n",
+ "fps step: 9472 fps step and policy inference: 7175 fps total: 5378 epoch: 62/500\n",
+ "fps step: 9861 fps step and policy inference: 7774 fps total: 5710 epoch: 63/500\n",
+ "fps step: 10183 fps step and policy inference: 7988 fps total: 5826 epoch: 64/500\n",
+ "fps step: 8558 fps step and policy inference: 6478 fps total: 4977 epoch: 65/500\n",
+ "fps step: 8705 fps step and policy inference: 6849 fps total: 5199 epoch: 66/500\n",
+ "fps step: 8354 fps step and policy inference: 6625 fps total: 5069 epoch: 67/500\n",
+ "fps step: 7300 fps step and policy inference: 5627 fps total: 4462 epoch: 68/500\n",
+ "fps step: 8379 fps step and policy inference: 6438 fps total: 4957 epoch: 69/500\n",
+ "fps step: 8769 fps step and policy inference: 6878 fps total: 5214 epoch: 70/500\n",
+ "fps step: 9155 fps step and policy inference: 7201 fps total: 5399 epoch: 71/500\n",
+ "fps step: 9007 fps step and policy inference: 6941 fps total: 5250 epoch: 72/500\n",
+ "fps step: 9948 fps step and policy inference: 7777 fps total: 5717 epoch: 73/500\n",
+ "fps step: 10541 fps step and policy inference: 8192 fps total: 5941 epoch: 74/500\n",
+ "fps step: 10072 fps step and policy inference: 7927 fps total: 5797 epoch: 75/500\n",
+ "fps step: 10543 fps step and policy inference: 8325 fps total: 6001 epoch: 76/500\n",
+ "fps step: 9620 fps step and policy inference: 7411 fps total: 5510 epoch: 77/500\n",
+ "fps step: 10720 fps step and policy inference: 8471 fps total: 6074 epoch: 78/500\n",
+ "fps step: 8939 fps step and policy inference: 6890 fps total: 5219 epoch: 79/500\n",
+ "fps step: 10368 fps step and policy inference: 8113 fps total: 5893 epoch: 80/500\n",
+ "fps step: 7668 fps step and policy inference: 5899 fps total: 4632 epoch: 81/500\n",
+ "fps step: 9229 fps step and policy inference: 7332 fps total: 5469 epoch: 82/500\n",
+ "fps step: 8178 fps step and policy inference: 6256 fps total: 4848 epoch: 83/500\n",
+ "fps step: 9179 fps step and policy inference: 7317 fps total: 5461 epoch: 84/500\n",
+ "fps step: 7815 fps step and policy inference: 6077 fps total: 4741 epoch: 85/500\n",
+ "fps step: 8977 fps step and policy inference: 6960 fps total: 5264 epoch: 86/500\n",
+ "fps step: 8373 fps step and policy inference: 6560 fps total: 5027 epoch: 87/500\n",
+ "fps step: 10279 fps step and policy inference: 7953 fps total: 5809 epoch: 88/500\n",
+ "fps step: 10166 fps step and policy inference: 7964 fps total: 5814 epoch: 89/500\n",
+ "fps step: 9232 fps step and policy inference: 7118 fps total: 5349 epoch: 90/500\n",
+ "fps step: 8951 fps step and policy inference: 6815 fps total: 5172 epoch: 91/500\n",
+ "fps step: 9574 fps step and policy inference: 7522 fps total: 5577 epoch: 92/500\n",
+ "fps step: 10061 fps step and policy inference: 7971 fps total: 5817 epoch: 93/500\n",
+ "fps step: 9071 fps step and policy inference: 6990 fps total: 5276 epoch: 94/500\n",
+ "fps step: 8400 fps step and policy inference: 6653 fps total: 5082 epoch: 95/500\n",
+ "fps step: 8288 fps step and policy inference: 6619 fps total: 5062 epoch: 96/500\n",
+ "fps step: 10456 fps step and policy inference: 8152 fps total: 5905 epoch: 97/500\n",
+ "fps step: 9020 fps step and policy inference: 7077 fps total: 5329 epoch: 98/500\n",
+ "fps step: 7633 fps step and policy inference: 6008 fps total: 4694 epoch: 99/500\n",
+ "fps step: 8805 fps step and policy inference: 6885 fps total: 5218 epoch: 100/500\n",
+ "saving next best rewards: [14.216911]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 7531 fps step and policy inference: 5608 fps total: 4441 epoch: 101/500\n",
+ "saving next best rewards: [14.405285]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9184 fps step and policy inference: 7164 fps total: 5369 epoch: 102/500\n",
+ "saving next best rewards: [14.5568285]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 10136 fps step and policy inference: 7790 fps total: 5724 epoch: 103/500\n",
+ "saving next best rewards: [14.601259]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9620 fps step and policy inference: 7463 fps total: 5540 epoch: 104/500\n",
+ "saving next best rewards: [14.688693]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9927 fps step and policy inference: 7709 fps total: 5673 epoch: 105/500\n",
+ "saving next best rewards: [14.764588]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9168 fps step and policy inference: 7149 fps total: 5368 epoch: 106/500\n",
+ "saving next best rewards: [14.931184]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9774 fps step and policy inference: 7435 fps total: 5521 epoch: 107/500\n",
+ "saving next best rewards: [14.951872]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9036 fps step and policy inference: 6879 fps total: 5211 epoch: 108/500\n",
+ "saving next best rewards: [15.092006]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9841 fps step and policy inference: 7708 fps total: 5670 epoch: 109/500\n",
+ "saving next best rewards: [15.322743]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 7745 fps step and policy inference: 5965 fps total: 4672 epoch: 110/500\n",
+ "saving next best rewards: [15.607838]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9167 fps step and policy inference: 7029 fps total: 5297 epoch: 111/500\n",
+ "saving next best rewards: [15.718489]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8733 fps step and policy inference: 6742 fps total: 5131 epoch: 112/500\n",
+ "saving next best rewards: [15.91862]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 7613 fps step and policy inference: 6036 fps total: 4714 epoch: 113/500\n",
+ "saving next best rewards: [16.332552]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8786 fps step and policy inference: 6768 fps total: 5148 epoch: 114/500\n",
+ "saving next best rewards: [16.428352]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9201 fps step and policy inference: 7167 fps total: 5375 epoch: 115/500\n",
+ "saving next best rewards: [16.499426]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 7106 fps step and policy inference: 5372 fps total: 4298 epoch: 116/500\n",
+ "saving next best rewards: [16.68881]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 10166 fps step and policy inference: 8030 fps total: 5849 epoch: 117/500\n",
+ "saving next best rewards: [16.819283]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9098 fps step and policy inference: 6879 fps total: 5207 epoch: 118/500\n",
+ "fps step: 9942 fps step and policy inference: 7776 fps total: 5708 epoch: 119/500\n",
+ "saving next best rewards: [16.85109]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 10083 fps step and policy inference: 7929 fps total: 5792 epoch: 120/500\n",
+ "saving next best rewards: [16.96527]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 10931 fps step and policy inference: 8568 fps total: 6126 epoch: 121/500\n",
+ "saving next best rewards: [16.985617]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9287 fps step and policy inference: 7354 fps total: 5479 epoch: 122/500\n",
+ "saving next best rewards: [17.13335]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9703 fps step and policy inference: 7468 fps total: 5543 epoch: 123/500\n",
+ "saving next best rewards: [17.301378]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 7415 fps step and policy inference: 5631 fps total: 4459 epoch: 124/500\n",
+ "saving next best rewards: [17.41403]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9389 fps step and policy inference: 7418 fps total: 5515 epoch: 125/500\n",
+ "saving next best rewards: [17.585842]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9387 fps step and policy inference: 7244 fps total: 5411 epoch: 126/500\n",
+ "saving next best rewards: [17.633883]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 7779 fps step and policy inference: 6017 fps total: 4700 epoch: 127/500\n",
+ "saving next best rewards: [17.794622]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9073 fps step and policy inference: 6975 fps total: 5265 epoch: 128/500\n",
+ "saving next best rewards: [17.860123]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8957 fps step and policy inference: 7022 fps total: 5292 epoch: 129/500\n",
+ "saving next best rewards: [17.985346]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8422 fps step and policy inference: 6510 fps total: 4991 epoch: 130/500\n",
+ "saving next best rewards: [18.073238]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8676 fps step and policy inference: 6740 fps total: 5128 epoch: 131/500\n",
+ "saving next best rewards: [18.15825]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9110 fps step and policy inference: 7074 fps total: 5314 epoch: 132/500\n",
+ "saving next best rewards: [18.248512]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 11077 fps step and policy inference: 8658 fps total: 6168 epoch: 133/500\n",
+ "saving next best rewards: [18.266026]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9061 fps step and policy inference: 6928 fps total: 5237 epoch: 134/500\n",
+ "saving next best rewards: [18.300531]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9847 fps step and policy inference: 7706 fps total: 5670 epoch: 135/500\n",
+ "saving next best rewards: [18.341106]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9361 fps step and policy inference: 7363 fps total: 5483 epoch: 136/500\n",
+ "saving next best rewards: [18.422405]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 10995 fps step and policy inference: 8631 fps total: 6157 epoch: 137/500\n",
+ "saving next best rewards: [18.43818]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 7884 fps step and policy inference: 6049 fps total: 4721 epoch: 138/500\n",
+ "saving next best rewards: [18.514875]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9436 fps step and policy inference: 7247 fps total: 5417 epoch: 139/500\n",
+ "saving next best rewards: [18.558983]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8176 fps step and policy inference: 6304 fps total: 4868 epoch: 140/500\n",
+ "saving next best rewards: [18.633701]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8744 fps step and policy inference: 6844 fps total: 5195 epoch: 141/500\n",
+ "saving next best rewards: [18.684343]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8721 fps step and policy inference: 6736 fps total: 5128 epoch: 142/500\n",
+ "saving next best rewards: [18.710203]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 7351 fps step and policy inference: 5633 fps total: 4460 epoch: 143/500\n",
+ "saving next best rewards: [18.771118]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8528 fps step and policy inference: 6561 fps total: 5023 epoch: 144/500\n",
+ "saving next best rewards: [18.797619]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8524 fps step and policy inference: 6670 fps total: 5088 epoch: 145/500\n",
+ "saving next best rewards: [18.80793]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9660 fps step and policy inference: 7600 fps total: 5610 epoch: 146/500\n",
+ "saving next best rewards: [18.8549]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 10568 fps step and policy inference: 8243 fps total: 5959 epoch: 147/500\n",
+ "saving next best rewards: [18.85635]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9410 fps step and policy inference: 7404 fps total: 5498 epoch: 148/500\n",
+ "saving next best rewards: [18.89416]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9399 fps step and policy inference: 7303 fps total: 5441 epoch: 149/500\n",
+ "fps step: 10094 fps step and policy inference: 7898 fps total: 5774 epoch: 150/500\n",
+ "fps step: 8099 fps step and policy inference: 6183 fps total: 4801 epoch: 151/500\n",
+ "saving next best rewards: [18.924257]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8548 fps step and policy inference: 6640 fps total: 5067 epoch: 152/500\n",
+ "saving next best rewards: [18.977512]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9955 fps step and policy inference: 7725 fps total: 5677 epoch: 153/500\n",
+ "saving next best rewards: [18.977736]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8966 fps step and policy inference: 6983 fps total: 5262 epoch: 154/500\n",
+ "saving next best rewards: [19.017935]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8207 fps step and policy inference: 6414 fps total: 4934 epoch: 155/500\n",
+ "saving next best rewards: [19.05534]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8936 fps step and policy inference: 6991 fps total: 5272 epoch: 156/500\n",
+ "saving next best rewards: [19.089886]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8777 fps step and policy inference: 6811 fps total: 5169 epoch: 157/500\n",
+ "saving next best rewards: [19.134487]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9019 fps step and policy inference: 6987 fps total: 5270 epoch: 158/500\n",
+ "saving next best rewards: [19.158789]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8764 fps step and policy inference: 6667 fps total: 5086 epoch: 159/500\n",
+ "saving next best rewards: [19.165627]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9951 fps step and policy inference: 7775 fps total: 5702 epoch: 160/500\n",
+ "saving next best rewards: [19.169094]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8993 fps step and policy inference: 6962 fps total: 5259 epoch: 161/500\n",
+ "fps step: 9070 fps step and policy inference: 7091 fps total: 5328 epoch: 162/500\n",
+ "saving next best rewards: [19.184063]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9192 fps step and policy inference: 7157 fps total: 5363 epoch: 163/500\n",
+ "saving next best rewards: [19.194647]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8663 fps step and policy inference: 6611 fps total: 5053 epoch: 164/500\n",
+ "saving next best rewards: [19.208666]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9517 fps step and policy inference: 7272 fps total: 5428 epoch: 165/500\n",
+ "saving next best rewards: [19.232168]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9693 fps step and policy inference: 7577 fps total: 5601 epoch: 166/500\n",
+ "saving next best rewards: [19.24507]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9494 fps step and policy inference: 7440 fps total: 5525 epoch: 167/500\n",
+ "fps step: 8207 fps step and policy inference: 6484 fps total: 4975 epoch: 168/500\n",
+ "fps step: 9580 fps step and policy inference: 7704 fps total: 5667 epoch: 169/500\n",
+ "saving next best rewards: [19.268833]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8058 fps step and policy inference: 6160 fps total: 4784 epoch: 170/500\n",
+ "saving next best rewards: [19.296295]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8166 fps step and policy inference: 6303 fps total: 4871 epoch: 171/500\n",
+ "saving next best rewards: [19.298754]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8982 fps step and policy inference: 6908 fps total: 5222 epoch: 172/500\n",
+ "saving next best rewards: [19.302809]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9467 fps step and policy inference: 7279 fps total: 5435 epoch: 173/500\n",
+ "saving next best rewards: [19.30978]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9246 fps step and policy inference: 7169 fps total: 5369 epoch: 174/500\n",
+ "saving next best rewards: [19.326975]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 10076 fps step and policy inference: 7859 fps total: 5753 epoch: 175/500\n",
+ "fps step: 9859 fps step and policy inference: 7744 fps total: 5691 epoch: 176/500\n",
+ "saving next best rewards: [19.337345]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 7263 fps step and policy inference: 5574 fps total: 4422 epoch: 177/500\n",
+ "saving next best rewards: [19.387865]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9523 fps step and policy inference: 7473 fps total: 5542 epoch: 178/500\n",
+ "saving next best rewards: [19.40414]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 10390 fps step and policy inference: 7931 fps total: 5789 epoch: 179/500\n",
+ "fps step: 8643 fps step and policy inference: 6564 fps total: 5018 epoch: 180/500\n",
+ "saving next best rewards: [19.411938]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8519 fps step and policy inference: 6466 fps total: 4967 epoch: 181/500\n",
+ "saving next best rewards: [19.42364]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8613 fps step and policy inference: 6725 fps total: 5120 epoch: 182/500\n",
+ "saving next best rewards: [19.43238]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9221 fps step and policy inference: 6992 fps total: 5274 epoch: 183/500\n",
+ "fps step: 9661 fps step and policy inference: 7591 fps total: 5605 epoch: 184/500\n",
+ "saving next best rewards: [19.434416]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9686 fps step and policy inference: 7670 fps total: 5648 epoch: 185/500\n",
+ "fps step: 7513 fps step and policy inference: 5723 fps total: 4518 epoch: 186/500\n",
+ "fps step: 9274 fps step and policy inference: 7365 fps total: 5484 epoch: 187/500\n",
+ "fps step: 10389 fps step and policy inference: 8195 fps total: 5927 epoch: 188/500\n",
+ "saving next best rewards: [19.442959]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8425 fps step and policy inference: 6418 fps total: 4941 epoch: 189/500\n",
+ "fps step: 9669 fps step and policy inference: 7565 fps total: 5592 epoch: 190/500\n",
+ "fps step: 9014 fps step and policy inference: 7024 fps total: 5292 epoch: 191/500\n",
+ "saving next best rewards: [19.446276]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8926 fps step and policy inference: 7021 fps total: 5288 epoch: 192/500\n",
+ "fps step: 8973 fps step and policy inference: 6995 fps total: 5275 epoch: 193/500\n",
+ "fps step: 10528 fps step and policy inference: 8326 fps total: 6001 epoch: 194/500\n",
+ "fps step: 10035 fps step and policy inference: 7859 fps total: 5744 epoch: 195/500\n",
+ "fps step: 10435 fps step and policy inference: 8334 fps total: 6003 epoch: 196/500\n",
+ "fps step: 11058 fps step and policy inference: 8694 fps total: 6182 epoch: 197/500\n",
+ "fps step: 8768 fps step and policy inference: 6897 fps total: 5215 epoch: 198/500\n",
+ "fps step: 8938 fps step and policy inference: 6974 fps total: 5265 epoch: 199/500\n",
+ "fps step: 9062 fps step and policy inference: 7196 fps total: 5383 epoch: 200/500\n",
+ "fps step: 8552 fps step and policy inference: 6588 fps total: 5041 epoch: 201/500\n",
+ "fps step: 8479 fps step and policy inference: 6840 fps total: 5188 epoch: 202/500\n",
+ "fps step: 9859 fps step and policy inference: 7567 fps total: 5594 epoch: 203/500\n",
+ "fps step: 10622 fps step and policy inference: 8316 fps total: 5994 epoch: 204/500\n",
+ "fps step: 9479 fps step and policy inference: 7558 fps total: 5588 epoch: 205/500\n",
+ "fps step: 8119 fps step and policy inference: 6379 fps total: 4917 epoch: 206/500\n",
+ "saving next best rewards: [19.467575]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8336 fps step and policy inference: 6555 fps total: 5020 epoch: 207/500\n",
+ "fps step: 9968 fps step and policy inference: 7796 fps total: 5711 epoch: 208/500\n",
+ "fps step: 8218 fps step and policy inference: 6257 fps total: 4841 epoch: 209/500\n",
+ "fps step: 9383 fps step and policy inference: 7348 fps total: 5472 epoch: 210/500\n",
+ "fps step: 10277 fps step and policy inference: 8035 fps total: 5843 epoch: 211/500\n",
+ "fps step: 9323 fps step and policy inference: 7274 fps total: 5425 epoch: 212/500\n",
+ "fps step: 9817 fps step and policy inference: 7792 fps total: 5712 epoch: 213/500\n",
+ "fps step: 8785 fps step and policy inference: 7013 fps total: 5282 epoch: 214/500\n",
+ "saving next best rewards: [19.483074]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9717 fps step and policy inference: 7670 fps total: 5650 epoch: 215/500\n",
+ "saving next best rewards: [19.508457]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8558 fps step and policy inference: 6772 fps total: 5144 epoch: 216/500\n",
+ "fps step: 9336 fps step and policy inference: 7173 fps total: 5374 epoch: 217/500\n",
+ "fps step: 10327 fps step and policy inference: 8188 fps total: 5916 epoch: 218/500\n",
+ "saving next best rewards: [19.522772]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8374 fps step and policy inference: 6444 fps total: 4955 epoch: 219/500\n",
+ "saving next best rewards: [19.536459]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9765 fps step and policy inference: 7666 fps total: 5647 epoch: 220/500\n",
+ "saving next best rewards: [19.550224]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 10088 fps step and policy inference: 7869 fps total: 5748 epoch: 221/500\n",
+ "saving next best rewards: [19.554722]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8023 fps step and policy inference: 6144 fps total: 4776 epoch: 222/500\n",
+ "fps step: 7912 fps step and policy inference: 6072 fps total: 4732 epoch: 223/500\n",
+ "fps step: 9776 fps step and policy inference: 7747 fps total: 5692 epoch: 224/500\n",
+ "fps step: 9531 fps step and policy inference: 7503 fps total: 5559 epoch: 225/500\n",
+ "fps step: 11245 fps step and policy inference: 8890 fps total: 6295 epoch: 226/500\n",
+ "fps step: 8614 fps step and policy inference: 6837 fps total: 5179 epoch: 227/500\n",
+ "saving next best rewards: [19.56048]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8639 fps step and policy inference: 6623 fps total: 5060 epoch: 228/500\n",
+ "fps step: 10255 fps step and policy inference: 8152 fps total: 5904 epoch: 229/500\n",
+ "saving next best rewards: [19.578505]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9311 fps step and policy inference: 7366 fps total: 5484 epoch: 230/500\n",
+ "saving next best rewards: [19.587627]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 10272 fps step and policy inference: 7938 fps total: 5791 epoch: 231/500\n",
+ "saving next best rewards: [19.591751]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9580 fps step and policy inference: 7644 fps total: 5634 epoch: 232/500\n",
+ "saving next best rewards: [19.605936]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9287 fps step and policy inference: 7137 fps total: 5355 epoch: 233/500\n",
+ "saving next best rewards: [19.607738]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 10730 fps step and policy inference: 8341 fps total: 6004 epoch: 234/500\n",
+ "saving next best rewards: [19.61166]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9623 fps step and policy inference: 7444 fps total: 5524 epoch: 235/500\n",
+ "saving next best rewards: [19.617258]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8015 fps step and policy inference: 6184 fps total: 4795 epoch: 236/500\n",
+ "saving next best rewards: [19.617605]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9876 fps step and policy inference: 7692 fps total: 5656 epoch: 237/500\n",
+ "saving next best rewards: [19.628962]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9867 fps step and policy inference: 7549 fps total: 5575 epoch: 238/500\n",
+ "saving next best rewards: [19.636345]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9527 fps step and policy inference: 7438 fps total: 5528 epoch: 239/500\n",
+ "fps step: 9985 fps step and policy inference: 7994 fps total: 5823 epoch: 240/500\n",
+ "fps step: 8558 fps step and policy inference: 6764 fps total: 5137 epoch: 241/500\n",
+ "fps step: 8772 fps step and policy inference: 6984 fps total: 5267 epoch: 242/500\n",
+ "fps step: 8608 fps step and policy inference: 6675 fps total: 5089 epoch: 243/500\n",
+ "fps step: 9347 fps step and policy inference: 7393 fps total: 5494 epoch: 244/500\n",
+ "fps step: 10408 fps step and policy inference: 8108 fps total: 5883 epoch: 245/500\n",
+ "fps step: 9886 fps step and policy inference: 7787 fps total: 5714 epoch: 246/500\n",
+ "fps step: 9687 fps step and policy inference: 7622 fps total: 5624 epoch: 247/500\n",
+ "fps step: 10836 fps step and policy inference: 8430 fps total: 6054 epoch: 248/500\n",
+ "fps step: 8733 fps step and policy inference: 6692 fps total: 5099 epoch: 249/500\n",
+ "fps step: 8717 fps step and policy inference: 6969 fps total: 5259 epoch: 250/500\n",
+ "fps step: 8695 fps step and policy inference: 6664 fps total: 5077 epoch: 251/500\n",
+ "fps step: 9096 fps step and policy inference: 7127 fps total: 5350 epoch: 252/500\n",
+ "fps step: 9969 fps step and policy inference: 7684 fps total: 5652 epoch: 253/500\n",
+ "fps step: 10101 fps step and policy inference: 7967 fps total: 5799 epoch: 254/500\n",
+ "fps step: 10228 fps step and policy inference: 8123 fps total: 5892 epoch: 255/500\n",
+ "fps step: 9103 fps step and policy inference: 7321 fps total: 5457 epoch: 256/500\n",
+ "fps step: 9583 fps step and policy inference: 7686 fps total: 5661 epoch: 257/500\n",
+ "fps step: 8855 fps step and policy inference: 7073 fps total: 5315 epoch: 258/500\n",
+ "fps step: 8984 fps step and policy inference: 7099 fps total: 5335 epoch: 259/500\n",
+ "fps step: 9831 fps step and policy inference: 7635 fps total: 5632 epoch: 260/500\n",
+ "fps step: 8767 fps step and policy inference: 6868 fps total: 5200 epoch: 261/500\n",
+ "fps step: 9661 fps step and policy inference: 7428 fps total: 5521 epoch: 262/500\n",
+ "fps step: 10817 fps step and policy inference: 8611 fps total: 6143 epoch: 263/500\n",
+ "fps step: 9661 fps step and policy inference: 7580 fps total: 5600 epoch: 264/500\n",
+ "fps step: 10069 fps step and policy inference: 8089 fps total: 5875 epoch: 265/500\n",
+ "fps step: 9988 fps step and policy inference: 7651 fps total: 5632 epoch: 266/500\n",
+ "fps step: 9342 fps step and policy inference: 7323 fps total: 5459 epoch: 267/500\n",
+ "fps step: 8443 fps step and policy inference: 6704 fps total: 5105 epoch: 268/500\n",
+ "fps step: 9867 fps step and policy inference: 7620 fps total: 5617 epoch: 269/500\n",
+ "fps step: 8991 fps step and policy inference: 7107 fps total: 5338 epoch: 270/500\n",
+ "fps step: 8308 fps step and policy inference: 6379 fps total: 4916 epoch: 271/500\n",
+ "fps step: 8884 fps step and policy inference: 7002 fps total: 5282 epoch: 272/500\n",
+ "fps step: 10481 fps step and policy inference: 8171 fps total: 5916 epoch: 273/500\n",
+ "fps step: 9349 fps step and policy inference: 7511 fps total: 5557 epoch: 274/500\n",
+ "fps step: 8598 fps step and policy inference: 6644 fps total: 5072 epoch: 275/500\n",
+ "fps step: 10002 fps step and policy inference: 7847 fps total: 5745 epoch: 276/500\n",
+ "fps step: 9773 fps step and policy inference: 7588 fps total: 5609 epoch: 277/500\n",
+ "fps step: 10620 fps step and policy inference: 8421 fps total: 6045 epoch: 278/500\n",
+ "fps step: 10199 fps step and policy inference: 7942 fps total: 5799 epoch: 279/500\n",
+ "fps step: 7878 fps step and policy inference: 5983 fps total: 4677 epoch: 280/500\n",
+ "fps step: 8014 fps step and policy inference: 6084 fps total: 4732 epoch: 281/500\n",
+ "fps step: 8508 fps step and policy inference: 6554 fps total: 5021 epoch: 282/500\n",
+ "fps step: 10028 fps step and policy inference: 7769 fps total: 5700 epoch: 283/500\n",
+ "fps step: 8880 fps step and policy inference: 7013 fps total: 5283 epoch: 284/500\n",
+ "fps step: 9593 fps step and policy inference: 7601 fps total: 5612 epoch: 285/500\n",
+ "fps step: 8361 fps step and policy inference: 6442 fps total: 4955 epoch: 286/500\n",
+ "fps step: 10058 fps step and policy inference: 7877 fps total: 5762 epoch: 287/500\n",
+ "fps step: 9509 fps step and policy inference: 7606 fps total: 5612 epoch: 288/500\n",
+ "fps step: 9549 fps step and policy inference: 7587 fps total: 5604 epoch: 289/500\n",
+ "fps step: 10914 fps step and policy inference: 8646 fps total: 6159 epoch: 290/500\n",
+ "fps step: 7823 fps step and policy inference: 5951 fps total: 4658 epoch: 291/500\n",
+ "fps step: 8512 fps step and policy inference: 6535 fps total: 5007 epoch: 292/500\n",
+ "fps step: 10715 fps step and policy inference: 8424 fps total: 6053 epoch: 293/500\n",
+ "fps step: 9755 fps step and policy inference: 7589 fps total: 5606 epoch: 294/500\n",
+ "fps step: 9484 fps step and policy inference: 7522 fps total: 5568 epoch: 295/500\n",
+ "fps step: 8462 fps step and policy inference: 6432 fps total: 4951 epoch: 296/500\n",
+ "fps step: 10158 fps step and policy inference: 8129 fps total: 5893 epoch: 297/500\n",
+ "fps step: 10065 fps step and policy inference: 7917 fps total: 5770 epoch: 298/500\n",
+ "fps step: 8776 fps step and policy inference: 6764 fps total: 5141 epoch: 299/500\n",
+ "fps step: 9717 fps step and policy inference: 7524 fps total: 5569 epoch: 300/500\n",
+ "fps step: 8312 fps step and policy inference: 6452 fps total: 4960 epoch: 301/500\n",
+ "fps step: 9235 fps step and policy inference: 7296 fps total: 5443 epoch: 302/500\n",
+ "fps step: 8778 fps step and policy inference: 6927 fps total: 5236 epoch: 303/500\n",
+ "fps step: 10154 fps step and policy inference: 8041 fps total: 5849 epoch: 304/500\n",
+ "fps step: 7919 fps step and policy inference: 6273 fps total: 4843 epoch: 305/500\n",
+ "fps step: 9274 fps step and policy inference: 7190 fps total: 5383 epoch: 306/500\n",
+ "fps step: 9363 fps step and policy inference: 7346 fps total: 5471 epoch: 307/500\n",
+ "fps step: 9347 fps step and policy inference: 7365 fps total: 5481 epoch: 308/500\n",
+ "fps step: 10251 fps step and policy inference: 8071 fps total: 5862 epoch: 309/500\n",
+ "fps step: 8079 fps step and policy inference: 6253 fps total: 4841 epoch: 310/500\n",
+ "fps step: 9708 fps step and policy inference: 7663 fps total: 5643 epoch: 311/500\n",
+ "fps step: 9461 fps step and policy inference: 7463 fps total: 5535 epoch: 312/500\n",
+ "fps step: 8929 fps step and policy inference: 6971 fps total: 5256 epoch: 313/500\n",
+ "fps step: 10609 fps step and policy inference: 8309 fps total: 5988 epoch: 314/500\n",
+ "fps step: 8276 fps step and policy inference: 6336 fps total: 4888 epoch: 315/500\n",
+ "fps step: 10481 fps step and policy inference: 8323 fps total: 5994 epoch: 316/500\n",
+ "saving next best rewards: [19.644829]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9005 fps step and policy inference: 6899 fps total: 5218 epoch: 317/500\n",
+ "saving next best rewards: [19.66227]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9450 fps step and policy inference: 7476 fps total: 5543 epoch: 318/500\n",
+ "saving next best rewards: [19.682032]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9613 fps step and policy inference: 7590 fps total: 5604 epoch: 319/500\n",
+ "saving next best rewards: [19.700638]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9262 fps step and policy inference: 7052 fps total: 5307 epoch: 320/500\n",
+ "saving next best rewards: [19.709528]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9928 fps step and policy inference: 7652 fps total: 5642 epoch: 321/500\n",
+ "saving next best rewards: [19.71531]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 7503 fps step and policy inference: 5779 fps total: 4552 epoch: 322/500\n",
+ "saving next best rewards: [19.72733]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8804 fps step and policy inference: 6715 fps total: 5111 epoch: 323/500\n",
+ "fps step: 10129 fps step and policy inference: 7934 fps total: 5790 epoch: 324/500\n",
+ "fps step: 8016 fps step and policy inference: 6175 fps total: 4794 epoch: 325/500\n",
+ "fps step: 9712 fps step and policy inference: 7605 fps total: 5613 epoch: 326/500\n",
+ "fps step: 9359 fps step and policy inference: 7426 fps total: 5511 epoch: 327/500\n",
+ "fps step: 10866 fps step and policy inference: 8389 fps total: 6028 epoch: 328/500\n",
+ "fps step: 9498 fps step and policy inference: 7354 fps total: 5475 epoch: 329/500\n",
+ "fps step: 7873 fps step and policy inference: 6051 fps total: 4714 epoch: 330/500\n",
+ "fps step: 8250 fps step and policy inference: 6424 fps total: 4939 epoch: 331/500\n",
+ "fps step: 9366 fps step and policy inference: 7266 fps total: 5429 epoch: 332/500\n",
+ "fps step: 8507 fps step and policy inference: 6638 fps total: 5069 epoch: 333/500\n",
+ "fps step: 9623 fps step and policy inference: 7587 fps total: 5604 epoch: 334/500\n",
+ "fps step: 8756 fps step and policy inference: 6813 fps total: 5169 epoch: 335/500\n",
+ "saving next best rewards: [19.72744]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 8868 fps step and policy inference: 6858 fps total: 5197 epoch: 336/500\n",
+ "fps step: 9576 fps step and policy inference: 7487 fps total: 5549 epoch: 337/500\n",
+ "fps step: 9786 fps step and policy inference: 7745 fps total: 5684 epoch: 338/500\n",
+ "saving next best rewards: [19.733776]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9133 fps step and policy inference: 7178 fps total: 5381 epoch: 339/500\n",
+ "fps step: 8417 fps step and policy inference: 6588 fps total: 5041 epoch: 340/500\n",
+ "fps step: 9032 fps step and policy inference: 7074 fps total: 5315 epoch: 341/500\n",
+ "saving next best rewards: [19.735119]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9990 fps step and policy inference: 7756 fps total: 5697 epoch: 342/500\n",
+ "fps step: 9977 fps step and policy inference: 7903 fps total: 5773 epoch: 343/500\n",
+ "saving next best rewards: [19.736557]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9047 fps step and policy inference: 7128 fps total: 5359 epoch: 344/500\n",
+ "saving next best rewards: [19.739565]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 10085 fps step and policy inference: 7781 fps total: 5707 epoch: 345/500\n",
+ "saving next best rewards: [19.7473]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9674 fps step and policy inference: 7620 fps total: 5623 epoch: 346/500\n",
+ "fps step: 10482 fps step and policy inference: 8313 fps total: 5988 epoch: 347/500\n",
+ "fps step: 9529 fps step and policy inference: 7641 fps total: 5630 epoch: 348/500\n",
+ "fps step: 9467 fps step and policy inference: 7482 fps total: 5545 epoch: 349/500\n",
+ "fps step: 9184 fps step and policy inference: 7351 fps total: 5473 epoch: 350/500\n",
+ "saving next best rewards: [19.757431]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "fps step: 9473 fps step and policy inference: 7221 fps total: 5402 epoch: 351/500\n",
+ "fps step: 9017 fps step and policy inference: 7019 fps total: 5286 epoch: 352/500\n",
+ "fps step: 9684 fps step and policy inference: 7601 fps total: 5605 epoch: 353/500\n",
+ "fps step: 8249 fps step and policy inference: 6342 fps total: 4893 epoch: 354/500\n",
+ "fps step: 9634 fps step and policy inference: 7563 fps total: 5591 epoch: 355/500\n",
+ "fps step: 9181 fps step and policy inference: 7364 fps total: 5479 epoch: 356/500\n",
+ "fps step: 9096 fps step and policy inference: 7083 fps total: 5323 epoch: 357/500\n",
+ "fps step: 9185 fps step and policy inference: 7352 fps total: 5475 epoch: 358/500\n",
+ "fps step: 9205 fps step and policy inference: 7222 fps total: 5399 epoch: 359/500\n",
+ "fps step: 9157 fps step and policy inference: 7214 fps total: 5397 epoch: 360/500\n",
+ "fps step: 9129 fps step and policy inference: 7066 fps total: 5306 epoch: 361/500\n",
+ "fps step: 8610 fps step and policy inference: 6718 fps total: 5114 epoch: 362/500\n",
+ "fps step: 9078 fps step and policy inference: 7023 fps total: 5285 epoch: 363/500\n",
+ "fps step: 9982 fps step and policy inference: 7789 fps total: 5712 epoch: 364/500\n",
+ "fps step: 9125 fps step and policy inference: 7026 fps total: 5288 epoch: 365/500\n",
+ "fps step: 9322 fps step and policy inference: 7389 fps total: 5493 epoch: 366/500\n",
+ "fps step: 9733 fps step and policy inference: 7591 fps total: 5606 epoch: 367/500\n",
+ "fps step: 10021 fps step and policy inference: 7770 fps total: 5695 epoch: 368/500\n",
+ "fps step: 10056 fps step and policy inference: 8063 fps total: 5858 epoch: 369/500\n",
+ "fps step: 10183 fps step and policy inference: 7984 fps total: 5809 epoch: 370/500\n",
+ "fps step: 8693 fps step and policy inference: 6934 fps total: 5241 epoch: 371/500\n",
+ "fps step: 9813 fps step and policy inference: 7764 fps total: 5700 epoch: 372/500\n",
+ "fps step: 8428 fps step and policy inference: 6586 fps total: 5040 epoch: 373/500\n",
+ "fps step: 9225 fps step and policy inference: 7228 fps total: 5402 epoch: 374/500\n",
+ "fps step: 8938 fps step and policy inference: 6849 fps total: 5190 epoch: 375/500\n",
+ "fps step: 9360 fps step and policy inference: 7202 fps total: 5387 epoch: 376/500\n",
+ "fps step: 10249 fps step and policy inference: 8137 fps total: 5890 epoch: 377/500\n",
+ "fps step: 10365 fps step and policy inference: 8058 fps total: 5853 epoch: 378/500\n",
+ "fps step: 8881 fps step and policy inference: 6907 fps total: 5223 epoch: 379/500\n",
+ "fps step: 8890 fps step and policy inference: 6923 fps total: 5228 epoch: 380/500\n",
+ "fps step: 9036 fps step and policy inference: 7172 fps total: 5372 epoch: 381/500\n",
+ "fps step: 10266 fps step and policy inference: 7892 fps total: 5768 epoch: 382/500\n",
+ "fps step: 7861 fps step and policy inference: 5971 fps total: 4668 epoch: 383/500\n",
+ "fps step: 10274 fps step and policy inference: 7954 fps total: 5796 epoch: 384/500\n",
+ "fps step: 9462 fps step and policy inference: 7491 fps total: 5552 epoch: 385/500\n",
+ "fps step: 9322 fps step and policy inference: 7290 fps total: 5439 epoch: 386/500\n",
+ "fps step: 8802 fps step and policy inference: 6833 fps total: 5183 epoch: 387/500\n",
+ "fps step: 8684 fps step and policy inference: 6771 fps total: 5146 epoch: 388/500\n",
+ "fps step: 9851 fps step and policy inference: 7789 fps total: 5713 epoch: 389/500\n",
+ "fps step: 9188 fps step and policy inference: 7134 fps total: 5350 epoch: 390/500\n",
+ "fps step: 10023 fps step and policy inference: 7921 fps total: 5783 epoch: 391/500\n",
+ "fps step: 9423 fps step and policy inference: 7388 fps total: 5488 epoch: 392/500\n",
+ "fps step: 9079 fps step and policy inference: 7128 fps total: 5345 epoch: 393/500\n",
+ "fps step: 10581 fps step and policy inference: 8278 fps total: 5964 epoch: 394/500\n",
+ "fps step: 9424 fps step and policy inference: 7269 fps total: 5426 epoch: 395/500\n",
+ "fps step: 8442 fps step and policy inference: 6482 fps total: 4977 epoch: 396/500\n",
+ "fps step: 8205 fps step and policy inference: 6439 fps total: 4951 epoch: 397/500\n",
+ "fps step: 11269 fps step and policy inference: 8894 fps total: 6284 epoch: 398/500\n",
+ "fps step: 9729 fps step and policy inference: 7552 fps total: 5586 epoch: 399/500\n",
+ "fps step: 7201 fps step and policy inference: 5514 fps total: 4385 epoch: 400/500\n",
+ "fps step: 10137 fps step and policy inference: 8104 fps total: 5873 epoch: 401/500\n",
+ "fps step: 7655 fps step and policy inference: 5755 fps total: 4537 epoch: 402/500\n",
+ "fps step: 7720 fps step and policy inference: 5969 fps total: 4669 epoch: 403/500\n",
+ "fps step: 9109 fps step and policy inference: 7186 fps total: 5381 epoch: 404/500\n",
+ "fps step: 9183 fps step and policy inference: 7040 fps total: 5303 epoch: 405/500\n",
+ "fps step: 8864 fps step and policy inference: 6824 fps total: 5177 epoch: 406/500\n",
+ "fps step: 9841 fps step and policy inference: 7624 fps total: 5621 epoch: 407/500\n",
+ "fps step: 9575 fps step and policy inference: 7304 fps total: 5443 epoch: 408/500\n",
+ "fps step: 9316 fps step and policy inference: 7307 fps total: 5448 epoch: 409/500\n",
+ "fps step: 9275 fps step and policy inference: 7266 fps total: 5428 epoch: 410/500\n",
+ "fps step: 9866 fps step and policy inference: 7874 fps total: 5761 epoch: 411/500\n",
+ "fps step: 9724 fps step and policy inference: 7604 fps total: 5609 epoch: 412/500\n",
+ "fps step: 9952 fps step and policy inference: 7872 fps total: 5750 epoch: 413/500\n",
+ "fps step: 8739 fps step and policy inference: 6743 fps total: 5130 epoch: 414/500\n",
+ "fps step: 10820 fps step and policy inference: 8493 fps total: 6084 epoch: 415/500\n",
+ "fps step: 9476 fps step and policy inference: 7542 fps total: 5576 epoch: 416/500\n",
+ "fps step: 9158 fps step and policy inference: 7168 fps total: 5371 epoch: 417/500\n",
+ "fps step: 8887 fps step and policy inference: 7077 fps total: 5319 epoch: 418/500\n",
+ "fps step: 9262 fps step and policy inference: 7242 fps total: 5408 epoch: 419/500\n",
+ "fps step: 10031 fps step and policy inference: 7874 fps total: 5759 epoch: 420/500\n",
+ "fps step: 9187 fps step and policy inference: 7105 fps total: 5336 epoch: 421/500\n",
+ "fps step: 9942 fps step and policy inference: 7723 fps total: 5675 epoch: 422/500\n",
+ "fps step: 9981 fps step and policy inference: 7808 fps total: 5720 epoch: 423/500\n",
+ "fps step: 9955 fps step and policy inference: 7818 fps total: 5723 epoch: 424/500\n",
+ "fps step: 8182 fps step and policy inference: 6408 fps total: 4934 epoch: 425/500\n",
+ "fps step: 8547 fps step and policy inference: 6618 fps total: 5053 epoch: 426/500\n",
+ "fps step: 9124 fps step and policy inference: 7196 fps total: 5388 epoch: 427/500\n",
+ "fps step: 8625 fps step and policy inference: 6674 fps total: 5089 epoch: 428/500\n",
+ "fps step: 9528 fps step and policy inference: 7435 fps total: 5512 epoch: 429/500\n",
+ "fps step: 8211 fps step and policy inference: 6358 fps total: 4903 epoch: 430/500\n",
+ "fps step: 8504 fps step and policy inference: 6663 fps total: 5083 epoch: 431/500\n",
+ "fps step: 10131 fps step and policy inference: 8095 fps total: 5877 epoch: 432/500\n",
+ "fps step: 8699 fps step and policy inference: 6728 fps total: 5120 epoch: 433/500\n",
+ "fps step: 8860 fps step and policy inference: 6886 fps total: 5211 epoch: 434/500\n",
+ "fps step: 8941 fps step and policy inference: 6954 fps total: 5250 epoch: 435/500\n",
+ "fps step: 8965 fps step and policy inference: 6825 fps total: 5175 epoch: 436/500\n",
+ "fps step: 9038 fps step and policy inference: 6970 fps total: 5260 epoch: 437/500\n",
+ "fps step: 10194 fps step and policy inference: 8022 fps total: 5837 epoch: 438/500\n",
+ "fps step: 9488 fps step and policy inference: 7505 fps total: 5551 epoch: 439/500\n",
+ "fps step: 10083 fps step and policy inference: 7849 fps total: 5745 epoch: 440/500\n",
+ "fps step: 9214 fps step and policy inference: 7201 fps total: 5390 epoch: 441/500\n",
+ "fps step: 10473 fps step and policy inference: 8341 fps total: 6004 epoch: 442/500\n",
+ "fps step: 9396 fps step and policy inference: 7395 fps total: 5499 epoch: 443/500\n",
+ "fps step: 10055 fps step and policy inference: 8069 fps total: 5862 epoch: 444/500\n",
+ "fps step: 8934 fps step and policy inference: 6996 fps total: 5273 epoch: 445/500\n",
+ "fps step: 8951 fps step and policy inference: 7031 fps total: 5292 epoch: 446/500\n",
+ "fps step: 7595 fps step and policy inference: 5877 fps total: 4609 epoch: 447/500\n",
+ "fps step: 9329 fps step and policy inference: 7407 fps total: 5505 epoch: 448/500\n",
+ "fps step: 9257 fps step and policy inference: 7228 fps total: 5406 epoch: 449/500\n",
+ "fps step: 8829 fps step and policy inference: 6824 fps total: 5172 epoch: 450/500\n",
+ "fps step: 10002 fps step and policy inference: 7821 fps total: 5731 epoch: 451/500\n",
+ "fps step: 9345 fps step and policy inference: 7353 fps total: 5475 epoch: 452/500\n",
+ "fps step: 9631 fps step and policy inference: 7628 fps total: 5623 epoch: 453/500\n",
+ "fps step: 9948 fps step and policy inference: 7781 fps total: 5709 epoch: 454/500\n",
+ "fps step: 8673 fps step and policy inference: 6715 fps total: 5112 epoch: 455/500\n",
+ "fps step: 11702 fps step and policy inference: 9220 fps total: 6445 epoch: 456/500\n",
+ "fps step: 9455 fps step and policy inference: 7429 fps total: 5513 epoch: 457/500\n",
+ "fps step: 8438 fps step and policy inference: 6536 fps total: 5008 epoch: 458/500\n",
+ "fps step: 9250 fps step and policy inference: 7240 fps total: 5412 epoch: 459/500\n",
+ "fps step: 9140 fps step and policy inference: 7105 fps total: 5327 epoch: 460/500\n",
+ "fps step: 9233 fps step and policy inference: 7256 fps total: 5431 epoch: 461/500\n",
+ "fps step: 7204 fps step and policy inference: 5524 fps total: 4394 epoch: 462/500\n",
+ "fps step: 9044 fps step and policy inference: 7245 fps total: 5413 epoch: 463/500\n",
+ "fps step: 9307 fps step and policy inference: 7200 fps total: 5381 epoch: 464/500\n",
+ "fps step: 11346 fps step and policy inference: 8914 fps total: 6295 epoch: 465/500\n",
+ "fps step: 9451 fps step and policy inference: 7264 fps total: 5426 epoch: 466/500\n",
+ "fps step: 8665 fps step and policy inference: 6620 fps total: 5061 epoch: 467/500\n",
+ "fps step: 7801 fps step and policy inference: 5960 fps total: 4663 epoch: 468/500\n",
+ "fps step: 9465 fps step and policy inference: 7485 fps total: 5551 epoch: 469/500\n",
+ "fps step: 9572 fps step and policy inference: 7412 fps total: 5506 epoch: 470/500\n",
+ "fps step: 9893 fps step and policy inference: 7773 fps total: 5707 epoch: 471/500\n",
+ "fps step: 9099 fps step and policy inference: 7167 fps total: 5368 epoch: 472/500\n",
+ "fps step: 8981 fps step and policy inference: 6827 fps total: 5174 epoch: 473/500\n",
+ "fps step: 9243 fps step and policy inference: 7059 fps total: 5309 epoch: 474/500\n",
+ "fps step: 7665 fps step and policy inference: 5901 fps total: 4627 epoch: 475/500\n",
+ "fps step: 7541 fps step and policy inference: 5954 fps total: 4661 epoch: 476/500\n",
+ "fps step: 8585 fps step and policy inference: 6646 fps total: 5072 epoch: 477/500\n",
+ "fps step: 10771 fps step and policy inference: 8422 fps total: 6035 epoch: 478/500\n",
+ "fps step: 9583 fps step and policy inference: 7447 fps total: 5527 epoch: 479/500\n",
+ "fps step: 10889 fps step and policy inference: 8614 fps total: 6146 epoch: 480/500\n",
+ "fps step: 9537 fps step and policy inference: 7374 fps total: 5479 epoch: 481/500\n",
+ "fps step: 8029 fps step and policy inference: 6193 fps total: 4802 epoch: 482/500\n",
+ "fps step: 9203 fps step and policy inference: 7209 fps total: 5392 epoch: 483/500\n",
+ "fps step: 10002 fps step and policy inference: 7792 fps total: 5714 epoch: 484/500\n",
+ "fps step: 9731 fps step and policy inference: 7482 fps total: 5546 epoch: 485/500\n",
+ "fps step: 9883 fps step and policy inference: 7762 fps total: 5698 epoch: 486/500\n",
+ "fps step: 8984 fps step and policy inference: 7122 fps total: 5340 epoch: 487/500\n",
+ "fps step: 9821 fps step and policy inference: 7826 fps total: 5733 epoch: 488/500\n",
+ "fps step: 9200 fps step and policy inference: 7139 fps total: 5352 epoch: 489/500\n",
+ "fps step: 9063 fps step and policy inference: 7226 fps total: 5398 epoch: 490/500\n",
+ "fps step: 8678 fps step and policy inference: 6683 fps total: 5095 epoch: 491/500\n",
+ "fps step: 9752 fps step and policy inference: 7756 fps total: 5695 epoch: 492/500\n",
+ "fps step: 10208 fps step and policy inference: 8090 fps total: 5870 epoch: 493/500\n",
+ "fps step: 8983 fps step and policy inference: 7010 fps total: 5281 epoch: 494/500\n",
+ "fps step: 9886 fps step and policy inference: 7772 fps total: 5698 epoch: 495/500\n",
+ "fps step: 8691 fps step and policy inference: 6727 fps total: 5121 epoch: 496/500\n",
+ "fps step: 8804 fps step and policy inference: 6815 fps total: 5173 epoch: 497/500\n",
+ "fps step: 8438 fps step and policy inference: 6575 fps total: 5033 epoch: 498/500\n",
+ "fps step: 9069 fps step and policy inference: 7043 fps total: 5299 epoch: 499/500\n",
+ "fps step: 9205 fps step and policy inference: 7004 fps total: 5274 epoch: 500/500\n",
+ "saving next best rewards: [19.764378]\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/Pong-v5_envpool.pth'\n",
+ "=> saving checkpoint 'runs/Pong-v5_envpool_23-09-07-59/nn/last_Pong-v5_envpoolep500rew[19.764378].pth'\n",
+ "MAX EPOCHS NUM!\n"
+ ]
+ }
+ ]
+ }
+ ]
+}
\ No newline at end of file