q_learning.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "1xEs450ZOUHA"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "import gym\n",
        "import random\n",
        "import time\n",
        "from IPython.display import clear_output, display\n",
        "import matplotlib.pyplot as plt\n",
        "from gym.envs.toy_text.frozen_lake import generate_random_map"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "i0gvf8uTRcQE",
        "outputId": "4e725845-ec0b-4553-c2b1-4c592bbb0666"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "[0]"
            ]
          },
          "execution_count": 2,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "random_map = generate_random_map(size=10, p=0.3)\n",
        "env = gym.make(\"FrozenLake-v1\",is_slippery=True, desc=random_map)\n",
        "env.seed(0)  # make results reproducible"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7KDR6gB8RhTE",
        "outputId": "3bfce5ed-7295-43e1-a3f3-915a07c88f7d"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "4 100\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "action_space_size = env.action_space.n\n",
        "state_space_size = env.observation_space.n\n",
        "\n",
        "print(action_space_size, state_space_size)\n",
        "plt.figure(3)\n",
        "plt.clf()\n",
        "env.reset()\n",
        "plt.imshow(env.render(\"rgb_array\"))\n",
        "time.sleep(0.3)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {},
      "outputs": [],
      "source": [
        "def plot_epsilon(epsilon, num_episodes):\n",
        "    steps = np.arange(0, num_episodes, 1)\n",
        "    plt.clf()\n",
        "    plt.plot(steps, epsilon, marker=\"o\", markersize=1)\n",
        "    plt.ylabel(\"Epsilon \")\n",
        "    plt.xlabel(\"Step\")\n",
        "    plt.tight_layout()\n",
        "    plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {},
      "outputs": [],
      "source": [
        "def plot_rewards(rewards, episodes):\n",
        "    # Plot reward per 1000 episodes\n",
        "    # Line plot to show how average reward changes over time\n",
        "    average_rewards = np.split(np.array(rewards), episodes/1000)\n",
        "    steps = 1000 * np.arange(len(average_rewards))\n",
        "    plt.clf()\n",
        "    plt.plot(steps, np.average(average_rewards, 1))\n",
        "    plt.ylabel(\"Average Reward\")\n",
        "    plt.xlabel(\"Episode\")\n",
        "    plt.tight_layout()\n",
        "    plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "# Agent not knowing about the reward at the right most corner"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {},
      "outputs": [],
      "source": [
        "total_episodes = 3000000\n",
        "learning_rate = 0.8  \n",
        "max_steps = 800\n",
        "gamma = 0.95\n",
        "epsilon = 1.0\n",
        "max_epsilon = 1.0\n",
        "min_epsilon = 0.001\n",
        "decay_rate = 0.000005\n",
        "q_table = np.zeros((state_space_size, action_space_size))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Score over time: 0.0\n",
            "[[0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]]\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Avg reward per 1000 episode \n",
            "\n",
            "1000 :  0.0\n",
            "2000 :  0.0\n",
            "3000 :  0.0\n",
            "4000 :  0.0\n",
            "5000 :  0.0\n",
            "6000 :  0.0\n",
            "7000 :  0.0\n",
            "8000 :  0.0\n",
            "9000 :  0.0\n",
            "10000 :  0.0\n",
            "11000 :  0.0\n",
            "12000 :  0.0\n",
            "13000 :  0.0\n",
            "14000 :  0.0\n",
            "15000 :  0.0\n",
            "16000 :  0.0\n",
            "17000 :  0.0\n",
            "18000 :  0.0\n",
            "19000 :  0.0\n",
            "20000 :  0.0\n",
            "21000 :  0.0\n",
            "22000 :  0.0\n",
            "23000 :  0.0\n",
            "24000 :  0.0\n",
            "25000 :  0.0\n",
            "26000 :  0.0\n",
            "27000 :  0.0\n",
            "28000 :  0.0\n",
            "29000 :  0.0\n",
            "30000 :  0.0\n",
            "31000 :  0.0\n",
            "32000 :  0.0\n",
            "33000 :  0.0\n",
            "34000 :  0.0\n",
            "35000 :  0.0\n",
            "36000 :  0.0\n",
            "37000 :  0.0\n",
            "38000 :  0.0\n",
            "39000 :  0.0\n",
            "40000 :  0.0\n",
            "41000 :  0.0\n",
            "42000 :  0.0\n",
            "43000 :  0.0\n",
            "44000 :  0.0\n",
            "45000 :  0.0\n",
            "46000 :  0.0\n",
            "47000 :  0.0\n",
            "48000 :  0.0\n",
            "49000 :  0.0\n",
            "50000 :  0.0\n",
            "51000 :  0.0\n",
            "52000 :  0.0\n",
            "53000 :  0.0\n",
            "54000 :  0.0\n",
            "55000 :  0.0\n",
            "56000 :  0.0\n",
            "57000 :  0.0\n",
            "58000 :  0.0\n",
            "59000 :  0.0\n",
            "60000 :  0.0\n",
            "61000 :  0.0\n",
            "62000 :  0.0\n",
            "63000 :  0.0\n",
            "64000 :  0.0\n",
            "65000 :  0.0\n",
            "66000 :  0.0\n",
            "67000 :  0.0\n",
            "68000 :  0.0\n",
            "69000 :  0.0\n",
            "70000 :  0.0\n",
            "71000 :  0.0\n",
            "72000 :  0.0\n",
            "73000 :  0.0\n",
            "74000 :  0.0\n",
            "75000 :  0.0\n",
            "76000 :  0.0\n",
            "77000 :  0.0\n",
            "78000 :  0.0\n",
            "79000 :  0.0\n",
            "80000 :  0.0\n",
            "81000 :  0.0\n",
            "82000 :  0.0\n",
            "83000 :  0.0\n",
            "84000 :  0.0\n",
            "85000 :  0.0\n",
            "86000 :  0.0\n",
            "87000 :  0.0\n",
            "88000 :  0.0\n",
            "89000 :  0.0\n",
            "90000 :  0.0\n",
            "91000 :  0.0\n",
            "92000 :  0.0\n",
            "93000 :  0.0\n",
            "94000 :  0.0\n",
            "95000 :  0.0\n",
            "96000 :  0.0\n",
            "97000 :  0.0\n",
            "98000 :  0.0\n",
            "99000 :  0.0\n",
            "100000 :  0.0\n",
            "101000 :  0.0\n",
            "102000 :  0.0\n",
            "103000 :  0.0\n",
            "104000 :  0.0\n",
            "105000 :  0.0\n",
            "106000 :  0.0\n",
            "107000 :  0.0\n",
            "108000 :  0.0\n",
            "109000 :  0.0\n",
            "110000 :  0.0\n",
            "111000 :  0.0\n",
            "112000 :  0.0\n",
            "113000 :  0.0\n",
            "114000 :  0.0\n",
            "115000 :  0.0\n",
            "116000 :  0.0\n",
            "117000 :  0.0\n",
            "118000 :  0.0\n",
            "119000 :  0.0\n",
            "120000 :  0.0\n",
            "121000 :  0.0\n",
            "122000 :  0.0\n",
            "123000 :  0.0\n",
            "124000 :  0.0\n",
            "125000 :  0.0\n",
            "126000 :  0.0\n",
            "127000 :  0.0\n",
            "128000 :  0.0\n",
            "129000 :  0.0\n",
            "130000 :  0.0\n",
            "131000 :  0.0\n",
            "132000 :  0.0\n",
            "133000 :  0.0\n",
            "134000 :  0.0\n",
            "135000 :  0.0\n",
            "136000 :  0.0\n",
            "137000 :  0.0\n",
            "138000 :  0.0\n",
            "139000 :  0.0\n",
            "140000 :  0.0\n",
            "141000 :  0.0\n",
            "142000 :  0.0\n",
            "143000 :  0.0\n",
            "144000 :  0.0\n",
            "145000 :  0.0\n",
            "146000 :  0.0\n",
            "147000 :  0.0\n",
            "148000 :  0.0\n",
            "149000 :  0.0\n",
            "150000 :  0.0\n",
            "151000 :  0.0\n",
            "152000 :  0.0\n",
            "153000 :  0.0\n",
            "154000 :  0.0\n",
            "155000 :  0.0\n",
            "156000 :  0.0\n",
            "157000 :  0.0\n",
            "158000 :  0.0\n",
            "159000 :  0.0\n",
            "160000 :  0.0\n",
            "161000 :  0.0\n",
            "162000 :  0.0\n",
            "163000 :  0.0\n",
            "164000 :  0.0\n",
            "165000 :  0.0\n",
            "166000 :  0.0\n",
            "167000 :  0.0\n",
            "168000 :  0.0\n",
            "169000 :  0.0\n",
            "170000 :  0.0\n",
            "171000 :  0.0\n",
            "172000 :  0.0\n",
            "173000 :  0.0\n",
            "174000 :  0.0\n",
            "175000 :  0.0\n",
            "176000 :  0.0\n",
            "177000 :  0.0\n",
            "178000 :  0.0\n",
            "179000 :  0.0\n",
            "180000 :  0.0\n",
            "181000 :  0.0\n",
            "182000 :  0.0\n",
            "183000 :  0.0\n",
            "184000 :  0.0\n",
            "185000 :  0.0\n",
            "186000 :  0.0\n",
            "187000 :  0.0\n",
            "188000 :  0.0\n",
            "189000 :  0.0\n",
            "190000 :  0.0\n",
            "191000 :  0.0\n",
            "192000 :  0.0\n",
            "193000 :  0.0\n",
            "194000 :  0.0\n",
            "195000 :  0.0\n",
            "196000 :  0.0\n",
            "197000 :  0.0\n",
            "198000 :  0.0\n",
            "199000 :  0.0\n",
            "200000 :  0.0\n",
            "201000 :  0.0\n",
            "202000 :  0.0\n",
            "203000 :  0.0\n",
            "204000 :  0.0\n",
            "205000 :  0.0\n",
            "206000 :  0.0\n",
            "207000 :  0.0\n",
            "208000 :  0.0\n",
            "209000 :  0.0\n",
            "210000 :  0.0\n",
            "211000 :  0.0\n",
            "212000 :  0.0\n",
            "213000 :  0.0\n",
            "214000 :  0.0\n",
            "215000 :  0.0\n",
            "216000 :  0.0\n",
            "217000 :  0.0\n",
            "218000 :  0.0\n",
            "219000 :  0.0\n",
            "220000 :  0.0\n",
            "221000 :  0.0\n",
            "222000 :  0.0\n",
            "223000 :  0.0\n",
            "224000 :  0.0\n",
            "225000 :  0.0\n",
            "226000 :  0.0\n",
            "227000 :  0.0\n",
            "228000 :  0.0\n",
            "229000 :  0.0\n",
            "230000 :  0.0\n",
            "231000 :  0.0\n",
            "232000 :  0.0\n",
            "233000 :  0.0\n",
            "234000 :  0.0\n",
            "235000 :  0.0\n",
            "236000 :  0.0\n",
            "237000 :  0.0\n",
            "238000 :  0.0\n",
            "239000 :  0.0\n",
            "240000 :  0.0\n",
            "241000 :  0.0\n",
            "242000 :  0.0\n",
            "243000 :  0.0\n",
            "244000 :  0.0\n",
            "245000 :  0.0\n",
            "246000 :  0.0\n",
            "247000 :  0.0\n",
            "248000 :  0.0\n",
            "249000 :  0.0\n",
            "250000 :  0.0\n",
            "251000 :  0.0\n",
            "252000 :  0.0\n",
            "253000 :  0.0\n",
            "254000 :  0.0\n",
            "255000 :  0.0\n",
            "256000 :  0.0\n",
            "257000 :  0.0\n",
            "258000 :  0.0\n",
            "259000 :  0.0\n",
            "260000 :  0.0\n",
            "261000 :  0.0\n",
            "262000 :  0.0\n",
            "263000 :  0.0\n",
            "264000 :  0.0\n",
            "265000 :  0.0\n",
            "266000 :  0.0\n",
            "267000 :  0.0\n",
            "268000 :  0.0\n",
            "269000 :  0.0\n",
            "270000 :  0.0\n",
            "271000 :  0.0\n",
            "272000 :  0.0\n",
            "273000 :  0.0\n",
            "274000 :  0.0\n",
            "275000 :  0.0\n",
            "276000 :  0.0\n",
            "277000 :  0.0\n",
            "278000 :  0.0\n",
            "279000 :  0.0\n",
            "280000 :  0.0\n",
            "281000 :  0.0\n",
            "282000 :  0.0\n",
            "283000 :  0.0\n",
            "284000 :  0.0\n",
            "285000 :  0.0\n",
            "286000 :  0.0\n",
            "287000 :  0.0\n",
            "288000 :  0.0\n",
            "289000 :  0.0\n",
            "290000 :  0.0\n",
            "291000 :  0.0\n",
            "292000 :  0.0\n",
            "293000 :  0.0\n",
            "294000 :  0.0\n",
            "295000 :  0.0\n",
            "296000 :  0.0\n",
            "297000 :  0.0\n",
            "298000 :  0.0\n",
            "299000 :  0.0\n",
            "300000 :  0.0\n",
            "301000 :  0.0\n",
            "302000 :  0.0\n",
            "303000 :  0.0\n",
            "304000 :  0.0\n",
            "305000 :  0.0\n",
            "306000 :  0.0\n",
            "307000 :  0.0\n",
            "308000 :  0.0\n",
            "309000 :  0.0\n",
            "310000 :  0.0\n",
            "311000 :  0.0\n",
            "312000 :  0.0\n",
            "313000 :  0.0\n",
            "314000 :  0.0\n",
            "315000 :  0.0\n",
            "316000 :  0.0\n",
            "317000 :  0.0\n",
            "318000 :  0.0\n",
            "319000 :  0.0\n",
            "320000 :  0.0\n",
            "321000 :  0.0\n",
            "322000 :  0.0\n",
            "323000 :  0.0\n",
            "324000 :  0.0\n",
            "325000 :  0.0\n",
            "326000 :  0.0\n",
            "327000 :  0.0\n",
            "328000 :  0.0\n",
            "329000 :  0.0\n",
            "330000 :  0.0\n",
            "331000 :  0.0\n",
            "332000 :  0.0\n",
            "333000 :  0.0\n",
            "334000 :  0.0\n",
            "335000 :  0.0\n",
            "336000 :  0.0\n",
            "337000 :  0.0\n",
            "338000 :  0.0\n",
            "339000 :  0.0\n",
            "340000 :  0.0\n",
            "341000 :  0.0\n",
            "342000 :  0.0\n",
            "343000 :  0.0\n",
            "344000 :  0.0\n",
            "345000 :  0.0\n",
            "346000 :  0.0\n",
            "347000 :  0.0\n",
            "348000 :  0.0\n",
            "349000 :  0.0\n",
            "350000 :  0.0\n",
            "351000 :  0.0\n",
            "352000 :  0.0\n",
            "353000 :  0.0\n",
            "354000 :  0.0\n",
            "355000 :  0.0\n",
            "356000 :  0.0\n",
            "357000 :  0.0\n",
            "358000 :  0.0\n",
            "359000 :  0.0\n",
            "360000 :  0.0\n",
            "361000 :  0.0\n",
            "362000 :  0.0\n",
            "363000 :  0.0\n",
            "364000 :  0.0\n",
            "365000 :  0.0\n",
            "366000 :  0.0\n",
            "367000 :  0.0\n",
            "368000 :  0.0\n",
            "369000 :  0.0\n",
            "370000 :  0.0\n",
            "371000 :  0.0\n",
            "372000 :  0.0\n",
            "373000 :  0.0\n",
            "374000 :  0.0\n",
            "375000 :  0.0\n",
            "376000 :  0.0\n",
            "377000 :  0.0\n",
            "378000 :  0.0\n",
            "379000 :  0.0\n",
            "380000 :  0.0\n",
            "381000 :  0.0\n",
            "382000 :  0.0\n",
            "383000 :  0.0\n",
            "384000 :  0.0\n",
            "385000 :  0.0\n",
            "386000 :  0.0\n",
            "387000 :  0.0\n",
            "388000 :  0.0\n",
            "389000 :  0.0\n",
            "390000 :  0.0\n",
            "391000 :  0.0\n",
            "392000 :  0.0\n",
            "393000 :  0.0\n",
            "394000 :  0.0\n",
            "395000 :  0.0\n",
            "396000 :  0.0\n",
            "397000 :  0.0\n",
            "398000 :  0.0\n",
            "399000 :  0.0\n",
            "400000 :  0.0\n",
            "401000 :  0.0\n",
            "402000 :  0.0\n",
            "403000 :  0.0\n",
            "404000 :  0.0\n",
            "405000 :  0.0\n",
            "406000 :  0.0\n",
            "407000 :  0.0\n",
            "408000 :  0.0\n",
            "409000 :  0.0\n",
            "410000 :  0.0\n",
            "411000 :  0.0\n",
            "412000 :  0.0\n",
            "413000 :  0.0\n",
            "414000 :  0.0\n",
            "415000 :  0.0\n",
            "416000 :  0.0\n",
            "417000 :  0.0\n",
            "418000 :  0.0\n",
            "419000 :  0.0\n",
            "420000 :  0.0\n",
            "421000 :  0.0\n",
            "422000 :  0.0\n",
            "423000 :  0.0\n",
            "424000 :  0.0\n",
            "425000 :  0.0\n",
            "426000 :  0.0\n",
            "427000 :  0.0\n",
            "428000 :  0.0\n",
            "429000 :  0.0\n",
            "430000 :  0.0\n",
            "431000 :  0.0\n",
            "432000 :  0.0\n",
            "433000 :  0.0\n",
            "434000 :  0.0\n",
            "435000 :  0.0\n",
            "436000 :  0.0\n",
            "437000 :  0.0\n",
            "438000 :  0.0\n",
            "439000 :  0.0\n",
            "440000 :  0.0\n",
            "441000 :  0.0\n",
            "442000 :  0.0\n",
            "443000 :  0.0\n",
            "444000 :  0.0\n",
            "445000 :  0.0\n",
            "446000 :  0.0\n",
            "447000 :  0.0\n",
            "448000 :  0.0\n",
            "449000 :  0.0\n",
            "450000 :  0.0\n",
            "451000 :  0.0\n",
            "452000 :  0.0\n",
            "453000 :  0.0\n",
            "454000 :  0.0\n",
            "455000 :  0.0\n",
            "456000 :  0.0\n",
            "457000 :  0.0\n",
            "458000 :  0.0\n",
            "459000 :  0.0\n",
            "460000 :  0.0\n",
            "461000 :  0.0\n",
            "462000 :  0.0\n",
            "463000 :  0.0\n",
            "464000 :  0.0\n",
            "465000 :  0.0\n",
            "466000 :  0.0\n",
            "467000 :  0.0\n",
            "468000 :  0.0\n",
            "469000 :  0.0\n",
            "470000 :  0.0\n",
            "471000 :  0.0\n",
            "472000 :  0.0\n",
            "473000 :  0.0\n",
            "474000 :  0.0\n",
            "475000 :  0.0\n",
            "476000 :  0.0\n",
            "477000 :  0.0\n",
            "478000 :  0.0\n",
            "479000 :  0.0\n",
            "480000 :  0.0\n",
            "481000 :  0.0\n",
            "482000 :  0.0\n",
            "483000 :  0.0\n",
            "484000 :  0.0\n",
            "485000 :  0.0\n",
            "486000 :  0.0\n",
            "487000 :  0.0\n",
            "488000 :  0.0\n",
            "489000 :  0.0\n",
            "490000 :  0.0\n",
            "491000 :  0.0\n",
            "492000 :  0.0\n",
            "493000 :  0.0\n",
            "494000 :  0.0\n",
            "495000 :  0.0\n",
            "496000 :  0.0\n",
            "497000 :  0.0\n",
            "498000 :  0.0\n",
            "499000 :  0.0\n",
            "500000 :  0.0\n",
            "501000 :  0.0\n",
            "502000 :  0.0\n",
            "503000 :  0.0\n",
            "504000 :  0.0\n",
            "505000 :  0.0\n",
            "506000 :  0.0\n",
            "507000 :  0.0\n",
            "508000 :  0.0\n",
            "509000 :  0.0\n",
            "510000 :  0.0\n",
            "511000 :  0.0\n",
            "512000 :  0.0\n",
            "513000 :  0.0\n",
            "514000 :  0.0\n",
            "515000 :  0.0\n",
            "516000 :  0.0\n",
            "517000 :  0.0\n",
            "518000 :  0.0\n",
            "519000 :  0.0\n",
            "520000 :  0.0\n",
            "521000 :  0.0\n",
            "522000 :  0.0\n",
            "523000 :  0.0\n",
            "524000 :  0.0\n",
            "525000 :  0.0\n",
            "526000 :  0.0\n",
            "527000 :  0.0\n",
            "528000 :  0.0\n",
            "529000 :  0.0\n",
            "530000 :  0.0\n",
            "531000 :  0.0\n",
            "532000 :  0.0\n",
            "533000 :  0.0\n",
            "534000 :  0.0\n",
            "535000 :  0.0\n",
            "536000 :  0.0\n",
            "537000 :  0.0\n",
            "538000 :  0.0\n",
            "539000 :  0.0\n",
            "540000 :  0.0\n",
            "541000 :  0.0\n",
            "542000 :  0.0\n",
            "543000 :  0.0\n",
            "544000 :  0.0\n",
            "545000 :  0.0\n",
            "546000 :  0.0\n",
            "547000 :  0.0\n",
            "548000 :  0.0\n",
            "549000 :  0.0\n",
            "550000 :  0.0\n",
            "551000 :  0.0\n",
            "552000 :  0.0\n",
            "553000 :  0.0\n",
            "554000 :  0.0\n",
            "555000 :  0.0\n",
            "556000 :  0.0\n",
            "557000 :  0.0\n",
            "558000 :  0.0\n",
            "559000 :  0.0\n",
            "560000 :  0.0\n",
            "561000 :  0.0\n",
            "562000 :  0.0\n",
            "563000 :  0.0\n",
            "564000 :  0.0\n",
            "565000 :  0.0\n",
            "566000 :  0.0\n",
            "567000 :  0.0\n",
            "568000 :  0.0\n",
            "569000 :  0.0\n",
            "570000 :  0.0\n",
            "571000 :  0.0\n",
            "572000 :  0.0\n",
            "573000 :  0.0\n",
            "574000 :  0.0\n",
            "575000 :  0.0\n",
            "576000 :  0.0\n",
            "577000 :  0.0\n",
            "578000 :  0.0\n",
            "579000 :  0.0\n",
            "580000 :  0.0\n",
            "581000 :  0.0\n",
            "582000 :  0.0\n",
            "583000 :  0.0\n",
            "584000 :  0.0\n",
            "585000 :  0.0\n",
            "586000 :  0.0\n",
            "587000 :  0.0\n",
            "588000 :  0.0\n",
            "589000 :  0.0\n",
            "590000 :  0.0\n",
            "591000 :  0.0\n",
            "592000 :  0.0\n",
            "593000 :  0.0\n",
            "594000 :  0.0\n",
            "595000 :  0.0\n",
            "596000 :  0.0\n",
            "597000 :  0.0\n",
            "598000 :  0.0\n",
            "599000 :  0.0\n",
            "600000 :  0.0\n",
            "601000 :  0.0\n",
            "602000 :  0.0\n",
            "603000 :  0.0\n",
            "604000 :  0.0\n",
            "605000 :  0.0\n",
            "606000 :  0.0\n",
            "607000 :  0.0\n",
            "608000 :  0.0\n",
            "609000 :  0.0\n",
            "610000 :  0.0\n",
            "611000 :  0.0\n",
            "612000 :  0.0\n",
            "613000 :  0.0\n",
            "614000 :  0.0\n",
            "615000 :  0.0\n",
            "616000 :  0.0\n",
            "617000 :  0.0\n",
            "618000 :  0.0\n",
            "619000 :  0.0\n",
            "620000 :  0.0\n",
            "621000 :  0.0\n",
            "622000 :  0.0\n",
            "623000 :  0.0\n",
            "624000 :  0.0\n",
            "625000 :  0.0\n",
            "626000 :  0.0\n",
            "627000 :  0.0\n",
            "628000 :  0.0\n",
            "629000 :  0.0\n",
            "630000 :  0.0\n",
            "631000 :  0.0\n",
            "632000 :  0.0\n",
            "633000 :  0.0\n",
            "634000 :  0.0\n",
            "635000 :  0.0\n",
            "636000 :  0.0\n",
            "637000 :  0.0\n",
            "638000 :  0.0\n",
            "639000 :  0.0\n",
            "640000 :  0.0\n",
            "641000 :  0.0\n",
            "642000 :  0.0\n",
            "643000 :  0.0\n",
            "644000 :  0.0\n",
            "645000 :  0.0\n",
            "646000 :  0.0\n",
            "647000 :  0.0\n",
            "648000 :  0.0\n",
            "649000 :  0.0\n",
            "650000 :  0.0\n",
            "651000 :  0.0\n",
            "652000 :  0.0\n",
            "653000 :  0.0\n",
            "654000 :  0.0\n",
            "655000 :  0.0\n",
            "656000 :  0.0\n",
            "657000 :  0.0\n",
            "658000 :  0.0\n",
            "659000 :  0.0\n",
            "660000 :  0.0\n",
            "661000 :  0.0\n",
            "662000 :  0.0\n",
            "663000 :  0.0\n",
            "664000 :  0.0\n",
            "665000 :  0.0\n",
            "666000 :  0.0\n",
            "667000 :  0.0\n",
            "668000 :  0.0\n",
            "669000 :  0.0\n",
            "670000 :  0.0\n",
            "671000 :  0.0\n",
            "672000 :  0.0\n",
            "673000 :  0.0\n",
            "674000 :  0.0\n",
            "675000 :  0.0\n",
            "676000 :  0.0\n",
            "677000 :  0.0\n",
            "678000 :  0.0\n",
            "679000 :  0.0\n",
            "680000 :  0.0\n",
            "681000 :  0.0\n",
            "682000 :  0.0\n",
            "683000 :  0.0\n",
            "684000 :  0.0\n",
            "685000 :  0.0\n",
            "686000 :  0.0\n",
            "687000 :  0.0\n",
            "688000 :  0.0\n",
            "689000 :  0.0\n",
            "690000 :  0.0\n",
            "691000 :  0.0\n",
            "692000 :  0.0\n",
            "693000 :  0.0\n",
            "694000 :  0.0\n",
            "695000 :  0.0\n",
            "696000 :  0.0\n",
            "697000 :  0.0\n",
            "698000 :  0.0\n",
            "699000 :  0.0\n",
            "700000 :  0.0\n",
            "701000 :  0.0\n",
            "702000 :  0.0\n",
            "703000 :  0.0\n",
            "704000 :  0.0\n",
            "705000 :  0.0\n",
            "706000 :  0.0\n",
            "707000 :  0.0\n",
            "708000 :  0.0\n",
            "709000 :  0.0\n",
            "710000 :  0.0\n",
            "711000 :  0.0\n",
            "712000 :  0.0\n",
            "713000 :  0.0\n",
            "714000 :  0.0\n",
            "715000 :  0.0\n",
            "716000 :  0.0\n",
            "717000 :  0.0\n",
            "718000 :  0.0\n",
            "719000 :  0.0\n",
            "720000 :  0.0\n",
            "721000 :  0.0\n",
            "722000 :  0.0\n",
            "723000 :  0.0\n",
            "724000 :  0.0\n",
            "725000 :  0.0\n",
            "726000 :  0.0\n",
            "727000 :  0.0\n",
            "728000 :  0.0\n",
            "729000 :  0.0\n",
            "730000 :  0.0\n",
            "731000 :  0.0\n",
            "732000 :  0.0\n",
            "733000 :  0.0\n",
            "734000 :  0.0\n",
            "735000 :  0.0\n",
            "736000 :  0.0\n",
            "737000 :  0.0\n",
            "738000 :  0.0\n",
            "739000 :  0.0\n",
            "740000 :  0.0\n",
            "741000 :  0.0\n",
            "742000 :  0.0\n",
            "743000 :  0.0\n",
            "744000 :  0.0\n",
            "745000 :  0.0\n",
            "746000 :  0.0\n",
            "747000 :  0.0\n",
            "748000 :  0.0\n",
            "749000 :  0.0\n",
            "750000 :  0.0\n",
            "751000 :  0.0\n",
            "752000 :  0.0\n",
            "753000 :  0.0\n",
            "754000 :  0.0\n",
            "755000 :  0.0\n",
            "756000 :  0.0\n",
            "757000 :  0.0\n",
            "758000 :  0.0\n",
            "759000 :  0.0\n",
            "760000 :  0.0\n",
            "761000 :  0.0\n",
            "762000 :  0.0\n",
            "763000 :  0.0\n",
            "764000 :  0.0\n",
            "765000 :  0.0\n",
            "766000 :  0.0\n",
            "767000 :  0.0\n",
            "768000 :  0.0\n",
            "769000 :  0.0\n",
            "770000 :  0.0\n",
            "771000 :  0.0\n",
            "772000 :  0.0\n",
            "773000 :  0.0\n",
            "774000 :  0.0\n",
            "775000 :  0.0\n",
            "776000 :  0.0\n",
            "777000 :  0.0\n",
            "778000 :  0.0\n",
            "779000 :  0.0\n",
            "780000 :  0.0\n",
            "781000 :  0.0\n",
            "782000 :  0.0\n",
            "783000 :  0.0\n",
            "784000 :  0.0\n",
            "785000 :  0.0\n",
            "786000 :  0.0\n",
            "787000 :  0.0\n",
            "788000 :  0.0\n",
            "789000 :  0.0\n",
            "790000 :  0.0\n",
            "791000 :  0.0\n",
            "792000 :  0.0\n",
            "793000 :  0.0\n",
            "794000 :  0.0\n",
            "795000 :  0.0\n",
            "796000 :  0.0\n",
            "797000 :  0.0\n",
            "798000 :  0.0\n",
            "799000 :  0.0\n",
            "800000 :  0.0\n",
            "801000 :  0.0\n",
            "802000 :  0.0\n",
            "803000 :  0.0\n",
            "804000 :  0.0\n",
            "805000 :  0.0\n",
            "806000 :  0.0\n",
            "807000 :  0.0\n",
            "808000 :  0.0\n",
            "809000 :  0.0\n",
            "810000 :  0.0\n",
            "811000 :  0.0\n",
            "812000 :  0.0\n",
            "813000 :  0.0\n",
            "814000 :  0.0\n",
            "815000 :  0.0\n",
            "816000 :  0.0\n",
            "817000 :  0.0\n",
            "818000 :  0.0\n",
            "819000 :  0.0\n",
            "820000 :  0.0\n",
            "821000 :  0.0\n",
            "822000 :  0.0\n",
            "823000 :  0.0\n",
            "824000 :  0.0\n",
            "825000 :  0.0\n",
            "826000 :  0.0\n",
            "827000 :  0.0\n",
            "828000 :  0.0\n",
            "829000 :  0.0\n",
            "830000 :  0.0\n",
            "831000 :  0.0\n",
            "832000 :  0.0\n",
            "833000 :  0.0\n",
            "834000 :  0.0\n",
            "835000 :  0.0\n",
            "836000 :  0.0\n",
            "837000 :  0.0\n",
            "838000 :  0.0\n",
            "839000 :  0.0\n",
            "840000 :  0.0\n",
            "841000 :  0.0\n",
            "842000 :  0.0\n",
            "843000 :  0.0\n",
            "844000 :  0.0\n",
            "845000 :  0.0\n",
            "846000 :  0.0\n",
            "847000 :  0.0\n",
            "848000 :  0.0\n",
            "849000 :  0.0\n",
            "850000 :  0.0\n",
            "851000 :  0.0\n",
            "852000 :  0.0\n",
            "853000 :  0.0\n",
            "854000 :  0.0\n",
            "855000 :  0.0\n",
            "856000 :  0.0\n",
            "857000 :  0.0\n",
            "858000 :  0.0\n",
            "859000 :  0.0\n",
            "860000 :  0.0\n",
            "861000 :  0.0\n",
            "862000 :  0.0\n",
            "863000 :  0.0\n",
            "864000 :  0.0\n",
            "865000 :  0.0\n",
            "866000 :  0.0\n",
            "867000 :  0.0\n",
            "868000 :  0.0\n",
            "869000 :  0.0\n",
            "870000 :  0.0\n",
            "871000 :  0.0\n",
            "872000 :  0.0\n",
            "873000 :  0.0\n",
            "874000 :  0.0\n",
            "875000 :  0.0\n",
            "876000 :  0.0\n",
            "877000 :  0.0\n",
            "878000 :  0.0\n",
            "879000 :  0.0\n",
            "880000 :  0.0\n",
            "881000 :  0.0\n",
            "882000 :  0.0\n",
            "883000 :  0.0\n",
            "884000 :  0.0\n",
            "885000 :  0.0\n",
            "886000 :  0.0\n",
            "887000 :  0.0\n",
            "888000 :  0.0\n",
            "889000 :  0.0\n",
            "890000 :  0.0\n",
            "891000 :  0.0\n",
            "892000 :  0.0\n",
            "893000 :  0.0\n",
            "894000 :  0.0\n",
            "895000 :  0.0\n",
            "896000 :  0.0\n",
            "897000 :  0.0\n",
            "898000 :  0.0\n",
            "899000 :  0.0\n",
            "900000 :  0.0\n",
            "901000 :  0.0\n",
            "902000 :  0.0\n",
            "903000 :  0.0\n",
            "904000 :  0.0\n",
            "905000 :  0.0\n",
            "906000 :  0.0\n",
            "907000 :  0.0\n",
            "908000 :  0.0\n",
            "909000 :  0.0\n",
            "910000 :  0.0\n",
            "911000 :  0.0\n",
            "912000 :  0.0\n",
            "913000 :  0.0\n",
            "914000 :  0.0\n",
            "915000 :  0.0\n",
            "916000 :  0.0\n",
            "917000 :  0.0\n",
            "918000 :  0.0\n",
            "919000 :  0.0\n",
            "920000 :  0.0\n",
            "921000 :  0.0\n",
            "922000 :  0.0\n",
            "923000 :  0.0\n",
            "924000 :  0.0\n",
            "925000 :  0.0\n",
            "926000 :  0.0\n",
            "927000 :  0.0\n",
            "928000 :  0.0\n",
            "929000 :  0.0\n",
            "930000 :  0.0\n",
            "931000 :  0.0\n",
            "932000 :  0.0\n",
            "933000 :  0.0\n",
            "934000 :  0.0\n",
            "935000 :  0.0\n",
            "936000 :  0.0\n",
            "937000 :  0.0\n",
            "938000 :  0.0\n",
            "939000 :  0.0\n",
            "940000 :  0.0\n",
            "941000 :  0.0\n",
            "942000 :  0.0\n",
            "943000 :  0.0\n",
            "944000 :  0.0\n",
            "945000 :  0.0\n",
            "946000 :  0.0\n",
            "947000 :  0.0\n",
            "948000 :  0.0\n",
            "949000 :  0.0\n",
            "950000 :  0.0\n",
            "951000 :  0.0\n",
            "952000 :  0.0\n",
            "953000 :  0.0\n",
            "954000 :  0.0\n",
            "955000 :  0.0\n",
            "956000 :  0.0\n",
            "957000 :  0.0\n",
            "958000 :  0.0\n",
            "959000 :  0.0\n",
            "960000 :  0.0\n",
            "961000 :  0.0\n",
            "962000 :  0.0\n",
            "963000 :  0.0\n",
            "964000 :  0.0\n",
            "965000 :  0.0\n",
            "966000 :  0.0\n",
            "967000 :  0.0\n",
            "968000 :  0.0\n",
            "969000 :  0.0\n",
            "970000 :  0.0\n",
            "971000 :  0.0\n",
            "972000 :  0.0\n",
            "973000 :  0.0\n",
            "974000 :  0.0\n",
            "975000 :  0.0\n",
            "976000 :  0.0\n",
            "977000 :  0.0\n",
            "978000 :  0.0\n",
            "979000 :  0.0\n",
            "980000 :  0.0\n",
            "981000 :  0.0\n",
            "982000 :  0.0\n",
            "983000 :  0.0\n",
            "984000 :  0.0\n",
            "985000 :  0.0\n",
            "986000 :  0.0\n",
            "987000 :  0.0\n",
            "988000 :  0.0\n",
            "989000 :  0.0\n",
            "990000 :  0.0\n",
            "991000 :  0.0\n",
            "992000 :  0.0\n",
            "993000 :  0.0\n",
            "994000 :  0.0\n",
            "995000 :  0.0\n",
            "996000 :  0.0\n",
            "997000 :  0.0\n",
            "998000 :  0.0\n",
            "999000 :  0.0\n",
            "1000000 :  0.0\n",
            "1001000 :  0.0\n",
            "1002000 :  0.0\n",
            "1003000 :  0.0\n",
            "1004000 :  0.0\n",
            "1005000 :  0.0\n",
            "1006000 :  0.0\n",
            "1007000 :  0.0\n",
            "1008000 :  0.0\n",
            "1009000 :  0.0\n",
            "1010000 :  0.0\n",
            "1011000 :  0.0\n",
            "1012000 :  0.0\n",
            "1013000 :  0.0\n",
            "1014000 :  0.0\n",
            "1015000 :  0.0\n",
            "1016000 :  0.0\n",
            "1017000 :  0.0\n",
            "1018000 :  0.0\n",
            "1019000 :  0.0\n",
            "1020000 :  0.0\n",
            "1021000 :  0.0\n",
            "1022000 :  0.0\n",
            "1023000 :  0.0\n",
            "1024000 :  0.0\n",
            "1025000 :  0.0\n",
            "1026000 :  0.0\n",
            "1027000 :  0.0\n",
            "1028000 :  0.0\n",
            "1029000 :  0.0\n",
            "1030000 :  0.0\n",
            "1031000 :  0.0\n",
            "1032000 :  0.0\n",
            "1033000 :  0.0\n",
            "1034000 :  0.0\n",
            "1035000 :  0.0\n",
            "1036000 :  0.0\n",
            "1037000 :  0.0\n",
            "1038000 :  0.0\n",
            "1039000 :  0.0\n",
            "1040000 :  0.0\n",
            "1041000 :  0.0\n",
            "1042000 :  0.0\n",
            "1043000 :  0.0\n",
            "1044000 :  0.0\n",
            "1045000 :  0.0\n",
            "1046000 :  0.0\n",
            "1047000 :  0.0\n",
            "1048000 :  0.0\n",
            "1049000 :  0.0\n",
            "1050000 :  0.0\n",
            "1051000 :  0.0\n",
            "1052000 :  0.0\n",
            "1053000 :  0.0\n",
            "1054000 :  0.0\n",
            "1055000 :  0.0\n",
            "1056000 :  0.0\n",
            "1057000 :  0.0\n",
            "1058000 :  0.0\n",
            "1059000 :  0.0\n",
            "1060000 :  0.0\n",
            "1061000 :  0.0\n",
            "1062000 :  0.0\n",
            "1063000 :  0.0\n",
            "1064000 :  0.0\n",
            "1065000 :  0.0\n",
            "1066000 :  0.0\n",
            "1067000 :  0.0\n",
            "1068000 :  0.0\n",
            "1069000 :  0.0\n",
            "1070000 :  0.0\n",
            "1071000 :  0.0\n",
            "1072000 :  0.0\n",
            "1073000 :  0.0\n",
            "1074000 :  0.0\n",
            "1075000 :  0.0\n",
            "1076000 :  0.0\n",
            "1077000 :  0.0\n",
            "1078000 :  0.0\n",
            "1079000 :  0.0\n",
            "1080000 :  0.0\n",
            "1081000 :  0.0\n",
            "1082000 :  0.0\n",
            "1083000 :  0.0\n",
            "1084000 :  0.0\n",
            "1085000 :  0.0\n",
            "1086000 :  0.0\n",
            "1087000 :  0.0\n",
            "1088000 :  0.0\n",
            "1089000 :  0.0\n",
            "1090000 :  0.0\n",
            "1091000 :  0.0\n",
            "1092000 :  0.0\n",
            "1093000 :  0.0\n",
            "1094000 :  0.0\n",
            "1095000 :  0.0\n",
            "1096000 :  0.0\n",
            "1097000 :  0.0\n",
            "1098000 :  0.0\n",
            "1099000 :  0.0\n",
            "1100000 :  0.0\n",
            "1101000 :  0.0\n",
            "1102000 :  0.0\n",
            "1103000 :  0.0\n",
            "1104000 :  0.0\n",
            "1105000 :  0.0\n",
            "1106000 :  0.0\n",
            "1107000 :  0.0\n",
            "1108000 :  0.0\n",
            "1109000 :  0.0\n",
            "1110000 :  0.0\n",
            "1111000 :  0.0\n",
            "1112000 :  0.0\n",
            "1113000 :  0.0\n",
            "1114000 :  0.0\n",
            "1115000 :  0.0\n",
            "1116000 :  0.0\n",
            "1117000 :  0.0\n",
            "1118000 :  0.0\n",
            "1119000 :  0.0\n",
            "1120000 :  0.0\n",
            "1121000 :  0.0\n",
            "1122000 :  0.0\n",
            "1123000 :  0.0\n",
            "1124000 :  0.0\n",
            "1125000 :  0.0\n",
            "1126000 :  0.0\n",
            "1127000 :  0.0\n",
            "1128000 :  0.0\n",
            "1129000 :  0.0\n",
            "1130000 :  0.0\n",
            "1131000 :  0.0\n",
            "1132000 :  0.0\n",
            "1133000 :  0.0\n",
            "1134000 :  0.0\n",
            "1135000 :  0.0\n",
            "1136000 :  0.0\n",
            "1137000 :  0.0\n",
            "1138000 :  0.0\n",
            "1139000 :  0.0\n",
            "1140000 :  0.0\n",
            "1141000 :  0.0\n",
            "1142000 :  0.0\n",
            "1143000 :  0.0\n",
            "1144000 :  0.0\n",
            "1145000 :  0.0\n",
            "1146000 :  0.0\n",
            "1147000 :  0.0\n",
            "1148000 :  0.0\n",
            "1149000 :  0.0\n",
            "1150000 :  0.0\n",
            "1151000 :  0.0\n",
            "1152000 :  0.0\n",
            "1153000 :  0.0\n",
            "1154000 :  0.0\n",
            "1155000 :  0.0\n",
            "1156000 :  0.0\n",
            "1157000 :  0.0\n",
            "1158000 :  0.0\n",
            "1159000 :  0.0\n",
            "1160000 :  0.0\n",
            "1161000 :  0.0\n",
            "1162000 :  0.0\n",
            "1163000 :  0.0\n",
            "1164000 :  0.0\n",
            "1165000 :  0.0\n",
            "1166000 :  0.0\n",
            "1167000 :  0.0\n",
            "1168000 :  0.0\n",
            "1169000 :  0.0\n",
            "1170000 :  0.0\n",
            "1171000 :  0.0\n",
            "1172000 :  0.0\n",
            "1173000 :  0.0\n",
            "1174000 :  0.0\n",
            "1175000 :  0.0\n",
            "1176000 :  0.0\n",
            "1177000 :  0.0\n",
            "1178000 :  0.0\n",
            "1179000 :  0.0\n",
            "1180000 :  0.0\n",
            "1181000 :  0.0\n",
            "1182000 :  0.0\n",
            "1183000 :  0.0\n",
            "1184000 :  0.0\n",
            "1185000 :  0.0\n",
            "1186000 :  0.0\n",
            "1187000 :  0.0\n",
            "1188000 :  0.0\n",
            "1189000 :  0.0\n",
            "1190000 :  0.0\n",
            "1191000 :  0.0\n",
            "1192000 :  0.0\n",
            "1193000 :  0.0\n",
            "1194000 :  0.0\n",
            "1195000 :  0.0\n",
            "1196000 :  0.0\n",
            "1197000 :  0.0\n",
            "1198000 :  0.0\n",
            "1199000 :  0.0\n",
            "1200000 :  0.0\n",
            "1201000 :  0.0\n",
            "1202000 :  0.0\n",
            "1203000 :  0.0\n",
            "1204000 :  0.0\n",
            "1205000 :  0.0\n",
            "1206000 :  0.0\n",
            "1207000 :  0.0\n",
            "1208000 :  0.0\n",
            "1209000 :  0.0\n",
            "1210000 :  0.0\n",
            "1211000 :  0.0\n",
            "1212000 :  0.0\n",
            "1213000 :  0.0\n",
            "1214000 :  0.0\n",
            "1215000 :  0.0\n",
            "1216000 :  0.0\n",
            "1217000 :  0.0\n",
            "1218000 :  0.0\n",
            "1219000 :  0.0\n",
            "1220000 :  0.0\n",
            "1221000 :  0.0\n",
            "1222000 :  0.0\n",
            "1223000 :  0.0\n",
            "1224000 :  0.0\n",
            "1225000 :  0.0\n",
            "1226000 :  0.0\n",
            "1227000 :  0.0\n",
            "1228000 :  0.0\n",
            "1229000 :  0.0\n",
            "1230000 :  0.0\n",
            "1231000 :  0.0\n",
            "1232000 :  0.0\n",
            "1233000 :  0.0\n",
            "1234000 :  0.0\n",
            "1235000 :  0.0\n",
            "1236000 :  0.0\n",
            "1237000 :  0.0\n",
            "1238000 :  0.0\n",
            "1239000 :  0.0\n",
            "1240000 :  0.0\n",
            "1241000 :  0.0\n",
            "1242000 :  0.0\n",
            "1243000 :  0.0\n",
            "1244000 :  0.0\n",
            "1245000 :  0.0\n",
            "1246000 :  0.0\n",
            "1247000 :  0.0\n",
            "1248000 :  0.0\n",
            "1249000 :  0.0\n",
            "1250000 :  0.0\n",
            "1251000 :  0.0\n",
            "1252000 :  0.0\n",
            "1253000 :  0.0\n",
            "1254000 :  0.0\n",
            "1255000 :  0.0\n",
            "1256000 :  0.0\n",
            "1257000 :  0.0\n",
            "1258000 :  0.0\n",
            "1259000 :  0.0\n",
            "1260000 :  0.0\n",
            "1261000 :  0.0\n",
            "1262000 :  0.0\n",
            "1263000 :  0.0\n",
            "1264000 :  0.0\n",
            "1265000 :  0.0\n",
            "1266000 :  0.0\n",
            "1267000 :  0.0\n",
            "1268000 :  0.0\n",
            "1269000 :  0.0\n",
            "1270000 :  0.0\n",
            "1271000 :  0.0\n",
            "1272000 :  0.0\n",
            "1273000 :  0.0\n",
            "1274000 :  0.0\n",
            "1275000 :  0.0\n",
            "1276000 :  0.0\n",
            "1277000 :  0.0\n",
            "1278000 :  0.0\n",
            "1279000 :  0.0\n",
            "1280000 :  0.0\n",
            "1281000 :  0.0\n",
            "1282000 :  0.0\n",
            "1283000 :  0.0\n",
            "1284000 :  0.0\n",
            "1285000 :  0.0\n",
            "1286000 :  0.0\n",
            "1287000 :  0.0\n",
            "1288000 :  0.0\n",
            "1289000 :  0.0\n",
            "1290000 :  0.0\n",
            "1291000 :  0.0\n",
            "1292000 :  0.0\n",
            "1293000 :  0.0\n",
            "1294000 :  0.0\n",
            "1295000 :  0.0\n",
            "1296000 :  0.0\n",
            "1297000 :  0.0\n",
            "1298000 :  0.0\n",
            "1299000 :  0.0\n",
            "1300000 :  0.0\n",
            "1301000 :  0.0\n",
            "1302000 :  0.0\n",
            "1303000 :  0.0\n",
            "1304000 :  0.0\n",
            "1305000 :  0.0\n",
            "1306000 :  0.0\n",
            "1307000 :  0.0\n",
            "1308000 :  0.0\n",
            "1309000 :  0.0\n",
            "1310000 :  0.0\n",
            "1311000 :  0.0\n",
            "1312000 :  0.0\n",
            "1313000 :  0.0\n",
            "1314000 :  0.0\n",
            "1315000 :  0.0\n",
            "1316000 :  0.0\n",
            "1317000 :  0.0\n",
            "1318000 :  0.0\n",
            "1319000 :  0.0\n",
            "1320000 :  0.0\n",
            "1321000 :  0.0\n",
            "1322000 :  0.0\n",
            "1323000 :  0.0\n",
            "1324000 :  0.0\n",
            "1325000 :  0.0\n",
            "1326000 :  0.0\n",
            "1327000 :  0.0\n",
            "1328000 :  0.0\n",
            "1329000 :  0.0\n",
            "1330000 :  0.0\n",
            "1331000 :  0.0\n",
            "1332000 :  0.0\n",
            "1333000 :  0.0\n",
            "1334000 :  0.0\n",
            "1335000 :  0.0\n",
            "1336000 :  0.0\n",
            "1337000 :  0.0\n",
            "1338000 :  0.0\n",
            "1339000 :  0.0\n",
            "1340000 :  0.0\n",
            "1341000 :  0.0\n",
            "1342000 :  0.0\n",
            "1343000 :  0.0\n",
            "1344000 :  0.0\n",
            "1345000 :  0.0\n",
            "1346000 :  0.0\n",
            "1347000 :  0.0\n",
            "1348000 :  0.0\n",
            "1349000 :  0.0\n",
            "1350000 :  0.0\n",
            "1351000 :  0.0\n",
            "1352000 :  0.0\n",
            "1353000 :  0.0\n",
            "1354000 :  0.0\n",
            "1355000 :  0.0\n",
            "1356000 :  0.0\n",
            "1357000 :  0.0\n",
            "1358000 :  0.0\n",
            "1359000 :  0.0\n",
            "1360000 :  0.0\n",
            "1361000 :  0.0\n",
            "1362000 :  0.0\n",
            "1363000 :  0.0\n",
            "1364000 :  0.0\n",
            "1365000 :  0.0\n",
            "1366000 :  0.0\n",
            "1367000 :  0.0\n",
            "1368000 :  0.0\n",
            "1369000 :  0.0\n",
            "1370000 :  0.0\n",
            "1371000 :  0.0\n",
            "1372000 :  0.0\n",
            "1373000 :  0.0\n",
            "1374000 :  0.0\n",
            "1375000 :  0.0\n",
            "1376000 :  0.0\n",
            "1377000 :  0.0\n",
            "1378000 :  0.0\n",
            "1379000 :  0.0\n",
            "1380000 :  0.0\n",
            "1381000 :  0.0\n",
            "1382000 :  0.0\n",
            "1383000 :  0.0\n",
            "1384000 :  0.0\n",
            "1385000 :  0.0\n",
            "1386000 :  0.0\n",
            "1387000 :  0.0\n",
            "1388000 :  0.0\n",
            "1389000 :  0.0\n",
            "1390000 :  0.0\n",
            "1391000 :  0.0\n",
            "1392000 :  0.0\n",
            "1393000 :  0.0\n",
            "1394000 :  0.0\n",
            "1395000 :  0.0\n",
            "1396000 :  0.0\n",
            "1397000 :  0.0\n",
            "1398000 :  0.0\n",
            "1399000 :  0.0\n",
            "1400000 :  0.0\n",
            "1401000 :  0.0\n",
            "1402000 :  0.0\n",
            "1403000 :  0.0\n",
            "1404000 :  0.0\n",
            "1405000 :  0.0\n",
            "1406000 :  0.0\n",
            "1407000 :  0.0\n",
            "1408000 :  0.0\n",
            "1409000 :  0.0\n",
            "1410000 :  0.0\n",
            "1411000 :  0.0\n",
            "1412000 :  0.0\n",
            "1413000 :  0.0\n",
            "1414000 :  0.0\n",
            "1415000 :  0.0\n",
            "1416000 :  0.0\n",
            "1417000 :  0.0\n",
            "1418000 :  0.0\n",
            "1419000 :  0.0\n",
            "1420000 :  0.0\n",
            "1421000 :  0.0\n",
            "1422000 :  0.0\n",
            "1423000 :  0.0\n",
            "1424000 :  0.0\n",
            "1425000 :  0.0\n",
            "1426000 :  0.0\n",
            "1427000 :  0.0\n",
            "1428000 :  0.0\n",
            "1429000 :  0.0\n",
            "1430000 :  0.0\n",
            "1431000 :  0.0\n",
            "1432000 :  0.0\n",
            "1433000 :  0.0\n",
            "1434000 :  0.0\n",
            "1435000 :  0.0\n",
            "1436000 :  0.0\n",
            "1437000 :  0.0\n",
            "1438000 :  0.0\n",
            "1439000 :  0.0\n",
            "1440000 :  0.0\n",
            "1441000 :  0.0\n",
            "1442000 :  0.0\n",
            "1443000 :  0.0\n",
            "1444000 :  0.0\n",
            "1445000 :  0.0\n",
            "1446000 :  0.0\n",
            "1447000 :  0.0\n",
            "1448000 :  0.0\n",
            "1449000 :  0.0\n",
            "1450000 :  0.0\n",
            "1451000 :  0.0\n",
            "1452000 :  0.0\n",
            "1453000 :  0.0\n",
            "1454000 :  0.0\n",
            "1455000 :  0.0\n",
            "1456000 :  0.0\n",
            "1457000 :  0.0\n",
            "1458000 :  0.0\n",
            "1459000 :  0.0\n",
            "1460000 :  0.0\n",
            "1461000 :  0.0\n",
            "1462000 :  0.0\n",
            "1463000 :  0.0\n",
            "1464000 :  0.0\n",
            "1465000 :  0.0\n",
            "1466000 :  0.0\n",
            "1467000 :  0.0\n",
            "1468000 :  0.0\n",
            "1469000 :  0.0\n",
            "1470000 :  0.0\n",
            "1471000 :  0.0\n",
            "1472000 :  0.0\n",
            "1473000 :  0.0\n",
            "1474000 :  0.0\n",
            "1475000 :  0.0\n",
            "1476000 :  0.0\n",
            "1477000 :  0.0\n",
            "1478000 :  0.0\n",
            "1479000 :  0.0\n",
            "1480000 :  0.0\n",
            "1481000 :  0.0\n",
            "1482000 :  0.0\n",
            "1483000 :  0.0\n",
            "1484000 :  0.0\n",
            "1485000 :  0.0\n",
            "1486000 :  0.0\n",
            "1487000 :  0.0\n",
            "1488000 :  0.0\n",
            "1489000 :  0.0\n",
            "1490000 :  0.0\n",
            "1491000 :  0.0\n",
            "1492000 :  0.0\n",
            "1493000 :  0.0\n",
            "1494000 :  0.0\n",
            "1495000 :  0.0\n",
            "1496000 :  0.0\n",
            "1497000 :  0.0\n",
            "1498000 :  0.0\n",
            "1499000 :  0.0\n",
            "1500000 :  0.0\n",
            "1501000 :  0.0\n",
            "1502000 :  0.0\n",
            "1503000 :  0.0\n",
            "1504000 :  0.0\n",
            "1505000 :  0.0\n",
            "1506000 :  0.0\n",
            "1507000 :  0.0\n",
            "1508000 :  0.0\n",
            "1509000 :  0.0\n",
            "1510000 :  0.0\n",
            "1511000 :  0.0\n",
            "1512000 :  0.0\n",
            "1513000 :  0.0\n",
            "1514000 :  0.0\n",
            "1515000 :  0.0\n",
            "1516000 :  0.0\n",
            "1517000 :  0.0\n",
            "1518000 :  0.0\n",
            "1519000 :  0.0\n",
            "1520000 :  0.0\n",
            "1521000 :  0.0\n",
            "1522000 :  0.0\n",
            "1523000 :  0.0\n",
            "1524000 :  0.0\n",
            "1525000 :  0.0\n",
            "1526000 :  0.0\n",
            "1527000 :  0.0\n",
            "1528000 :  0.0\n",
            "1529000 :  0.0\n",
            "1530000 :  0.0\n",
            "1531000 :  0.0\n",
            "1532000 :  0.0\n",
            "1533000 :  0.0\n",
            "1534000 :  0.0\n",
            "1535000 :  0.0\n",
            "1536000 :  0.0\n",
            "1537000 :  0.0\n",
            "1538000 :  0.0\n",
            "1539000 :  0.0\n",
            "1540000 :  0.0\n",
            "1541000 :  0.0\n",
            "1542000 :  0.0\n",
            "1543000 :  0.0\n",
            "1544000 :  0.0\n",
            "1545000 :  0.0\n",
            "1546000 :  0.0\n",
            "1547000 :  0.0\n",
            "1548000 :  0.0\n",
            "1549000 :  0.0\n",
            "1550000 :  0.0\n",
            "1551000 :  0.0\n",
            "1552000 :  0.0\n",
            "1553000 :  0.0\n",
            "1554000 :  0.0\n",
            "1555000 :  0.0\n",
            "1556000 :  0.0\n",
            "1557000 :  0.0\n",
            "1558000 :  0.0\n",
            "1559000 :  0.0\n",
            "1560000 :  0.0\n",
            "1561000 :  0.0\n",
            "1562000 :  0.0\n",
            "1563000 :  0.0\n",
            "1564000 :  0.0\n",
            "1565000 :  0.0\n",
            "1566000 :  0.0\n",
            "1567000 :  0.0\n",
            "1568000 :  0.0\n",
            "1569000 :  0.0\n",
            "1570000 :  0.0\n",
            "1571000 :  0.0\n",
            "1572000 :  0.0\n",
            "1573000 :  0.0\n",
            "1574000 :  0.0\n",
            "1575000 :  0.0\n",
            "1576000 :  0.0\n",
            "1577000 :  0.0\n",
            "1578000 :  0.0\n",
            "1579000 :  0.0\n",
            "1580000 :  0.0\n",
            "1581000 :  0.0\n",
            "1582000 :  0.0\n",
            "1583000 :  0.0\n",
            "1584000 :  0.0\n",
            "1585000 :  0.0\n",
            "1586000 :  0.0\n",
            "1587000 :  0.0\n",
            "1588000 :  0.0\n",
            "1589000 :  0.0\n",
            "1590000 :  0.0\n",
            "1591000 :  0.0\n",
            "1592000 :  0.0\n",
            "1593000 :  0.0\n",
            "1594000 :  0.0\n",
            "1595000 :  0.0\n",
            "1596000 :  0.0\n",
            "1597000 :  0.0\n",
            "1598000 :  0.0\n",
            "1599000 :  0.0\n",
            "1600000 :  0.0\n",
            "1601000 :  0.0\n",
            "1602000 :  0.0\n",
            "1603000 :  0.0\n",
            "1604000 :  0.0\n",
            "1605000 :  0.0\n",
            "1606000 :  0.0\n",
            "1607000 :  0.0\n",
            "1608000 :  0.0\n",
            "1609000 :  0.0\n",
            "1610000 :  0.0\n",
            "1611000 :  0.0\n",
            "1612000 :  0.0\n",
            "1613000 :  0.0\n",
            "1614000 :  0.0\n",
            "1615000 :  0.0\n",
            "1616000 :  0.0\n",
            "1617000 :  0.0\n",
            "1618000 :  0.0\n",
            "1619000 :  0.0\n",
            "1620000 :  0.0\n",
            "1621000 :  0.0\n",
            "1622000 :  0.0\n",
            "1623000 :  0.0\n",
            "1624000 :  0.0\n",
            "1625000 :  0.0\n",
            "1626000 :  0.0\n",
            "1627000 :  0.0\n",
            "1628000 :  0.0\n",
            "1629000 :  0.0\n",
            "1630000 :  0.0\n",
            "1631000 :  0.0\n",
            "1632000 :  0.0\n",
            "1633000 :  0.0\n",
            "1634000 :  0.0\n",
            "1635000 :  0.0\n",
            "1636000 :  0.0\n",
            "1637000 :  0.0\n",
            "1638000 :  0.0\n",
            "1639000 :  0.0\n",
            "1640000 :  0.0\n",
            "1641000 :  0.0\n",
            "1642000 :  0.0\n",
            "1643000 :  0.0\n",
            "1644000 :  0.0\n",
            "1645000 :  0.0\n",
            "1646000 :  0.0\n",
            "1647000 :  0.0\n",
            "1648000 :  0.0\n",
            "1649000 :  0.0\n",
            "1650000 :  0.0\n",
            "1651000 :  0.0\n",
            "1652000 :  0.0\n",
            "1653000 :  0.0\n",
            "1654000 :  0.0\n",
            "1655000 :  0.0\n",
            "1656000 :  0.0\n",
            "1657000 :  0.0\n",
            "1658000 :  0.0\n",
            "1659000 :  0.0\n",
            "1660000 :  0.0\n",
            "1661000 :  0.0\n",
            "1662000 :  0.0\n",
            "1663000 :  0.0\n",
            "1664000 :  0.0\n",
            "1665000 :  0.0\n",
            "1666000 :  0.0\n",
            "1667000 :  0.0\n",
            "1668000 :  0.0\n",
            "1669000 :  0.0\n",
            "1670000 :  0.0\n",
            "1671000 :  0.0\n",
            "1672000 :  0.0\n",
            "1673000 :  0.0\n",
            "1674000 :  0.0\n",
            "1675000 :  0.0\n",
            "1676000 :  0.0\n",
            "1677000 :  0.0\n",
            "1678000 :  0.0\n",
            "1679000 :  0.0\n",
            "1680000 :  0.0\n",
            "1681000 :  0.0\n",
            "1682000 :  0.0\n",
            "1683000 :  0.0\n",
            "1684000 :  0.0\n",
            "1685000 :  0.0\n",
            "1686000 :  0.0\n",
            "1687000 :  0.0\n",
            "1688000 :  0.0\n",
            "1689000 :  0.0\n",
            "1690000 :  0.0\n",
            "1691000 :  0.0\n",
            "1692000 :  0.0\n",
            "1693000 :  0.0\n",
            "1694000 :  0.0\n",
            "1695000 :  0.0\n",
            "1696000 :  0.0\n",
            "1697000 :  0.0\n",
            "1698000 :  0.0\n",
            "1699000 :  0.0\n",
            "1700000 :  0.0\n",
            "1701000 :  0.0\n",
            "1702000 :  0.0\n",
            "1703000 :  0.0\n",
            "1704000 :  0.0\n",
            "1705000 :  0.0\n",
            "1706000 :  0.0\n",
            "1707000 :  0.0\n",
            "1708000 :  0.0\n",
            "1709000 :  0.0\n",
            "1710000 :  0.0\n",
            "1711000 :  0.0\n",
            "1712000 :  0.0\n",
            "1713000 :  0.0\n",
            "1714000 :  0.0\n",
            "1715000 :  0.0\n",
            "1716000 :  0.0\n",
            "1717000 :  0.0\n",
            "1718000 :  0.0\n",
            "1719000 :  0.0\n",
            "1720000 :  0.0\n",
            "1721000 :  0.0\n",
            "1722000 :  0.0\n",
            "1723000 :  0.0\n",
            "1724000 :  0.0\n",
            "1725000 :  0.0\n",
            "1726000 :  0.0\n",
            "1727000 :  0.0\n",
            "1728000 :  0.0\n",
            "1729000 :  0.0\n",
            "1730000 :  0.0\n",
            "1731000 :  0.0\n",
            "1732000 :  0.0\n",
            "1733000 :  0.0\n",
            "1734000 :  0.0\n",
            "1735000 :  0.0\n",
            "1736000 :  0.0\n",
            "1737000 :  0.0\n",
            "1738000 :  0.0\n",
            "1739000 :  0.0\n",
            "1740000 :  0.0\n",
            "1741000 :  0.0\n",
            "1742000 :  0.0\n",
            "1743000 :  0.0\n",
            "1744000 :  0.0\n",
            "1745000 :  0.0\n",
            "1746000 :  0.0\n",
            "1747000 :  0.0\n",
            "1748000 :  0.0\n",
            "1749000 :  0.0\n",
            "1750000 :  0.0\n",
            "1751000 :  0.0\n",
            "1752000 :  0.0\n",
            "1753000 :  0.0\n",
            "1754000 :  0.0\n",
            "1755000 :  0.0\n",
            "1756000 :  0.0\n",
            "1757000 :  0.0\n",
            "1758000 :  0.0\n",
            "1759000 :  0.0\n",
            "1760000 :  0.0\n",
            "1761000 :  0.0\n",
            "1762000 :  0.0\n",
            "1763000 :  0.0\n",
            "1764000 :  0.0\n",
            "1765000 :  0.0\n",
            "1766000 :  0.0\n",
            "1767000 :  0.0\n",
            "1768000 :  0.0\n",
            "1769000 :  0.0\n",
            "1770000 :  0.0\n",
            "1771000 :  0.0\n",
            "1772000 :  0.0\n",
            "1773000 :  0.0\n",
            "1774000 :  0.0\n",
            "1775000 :  0.0\n",
            "1776000 :  0.0\n",
            "1777000 :  0.0\n",
            "1778000 :  0.0\n",
            "1779000 :  0.0\n",
            "1780000 :  0.0\n",
            "1781000 :  0.0\n",
            "1782000 :  0.0\n",
            "1783000 :  0.0\n",
            "1784000 :  0.0\n",
            "1785000 :  0.0\n",
            "1786000 :  0.0\n",
            "1787000 :  0.0\n",
            "1788000 :  0.0\n",
            "1789000 :  0.0\n",
            "1790000 :  0.0\n",
            "1791000 :  0.0\n",
            "1792000 :  0.0\n",
            "1793000 :  0.0\n",
            "1794000 :  0.0\n",
            "1795000 :  0.0\n",
            "1796000 :  0.0\n",
            "1797000 :  0.0\n",
            "1798000 :  0.0\n",
            "1799000 :  0.0\n",
            "1800000 :  0.0\n",
            "1801000 :  0.0\n",
            "1802000 :  0.0\n",
            "1803000 :  0.0\n",
            "1804000 :  0.0\n",
            "1805000 :  0.0\n",
            "1806000 :  0.0\n",
            "1807000 :  0.0\n",
            "1808000 :  0.0\n",
            "1809000 :  0.0\n",
            "1810000 :  0.0\n",
            "1811000 :  0.0\n",
            "1812000 :  0.0\n",
            "1813000 :  0.0\n",
            "1814000 :  0.0\n",
            "1815000 :  0.0\n",
            "1816000 :  0.0\n",
            "1817000 :  0.0\n",
            "1818000 :  0.0\n",
            "1819000 :  0.0\n",
            "1820000 :  0.0\n",
            "1821000 :  0.0\n",
            "1822000 :  0.0\n",
            "1823000 :  0.0\n",
            "1824000 :  0.0\n",
            "1825000 :  0.0\n",
            "1826000 :  0.0\n",
            "1827000 :  0.0\n",
            "1828000 :  0.0\n",
            "1829000 :  0.0\n",
            "1830000 :  0.0\n",
            "1831000 :  0.0\n",
            "1832000 :  0.0\n",
            "1833000 :  0.0\n",
            "1834000 :  0.0\n",
            "1835000 :  0.0\n",
            "1836000 :  0.0\n",
            "1837000 :  0.0\n",
            "1838000 :  0.0\n",
            "1839000 :  0.0\n",
            "1840000 :  0.0\n",
            "1841000 :  0.0\n",
            "1842000 :  0.0\n",
            "1843000 :  0.0\n",
            "1844000 :  0.0\n",
            "1845000 :  0.0\n",
            "1846000 :  0.0\n",
            "1847000 :  0.0\n",
            "1848000 :  0.0\n",
            "1849000 :  0.0\n",
            "1850000 :  0.0\n",
            "1851000 :  0.0\n",
            "1852000 :  0.0\n",
            "1853000 :  0.0\n",
            "1854000 :  0.0\n",
            "1855000 :  0.0\n",
            "1856000 :  0.0\n",
            "1857000 :  0.0\n",
            "1858000 :  0.0\n",
            "1859000 :  0.0\n",
            "1860000 :  0.0\n",
            "1861000 :  0.0\n",
            "1862000 :  0.0\n",
            "1863000 :  0.0\n",
            "1864000 :  0.0\n",
            "1865000 :  0.0\n",
            "1866000 :  0.0\n",
            "1867000 :  0.0\n",
            "1868000 :  0.0\n",
            "1869000 :  0.0\n",
            "1870000 :  0.0\n",
            "1871000 :  0.0\n",
            "1872000 :  0.0\n",
            "1873000 :  0.0\n",
            "1874000 :  0.0\n",
            "1875000 :  0.0\n",
            "1876000 :  0.0\n",
            "1877000 :  0.0\n",
            "1878000 :  0.0\n",
            "1879000 :  0.0\n",
            "1880000 :  0.0\n",
            "1881000 :  0.0\n",
            "1882000 :  0.0\n",
            "1883000 :  0.0\n",
            "1884000 :  0.0\n",
            "1885000 :  0.0\n",
            "1886000 :  0.0\n",
            "1887000 :  0.0\n",
            "1888000 :  0.0\n",
            "1889000 :  0.0\n",
            "1890000 :  0.0\n",
            "1891000 :  0.0\n",
            "1892000 :  0.0\n",
            "1893000 :  0.0\n",
            "1894000 :  0.0\n",
            "1895000 :  0.0\n",
            "1896000 :  0.0\n",
            "1897000 :  0.0\n",
            "1898000 :  0.0\n",
            "1899000 :  0.0\n",
            "1900000 :  0.0\n",
            "1901000 :  0.0\n",
            "1902000 :  0.0\n",
            "1903000 :  0.0\n",
            "1904000 :  0.0\n",
            "1905000 :  0.0\n",
            "1906000 :  0.0\n",
            "1907000 :  0.0\n",
            "1908000 :  0.0\n",
            "1909000 :  0.0\n",
            "1910000 :  0.0\n",
            "1911000 :  0.0\n",
            "1912000 :  0.0\n",
            "1913000 :  0.0\n",
            "1914000 :  0.0\n",
            "1915000 :  0.0\n",
            "1916000 :  0.0\n",
            "1917000 :  0.0\n",
            "1918000 :  0.0\n",
            "1919000 :  0.0\n",
            "1920000 :  0.0\n",
            "1921000 :  0.0\n",
            "1922000 :  0.0\n",
            "1923000 :  0.0\n",
            "1924000 :  0.0\n",
            "1925000 :  0.0\n",
            "1926000 :  0.0\n",
            "1927000 :  0.0\n",
            "1928000 :  0.0\n",
            "1929000 :  0.0\n",
            "1930000 :  0.0\n",
            "1931000 :  0.0\n",
            "1932000 :  0.0\n",
            "1933000 :  0.0\n",
            "1934000 :  0.0\n",
            "1935000 :  0.0\n",
            "1936000 :  0.0\n",
            "1937000 :  0.0\n",
            "1938000 :  0.0\n",
            "1939000 :  0.0\n",
            "1940000 :  0.0\n",
            "1941000 :  0.0\n",
            "1942000 :  0.0\n",
            "1943000 :  0.0\n",
            "1944000 :  0.0\n",
            "1945000 :  0.0\n",
            "1946000 :  0.0\n",
            "1947000 :  0.0\n",
            "1948000 :  0.0\n",
            "1949000 :  0.0\n",
            "1950000 :  0.0\n",
            "1951000 :  0.0\n",
            "1952000 :  0.0\n",
            "1953000 :  0.0\n",
            "1954000 :  0.0\n",
            "1955000 :  0.0\n",
            "1956000 :  0.0\n",
            "1957000 :  0.0\n",
            "1958000 :  0.0\n",
            "1959000 :  0.0\n",
            "1960000 :  0.0\n",
            "1961000 :  0.0\n",
            "1962000 :  0.0\n",
            "1963000 :  0.0\n",
            "1964000 :  0.0\n",
            "1965000 :  0.0\n",
            "1966000 :  0.0\n",
            "1967000 :  0.0\n",
            "1968000 :  0.0\n",
            "1969000 :  0.0\n",
            "1970000 :  0.0\n",
            "1971000 :  0.0\n",
            "1972000 :  0.0\n",
            "1973000 :  0.0\n",
            "1974000 :  0.0\n",
            "1975000 :  0.0\n",
            "1976000 :  0.0\n",
            "1977000 :  0.0\n",
            "1978000 :  0.0\n",
            "1979000 :  0.0\n",
            "1980000 :  0.0\n",
            "1981000 :  0.0\n",
            "1982000 :  0.0\n",
            "1983000 :  0.0\n",
            "1984000 :  0.0\n",
            "1985000 :  0.0\n",
            "1986000 :  0.0\n",
            "1987000 :  0.0\n",
            "1988000 :  0.0\n",
            "1989000 :  0.0\n",
            "1990000 :  0.0\n",
            "1991000 :  0.0\n",
            "1992000 :  0.0\n",
            "1993000 :  0.0\n",
            "1994000 :  0.0\n",
            "1995000 :  0.0\n",
            "1996000 :  0.0\n",
            "1997000 :  0.0\n",
            "1998000 :  0.0\n",
            "1999000 :  0.0\n",
            "2000000 :  0.0\n",
            "2001000 :  0.0\n",
            "2002000 :  0.0\n",
            "2003000 :  0.0\n",
            "2004000 :  0.0\n",
            "2005000 :  0.0\n",
            "2006000 :  0.0\n",
            "2007000 :  0.0\n",
            "2008000 :  0.0\n",
            "2009000 :  0.0\n",
            "2010000 :  0.0\n",
            "2011000 :  0.0\n",
            "2012000 :  0.0\n",
            "2013000 :  0.0\n",
            "2014000 :  0.0\n",
            "2015000 :  0.0\n",
            "2016000 :  0.0\n",
            "2017000 :  0.0\n",
            "2018000 :  0.0\n",
            "2019000 :  0.0\n",
            "2020000 :  0.0\n",
            "2021000 :  0.0\n",
            "2022000 :  0.0\n",
            "2023000 :  0.0\n",
            "2024000 :  0.0\n",
            "2025000 :  0.0\n",
            "2026000 :  0.0\n",
            "2027000 :  0.0\n",
            "2028000 :  0.0\n",
            "2029000 :  0.0\n",
            "2030000 :  0.0\n",
            "2031000 :  0.0\n",
            "2032000 :  0.0\n",
            "2033000 :  0.0\n",
            "2034000 :  0.0\n",
            "2035000 :  0.0\n",
            "2036000 :  0.0\n",
            "2037000 :  0.0\n",
            "2038000 :  0.0\n",
            "2039000 :  0.0\n",
            "2040000 :  0.0\n",
            "2041000 :  0.0\n",
            "2042000 :  0.0\n",
            "2043000 :  0.0\n",
            "2044000 :  0.0\n",
            "2045000 :  0.0\n",
            "2046000 :  0.0\n",
            "2047000 :  0.0\n",
            "2048000 :  0.0\n",
            "2049000 :  0.0\n",
            "2050000 :  0.0\n",
            "2051000 :  0.0\n",
            "2052000 :  0.0\n",
            "2053000 :  0.0\n",
            "2054000 :  0.0\n",
            "2055000 :  0.0\n",
            "2056000 :  0.0\n",
            "2057000 :  0.0\n",
            "2058000 :  0.0\n",
            "2059000 :  0.0\n",
            "2060000 :  0.0\n",
            "2061000 :  0.0\n",
            "2062000 :  0.0\n",
            "2063000 :  0.0\n",
            "2064000 :  0.0\n",
            "2065000 :  0.0\n",
            "2066000 :  0.0\n",
            "2067000 :  0.0\n",
            "2068000 :  0.0\n",
            "2069000 :  0.0\n",
            "2070000 :  0.0\n",
            "2071000 :  0.0\n",
            "2072000 :  0.0\n",
            "2073000 :  0.0\n",
            "2074000 :  0.0\n",
            "2075000 :  0.0\n",
            "2076000 :  0.0\n",
            "2077000 :  0.0\n",
            "2078000 :  0.0\n",
            "2079000 :  0.0\n",
            "2080000 :  0.0\n",
            "2081000 :  0.0\n",
            "2082000 :  0.0\n",
            "2083000 :  0.0\n",
            "2084000 :  0.0\n",
            "2085000 :  0.0\n",
            "2086000 :  0.0\n",
            "2087000 :  0.0\n",
            "2088000 :  0.0\n",
            "2089000 :  0.0\n",
            "2090000 :  0.0\n",
            "2091000 :  0.0\n",
            "2092000 :  0.0\n",
            "2093000 :  0.0\n",
            "2094000 :  0.0\n",
            "2095000 :  0.0\n",
            "2096000 :  0.0\n",
            "2097000 :  0.0\n",
            "2098000 :  0.0\n",
            "2099000 :  0.0\n",
            "2100000 :  0.0\n",
            "2101000 :  0.0\n",
            "2102000 :  0.0\n",
            "2103000 :  0.0\n",
            "2104000 :  0.0\n",
            "2105000 :  0.0\n",
            "2106000 :  0.0\n",
            "2107000 :  0.0\n",
            "2108000 :  0.0\n",
            "2109000 :  0.0\n",
            "2110000 :  0.0\n",
            "2111000 :  0.0\n",
            "2112000 :  0.0\n",
            "2113000 :  0.0\n",
            "2114000 :  0.0\n",
            "2115000 :  0.0\n",
            "2116000 :  0.0\n",
            "2117000 :  0.0\n",
            "2118000 :  0.0\n",
            "2119000 :  0.0\n",
            "2120000 :  0.0\n",
            "2121000 :  0.0\n",
            "2122000 :  0.0\n",
            "2123000 :  0.0\n",
            "2124000 :  0.0\n",
            "2125000 :  0.0\n",
            "2126000 :  0.0\n",
            "2127000 :  0.0\n",
            "2128000 :  0.0\n",
            "2129000 :  0.0\n",
            "2130000 :  0.0\n",
            "2131000 :  0.0\n",
            "2132000 :  0.0\n",
            "2133000 :  0.0\n",
            "2134000 :  0.0\n",
            "2135000 :  0.0\n",
            "2136000 :  0.0\n",
            "2137000 :  0.0\n",
            "2138000 :  0.0\n",
            "2139000 :  0.0\n",
            "2140000 :  0.0\n",
            "2141000 :  0.0\n",
            "2142000 :  0.0\n",
            "2143000 :  0.0\n",
            "2144000 :  0.0\n",
            "2145000 :  0.0\n",
            "2146000 :  0.0\n",
            "2147000 :  0.0\n",
            "2148000 :  0.0\n",
            "2149000 :  0.0\n",
            "2150000 :  0.0\n",
            "2151000 :  0.0\n",
            "2152000 :  0.0\n",
            "2153000 :  0.0\n",
            "2154000 :  0.0\n",
            "2155000 :  0.0\n",
            "2156000 :  0.0\n",
            "2157000 :  0.0\n",
            "2158000 :  0.0\n",
            "2159000 :  0.0\n",
            "2160000 :  0.0\n",
            "2161000 :  0.0\n",
            "2162000 :  0.0\n",
            "2163000 :  0.0\n",
            "2164000 :  0.0\n",
            "2165000 :  0.0\n",
            "2166000 :  0.0\n",
            "2167000 :  0.0\n",
            "2168000 :  0.0\n",
            "2169000 :  0.0\n",
            "2170000 :  0.0\n",
            "2171000 :  0.0\n",
            "2172000 :  0.0\n",
            "2173000 :  0.0\n",
            "2174000 :  0.0\n",
            "2175000 :  0.0\n",
            "2176000 :  0.0\n",
            "2177000 :  0.0\n",
            "2178000 :  0.0\n",
            "2179000 :  0.0\n",
            "2180000 :  0.0\n",
            "2181000 :  0.0\n",
            "2182000 :  0.0\n",
            "2183000 :  0.0\n",
            "2184000 :  0.0\n",
            "2185000 :  0.0\n",
            "2186000 :  0.0\n",
            "2187000 :  0.0\n",
            "2188000 :  0.0\n",
            "2189000 :  0.0\n",
            "2190000 :  0.0\n",
            "2191000 :  0.0\n",
            "2192000 :  0.0\n",
            "2193000 :  0.0\n",
            "2194000 :  0.0\n",
            "2195000 :  0.0\n",
            "2196000 :  0.0\n",
            "2197000 :  0.0\n",
            "2198000 :  0.0\n",
            "2199000 :  0.0\n",
            "2200000 :  0.0\n",
            "2201000 :  0.0\n",
            "2202000 :  0.0\n",
            "2203000 :  0.0\n",
            "2204000 :  0.0\n",
            "2205000 :  0.0\n",
            "2206000 :  0.0\n",
            "2207000 :  0.0\n",
            "2208000 :  0.0\n",
            "2209000 :  0.0\n",
            "2210000 :  0.0\n",
            "2211000 :  0.0\n",
            "2212000 :  0.0\n",
            "2213000 :  0.0\n",
            "2214000 :  0.0\n",
            "2215000 :  0.0\n",
            "2216000 :  0.0\n",
            "2217000 :  0.0\n",
            "2218000 :  0.0\n",
            "2219000 :  0.0\n",
            "2220000 :  0.0\n",
            "2221000 :  0.0\n",
            "2222000 :  0.0\n",
            "2223000 :  0.0\n",
            "2224000 :  0.0\n",
            "2225000 :  0.0\n",
            "2226000 :  0.0\n",
            "2227000 :  0.0\n",
            "2228000 :  0.0\n",
            "2229000 :  0.0\n",
            "2230000 :  0.0\n",
            "2231000 :  0.0\n",
            "2232000 :  0.0\n",
            "2233000 :  0.0\n",
            "2234000 :  0.0\n",
            "2235000 :  0.0\n",
            "2236000 :  0.0\n",
            "2237000 :  0.0\n",
            "2238000 :  0.0\n",
            "2239000 :  0.0\n",
            "2240000 :  0.0\n",
            "2241000 :  0.0\n",
            "2242000 :  0.0\n",
            "2243000 :  0.0\n",
            "2244000 :  0.0\n",
            "2245000 :  0.0\n",
            "2246000 :  0.0\n",
            "2247000 :  0.0\n",
            "2248000 :  0.0\n",
            "2249000 :  0.0\n",
            "2250000 :  0.0\n",
            "2251000 :  0.0\n",
            "2252000 :  0.0\n",
            "2253000 :  0.0\n",
            "2254000 :  0.0\n",
            "2255000 :  0.0\n",
            "2256000 :  0.0\n",
            "2257000 :  0.0\n",
            "2258000 :  0.0\n",
            "2259000 :  0.0\n",
            "2260000 :  0.0\n",
            "2261000 :  0.0\n",
            "2262000 :  0.0\n",
            "2263000 :  0.0\n",
            "2264000 :  0.0\n",
            "2265000 :  0.0\n",
            "2266000 :  0.0\n",
            "2267000 :  0.0\n",
            "2268000 :  0.0\n",
            "2269000 :  0.0\n",
            "2270000 :  0.0\n",
            "2271000 :  0.0\n",
            "2272000 :  0.0\n",
            "2273000 :  0.0\n",
            "2274000 :  0.0\n",
            "2275000 :  0.0\n",
            "2276000 :  0.0\n",
            "2277000 :  0.0\n",
            "2278000 :  0.0\n",
            "2279000 :  0.0\n",
            "2280000 :  0.0\n",
            "2281000 :  0.0\n",
            "2282000 :  0.0\n",
            "2283000 :  0.0\n",
            "2284000 :  0.0\n",
            "2285000 :  0.0\n",
            "2286000 :  0.0\n",
            "2287000 :  0.0\n",
            "2288000 :  0.0\n",
            "2289000 :  0.0\n",
            "2290000 :  0.0\n",
            "2291000 :  0.0\n",
            "2292000 :  0.0\n",
            "2293000 :  0.0\n",
            "2294000 :  0.0\n",
            "2295000 :  0.0\n",
            "2296000 :  0.0\n",
            "2297000 :  0.0\n",
            "2298000 :  0.0\n",
            "2299000 :  0.0\n",
            "2300000 :  0.0\n",
            "2301000 :  0.0\n",
            "2302000 :  0.0\n",
            "2303000 :  0.0\n",
            "2304000 :  0.0\n",
            "2305000 :  0.0\n",
            "2306000 :  0.0\n",
            "2307000 :  0.0\n",
            "2308000 :  0.0\n",
            "2309000 :  0.0\n",
            "2310000 :  0.0\n",
            "2311000 :  0.0\n",
            "2312000 :  0.0\n",
            "2313000 :  0.0\n",
            "2314000 :  0.0\n",
            "2315000 :  0.0\n",
            "2316000 :  0.0\n",
            "2317000 :  0.0\n",
            "2318000 :  0.0\n",
            "2319000 :  0.0\n",
            "2320000 :  0.0\n",
            "2321000 :  0.0\n",
            "2322000 :  0.0\n",
            "2323000 :  0.0\n",
            "2324000 :  0.0\n",
            "2325000 :  0.0\n",
            "2326000 :  0.0\n",
            "2327000 :  0.0\n",
            "2328000 :  0.0\n",
            "2329000 :  0.0\n",
            "2330000 :  0.0\n",
            "2331000 :  0.0\n",
            "2332000 :  0.0\n",
            "2333000 :  0.0\n",
            "2334000 :  0.0\n",
            "2335000 :  0.0\n",
            "2336000 :  0.0\n",
            "2337000 :  0.0\n",
            "2338000 :  0.0\n",
            "2339000 :  0.0\n",
            "2340000 :  0.0\n",
            "2341000 :  0.0\n",
            "2342000 :  0.0\n",
            "2343000 :  0.0\n",
            "2344000 :  0.0\n",
            "2345000 :  0.0\n",
            "2346000 :  0.0\n",
            "2347000 :  0.0\n",
            "2348000 :  0.0\n",
            "2349000 :  0.0\n",
            "2350000 :  0.0\n",
            "2351000 :  0.0\n",
            "2352000 :  0.0\n",
            "2353000 :  0.0\n",
            "2354000 :  0.0\n",
            "2355000 :  0.0\n",
            "2356000 :  0.0\n",
            "2357000 :  0.0\n",
            "2358000 :  0.0\n",
            "2359000 :  0.0\n",
            "2360000 :  0.0\n",
            "2361000 :  0.0\n",
            "2362000 :  0.0\n",
            "2363000 :  0.0\n",
            "2364000 :  0.0\n",
            "2365000 :  0.0\n",
            "2366000 :  0.0\n",
            "2367000 :  0.0\n",
            "2368000 :  0.0\n",
            "2369000 :  0.0\n",
            "2370000 :  0.0\n",
            "2371000 :  0.0\n",
            "2372000 :  0.0\n",
            "2373000 :  0.0\n",
            "2374000 :  0.0\n",
            "2375000 :  0.0\n",
            "2376000 :  0.0\n",
            "2377000 :  0.0\n",
            "2378000 :  0.0\n",
            "2379000 :  0.0\n",
            "2380000 :  0.0\n",
            "2381000 :  0.0\n",
            "2382000 :  0.0\n",
            "2383000 :  0.0\n",
            "2384000 :  0.0\n",
            "2385000 :  0.0\n",
            "2386000 :  0.0\n",
            "2387000 :  0.0\n",
            "2388000 :  0.0\n",
            "2389000 :  0.0\n",
            "2390000 :  0.0\n",
            "2391000 :  0.0\n",
            "2392000 :  0.0\n",
            "2393000 :  0.0\n",
            "2394000 :  0.0\n",
            "2395000 :  0.0\n",
            "2396000 :  0.0\n",
            "2397000 :  0.0\n",
            "2398000 :  0.0\n",
            "2399000 :  0.0\n",
            "2400000 :  0.0\n",
            "2401000 :  0.0\n",
            "2402000 :  0.0\n",
            "2403000 :  0.0\n",
            "2404000 :  0.0\n",
            "2405000 :  0.0\n",
            "2406000 :  0.0\n",
            "2407000 :  0.0\n",
            "2408000 :  0.0\n",
            "2409000 :  0.0\n",
            "2410000 :  0.0\n",
            "2411000 :  0.0\n",
            "2412000 :  0.0\n",
            "2413000 :  0.0\n",
            "2414000 :  0.0\n",
            "2415000 :  0.0\n",
            "2416000 :  0.0\n",
            "2417000 :  0.0\n",
            "2418000 :  0.0\n",
            "2419000 :  0.0\n",
            "2420000 :  0.0\n",
            "2421000 :  0.0\n",
            "2422000 :  0.0\n",
            "2423000 :  0.0\n",
            "2424000 :  0.0\n",
            "2425000 :  0.0\n",
            "2426000 :  0.0\n",
            "2427000 :  0.0\n",
            "2428000 :  0.0\n",
            "2429000 :  0.0\n",
            "2430000 :  0.0\n",
            "2431000 :  0.0\n",
            "2432000 :  0.0\n",
            "2433000 :  0.0\n",
            "2434000 :  0.0\n",
            "2435000 :  0.0\n",
            "2436000 :  0.0\n",
            "2437000 :  0.0\n",
            "2438000 :  0.0\n",
            "2439000 :  0.0\n",
            "2440000 :  0.0\n",
            "2441000 :  0.0\n",
            "2442000 :  0.0\n",
            "2443000 :  0.0\n",
            "2444000 :  0.0\n",
            "2445000 :  0.0\n",
            "2446000 :  0.0\n",
            "2447000 :  0.0\n",
            "2448000 :  0.0\n",
            "2449000 :  0.0\n",
            "2450000 :  0.0\n",
            "2451000 :  0.0\n",
            "2452000 :  0.0\n",
            "2453000 :  0.0\n",
            "2454000 :  0.0\n",
            "2455000 :  0.0\n",
            "2456000 :  0.0\n",
            "2457000 :  0.0\n",
            "2458000 :  0.0\n",
            "2459000 :  0.0\n",
            "2460000 :  0.0\n",
            "2461000 :  0.0\n",
            "2462000 :  0.0\n",
            "2463000 :  0.0\n",
            "2464000 :  0.0\n",
            "2465000 :  0.0\n",
            "2466000 :  0.0\n",
            "2467000 :  0.0\n",
            "2468000 :  0.0\n",
            "2469000 :  0.0\n",
            "2470000 :  0.0\n",
            "2471000 :  0.0\n",
            "2472000 :  0.0\n",
            "2473000 :  0.0\n",
            "2474000 :  0.0\n",
            "2475000 :  0.0\n",
            "2476000 :  0.0\n",
            "2477000 :  0.0\n",
            "2478000 :  0.0\n",
            "2479000 :  0.0\n",
            "2480000 :  0.0\n",
            "2481000 :  0.0\n",
            "2482000 :  0.0\n",
            "2483000 :  0.0\n",
            "2484000 :  0.0\n",
            "2485000 :  0.0\n",
            "2486000 :  0.0\n",
            "2487000 :  0.0\n",
            "2488000 :  0.0\n",
            "2489000 :  0.0\n",
            "2490000 :  0.0\n",
            "2491000 :  0.0\n",
            "2492000 :  0.0\n",
            "2493000 :  0.0\n",
            "2494000 :  0.0\n",
            "2495000 :  0.0\n",
            "2496000 :  0.0\n",
            "2497000 :  0.0\n",
            "2498000 :  0.0\n",
            "2499000 :  0.0\n",
            "2500000 :  0.0\n",
            "2501000 :  0.0\n",
            "2502000 :  0.0\n",
            "2503000 :  0.0\n",
            "2504000 :  0.0\n",
            "2505000 :  0.0\n",
            "2506000 :  0.0\n",
            "2507000 :  0.0\n",
            "2508000 :  0.0\n",
            "2509000 :  0.0\n",
            "2510000 :  0.0\n",
            "2511000 :  0.0\n",
            "2512000 :  0.0\n",
            "2513000 :  0.0\n",
            "2514000 :  0.0\n",
            "2515000 :  0.0\n",
            "2516000 :  0.0\n",
            "2517000 :  0.0\n",
            "2518000 :  0.0\n",
            "2519000 :  0.0\n",
            "2520000 :  0.0\n",
            "2521000 :  0.0\n",
            "2522000 :  0.0\n",
            "2523000 :  0.0\n",
            "2524000 :  0.0\n",
            "2525000 :  0.0\n",
            "2526000 :  0.0\n",
            "2527000 :  0.0\n",
            "2528000 :  0.0\n",
            "2529000 :  0.0\n",
            "2530000 :  0.0\n",
            "2531000 :  0.0\n",
            "2532000 :  0.0\n",
            "2533000 :  0.0\n",
            "2534000 :  0.0\n",
            "2535000 :  0.0\n",
            "2536000 :  0.0\n",
            "2537000 :  0.0\n",
            "2538000 :  0.0\n",
            "2539000 :  0.0\n",
            "2540000 :  0.0\n",
            "2541000 :  0.0\n",
            "2542000 :  0.0\n",
            "2543000 :  0.0\n",
            "2544000 :  0.0\n",
            "2545000 :  0.0\n",
            "2546000 :  0.0\n",
            "2547000 :  0.0\n",
            "2548000 :  0.0\n",
            "2549000 :  0.0\n",
            "2550000 :  0.0\n",
            "2551000 :  0.0\n",
            "2552000 :  0.0\n",
            "2553000 :  0.0\n",
            "2554000 :  0.0\n",
            "2555000 :  0.0\n",
            "2556000 :  0.0\n",
            "2557000 :  0.0\n",
            "2558000 :  0.0\n",
            "2559000 :  0.0\n",
            "2560000 :  0.0\n",
            "2561000 :  0.0\n",
            "2562000 :  0.0\n",
            "2563000 :  0.0\n",
            "2564000 :  0.0\n",
            "2565000 :  0.0\n",
            "2566000 :  0.0\n",
            "2567000 :  0.0\n",
            "2568000 :  0.0\n",
            "2569000 :  0.0\n",
            "2570000 :  0.0\n",
            "2571000 :  0.0\n",
            "2572000 :  0.0\n",
            "2573000 :  0.0\n",
            "2574000 :  0.0\n",
            "2575000 :  0.0\n",
            "2576000 :  0.0\n",
            "2577000 :  0.0\n",
            "2578000 :  0.0\n",
            "2579000 :  0.0\n",
            "2580000 :  0.0\n",
            "2581000 :  0.0\n",
            "2582000 :  0.0\n",
            "2583000 :  0.0\n",
            "2584000 :  0.0\n",
            "2585000 :  0.0\n",
            "2586000 :  0.0\n",
            "2587000 :  0.0\n",
            "2588000 :  0.0\n",
            "2589000 :  0.0\n",
            "2590000 :  0.0\n",
            "2591000 :  0.0\n",
            "2592000 :  0.0\n",
            "2593000 :  0.0\n",
            "2594000 :  0.0\n",
            "2595000 :  0.0\n",
            "2596000 :  0.0\n",
            "2597000 :  0.0\n",
            "2598000 :  0.0\n",
            "2599000 :  0.0\n",
            "2600000 :  0.0\n",
            "2601000 :  0.0\n",
            "2602000 :  0.0\n",
            "2603000 :  0.0\n",
            "2604000 :  0.0\n",
            "2605000 :  0.0\n",
            "2606000 :  0.0\n",
            "2607000 :  0.0\n",
            "2608000 :  0.0\n",
            "2609000 :  0.0\n",
            "2610000 :  0.0\n",
            "2611000 :  0.0\n",
            "2612000 :  0.0\n",
            "2613000 :  0.0\n",
            "2614000 :  0.0\n",
            "2615000 :  0.0\n",
            "2616000 :  0.0\n",
            "2617000 :  0.0\n",
            "2618000 :  0.0\n",
            "2619000 :  0.0\n",
            "2620000 :  0.0\n",
            "2621000 :  0.0\n",
            "2622000 :  0.0\n",
            "2623000 :  0.0\n",
            "2624000 :  0.0\n",
            "2625000 :  0.0\n",
            "2626000 :  0.0\n",
            "2627000 :  0.0\n",
            "2628000 :  0.0\n",
            "2629000 :  0.0\n",
            "2630000 :  0.0\n",
            "2631000 :  0.0\n",
            "2632000 :  0.0\n",
            "2633000 :  0.0\n",
            "2634000 :  0.0\n",
            "2635000 :  0.0\n",
            "2636000 :  0.0\n",
            "2637000 :  0.0\n",
            "2638000 :  0.0\n",
            "2639000 :  0.0\n",
            "2640000 :  0.0\n",
            "2641000 :  0.0\n",
            "2642000 :  0.0\n",
            "2643000 :  0.0\n",
            "2644000 :  0.0\n",
            "2645000 :  0.0\n",
            "2646000 :  0.0\n",
            "2647000 :  0.0\n",
            "2648000 :  0.0\n",
            "2649000 :  0.0\n",
            "2650000 :  0.0\n",
            "2651000 :  0.0\n",
            "2652000 :  0.0\n",
            "2653000 :  0.0\n",
            "2654000 :  0.0\n",
            "2655000 :  0.0\n",
            "2656000 :  0.0\n",
            "2657000 :  0.0\n",
            "2658000 :  0.0\n",
            "2659000 :  0.0\n",
            "2660000 :  0.0\n",
            "2661000 :  0.0\n",
            "2662000 :  0.0\n",
            "2663000 :  0.0\n",
            "2664000 :  0.0\n",
            "2665000 :  0.0\n",
            "2666000 :  0.0\n",
            "2667000 :  0.0\n",
            "2668000 :  0.0\n",
            "2669000 :  0.0\n",
            "2670000 :  0.0\n",
            "2671000 :  0.0\n",
            "2672000 :  0.0\n",
            "2673000 :  0.0\n",
            "2674000 :  0.0\n",
            "2675000 :  0.0\n",
            "2676000 :  0.0\n",
            "2677000 :  0.0\n",
            "2678000 :  0.0\n",
            "2679000 :  0.0\n",
            "2680000 :  0.0\n",
            "2681000 :  0.0\n",
            "2682000 :  0.0\n",
            "2683000 :  0.0\n",
            "2684000 :  0.0\n",
            "2685000 :  0.0\n",
            "2686000 :  0.0\n",
            "2687000 :  0.0\n",
            "2688000 :  0.0\n",
            "2689000 :  0.0\n",
            "2690000 :  0.0\n",
            "2691000 :  0.0\n",
            "2692000 :  0.0\n",
            "2693000 :  0.0\n",
            "2694000 :  0.0\n",
            "2695000 :  0.0\n",
            "2696000 :  0.0\n",
            "2697000 :  0.0\n",
            "2698000 :  0.0\n",
            "2699000 :  0.0\n",
            "2700000 :  0.0\n",
            "2701000 :  0.0\n",
            "2702000 :  0.0\n",
            "2703000 :  0.0\n",
            "2704000 :  0.0\n",
            "2705000 :  0.0\n",
            "2706000 :  0.0\n",
            "2707000 :  0.0\n",
            "2708000 :  0.0\n",
            "2709000 :  0.0\n",
            "2710000 :  0.0\n",
            "2711000 :  0.0\n",
            "2712000 :  0.0\n",
            "2713000 :  0.0\n",
            "2714000 :  0.0\n",
            "2715000 :  0.0\n",
            "2716000 :  0.0\n",
            "2717000 :  0.0\n",
            "2718000 :  0.0\n",
            "2719000 :  0.0\n",
            "2720000 :  0.0\n",
            "2721000 :  0.0\n",
            "2722000 :  0.0\n",
            "2723000 :  0.0\n",
            "2724000 :  0.0\n",
            "2725000 :  0.0\n",
            "2726000 :  0.0\n",
            "2727000 :  0.0\n",
            "2728000 :  0.0\n",
            "2729000 :  0.0\n",
            "2730000 :  0.0\n",
            "2731000 :  0.0\n",
            "2732000 :  0.0\n",
            "2733000 :  0.0\n",
            "2734000 :  0.0\n",
            "2735000 :  0.0\n",
            "2736000 :  0.0\n",
            "2737000 :  0.0\n",
            "2738000 :  0.0\n",
            "2739000 :  0.0\n",
            "2740000 :  0.0\n",
            "2741000 :  0.0\n",
            "2742000 :  0.0\n",
            "2743000 :  0.0\n",
            "2744000 :  0.0\n",
            "2745000 :  0.0\n",
            "2746000 :  0.0\n",
            "2747000 :  0.0\n",
            "2748000 :  0.0\n",
            "2749000 :  0.0\n",
            "2750000 :  0.0\n",
            "2751000 :  0.0\n",
            "2752000 :  0.0\n",
            "2753000 :  0.0\n",
            "2754000 :  0.0\n",
            "2755000 :  0.0\n",
            "2756000 :  0.0\n",
            "2757000 :  0.0\n",
            "2758000 :  0.0\n",
            "2759000 :  0.0\n",
            "2760000 :  0.0\n",
            "2761000 :  0.0\n",
            "2762000 :  0.0\n",
            "2763000 :  0.0\n",
            "2764000 :  0.0\n",
            "2765000 :  0.0\n",
            "2766000 :  0.0\n",
            "2767000 :  0.0\n",
            "2768000 :  0.0\n",
            "2769000 :  0.0\n",
            "2770000 :  0.0\n",
            "2771000 :  0.0\n",
            "2772000 :  0.0\n",
            "2773000 :  0.0\n",
            "2774000 :  0.0\n",
            "2775000 :  0.0\n",
            "2776000 :  0.0\n",
            "2777000 :  0.0\n",
            "2778000 :  0.0\n",
            "2779000 :  0.0\n",
            "2780000 :  0.0\n",
            "2781000 :  0.0\n",
            "2782000 :  0.0\n",
            "2783000 :  0.0\n",
            "2784000 :  0.0\n",
            "2785000 :  0.0\n",
            "2786000 :  0.0\n",
            "2787000 :  0.0\n",
            "2788000 :  0.0\n",
            "2789000 :  0.0\n",
            "2790000 :  0.0\n",
            "2791000 :  0.0\n",
            "2792000 :  0.0\n",
            "2793000 :  0.0\n",
            "2794000 :  0.0\n",
            "2795000 :  0.0\n",
            "2796000 :  0.0\n",
            "2797000 :  0.0\n",
            "2798000 :  0.0\n",
            "2799000 :  0.0\n",
            "2800000 :  0.0\n",
            "2801000 :  0.0\n",
            "2802000 :  0.0\n",
            "2803000 :  0.0\n",
            "2804000 :  0.0\n",
            "2805000 :  0.0\n",
            "2806000 :  0.0\n",
            "2807000 :  0.0\n",
            "2808000 :  0.0\n",
            "2809000 :  0.0\n",
            "2810000 :  0.0\n",
            "2811000 :  0.0\n",
            "2812000 :  0.0\n",
            "2813000 :  0.0\n",
            "2814000 :  0.0\n",
            "2815000 :  0.0\n",
            "2816000 :  0.0\n",
            "2817000 :  0.0\n",
            "2818000 :  0.0\n",
            "2819000 :  0.0\n",
            "2820000 :  0.0\n",
            "2821000 :  0.0\n",
            "2822000 :  0.0\n",
            "2823000 :  0.0\n",
            "2824000 :  0.0\n",
            "2825000 :  0.0\n",
            "2826000 :  0.0\n",
            "2827000 :  0.0\n",
            "2828000 :  0.0\n",
            "2829000 :  0.0\n",
            "2830000 :  0.0\n",
            "2831000 :  0.0\n",
            "2832000 :  0.0\n",
            "2833000 :  0.0\n",
            "2834000 :  0.0\n",
            "2835000 :  0.0\n",
            "2836000 :  0.0\n",
            "2837000 :  0.0\n",
            "2838000 :  0.0\n",
            "2839000 :  0.0\n",
            "2840000 :  0.0\n",
            "2841000 :  0.0\n",
            "2842000 :  0.0\n",
            "2843000 :  0.0\n",
            "2844000 :  0.0\n",
            "2845000 :  0.0\n",
            "2846000 :  0.0\n",
            "2847000 :  0.0\n",
            "2848000 :  0.0\n",
            "2849000 :  0.0\n",
            "2850000 :  0.0\n",
            "2851000 :  0.0\n",
            "2852000 :  0.0\n",
            "2853000 :  0.0\n",
            "2854000 :  0.0\n",
            "2855000 :  0.0\n",
            "2856000 :  0.0\n",
            "2857000 :  0.0\n",
            "2858000 :  0.0\n",
            "2859000 :  0.0\n",
            "2860000 :  0.0\n",
            "2861000 :  0.0\n",
            "2862000 :  0.0\n",
            "2863000 :  0.0\n",
            "2864000 :  0.0\n",
            "2865000 :  0.0\n",
            "2866000 :  0.0\n",
            "2867000 :  0.0\n",
            "2868000 :  0.0\n",
            "2869000 :  0.0\n",
            "2870000 :  0.0\n",
            "2871000 :  0.0\n",
            "2872000 :  0.0\n",
            "2873000 :  0.0\n",
            "2874000 :  0.0\n",
            "2875000 :  0.0\n",
            "2876000 :  0.0\n",
            "2877000 :  0.0\n",
            "2878000 :  0.0\n",
            "2879000 :  0.0\n",
            "2880000 :  0.0\n",
            "2881000 :  0.0\n",
            "2882000 :  0.0\n",
            "2883000 :  0.0\n",
            "2884000 :  0.0\n",
            "2885000 :  0.0\n",
            "2886000 :  0.0\n",
            "2887000 :  0.0\n",
            "2888000 :  0.0\n",
            "2889000 :  0.0\n",
            "2890000 :  0.0\n",
            "2891000 :  0.0\n",
            "2892000 :  0.0\n",
            "2893000 :  0.0\n",
            "2894000 :  0.0\n",
            "2895000 :  0.0\n",
            "2896000 :  0.0\n",
            "2897000 :  0.0\n",
            "2898000 :  0.0\n",
            "2899000 :  0.0\n",
            "2900000 :  0.0\n",
            "2901000 :  0.0\n",
            "2902000 :  0.0\n",
            "2903000 :  0.0\n",
            "2904000 :  0.0\n",
            "2905000 :  0.0\n",
            "2906000 :  0.0\n",
            "2907000 :  0.0\n",
            "2908000 :  0.0\n",
            "2909000 :  0.0\n",
            "2910000 :  0.0\n",
            "2911000 :  0.0\n",
            "2912000 :  0.0\n",
            "2913000 :  0.0\n",
            "2914000 :  0.0\n",
            "2915000 :  0.0\n",
            "2916000 :  0.0\n",
            "2917000 :  0.0\n",
            "2918000 :  0.0\n",
            "2919000 :  0.0\n",
            "2920000 :  0.0\n",
            "2921000 :  0.0\n",
            "2922000 :  0.0\n",
            "2923000 :  0.0\n",
            "2924000 :  0.0\n",
            "2925000 :  0.0\n",
            "2926000 :  0.0\n",
            "2927000 :  0.0\n",
            "2928000 :  0.0\n",
            "2929000 :  0.0\n",
            "2930000 :  0.0\n",
            "2931000 :  0.0\n",
            "2932000 :  0.0\n",
            "2933000 :  0.0\n",
            "2934000 :  0.0\n",
            "2935000 :  0.0\n",
            "2936000 :  0.0\n",
            "2937000 :  0.0\n",
            "2938000 :  0.0\n",
            "2939000 :  0.0\n",
            "2940000 :  0.0\n",
            "2941000 :  0.0\n",
            "2942000 :  0.0\n",
            "2943000 :  0.0\n",
            "2944000 :  0.0\n",
            "2945000 :  0.0\n",
            "2946000 :  0.0\n",
            "2947000 :  0.0\n",
            "2948000 :  0.0\n",
            "2949000 :  0.0\n",
            "2950000 :  0.0\n",
            "2951000 :  0.0\n",
            "2952000 :  0.0\n",
            "2953000 :  0.0\n",
            "2954000 :  0.0\n",
            "2955000 :  0.0\n",
            "2956000 :  0.0\n",
            "2957000 :  0.0\n",
            "2958000 :  0.0\n",
            "2959000 :  0.0\n",
            "2960000 :  0.0\n",
            "2961000 :  0.0\n",
            "2962000 :  0.0\n",
            "2963000 :  0.0\n",
            "2964000 :  0.0\n",
            "2965000 :  0.0\n",
            "2966000 :  0.0\n",
            "2967000 :  0.0\n",
            "2968000 :  0.0\n",
            "2969000 :  0.0\n",
            "2970000 :  0.0\n",
            "2971000 :  0.0\n",
            "2972000 :  0.0\n",
            "2973000 :  0.0\n",
            "2974000 :  0.0\n",
            "2975000 :  0.0\n",
            "2976000 :  0.0\n",
            "2977000 :  0.0\n",
            "2978000 :  0.0\n",
            "2979000 :  0.0\n",
            "2980000 :  0.0\n",
            "2981000 :  0.0\n",
            "2982000 :  0.0\n",
            "2983000 :  0.0\n",
            "2984000 :  0.0\n",
            "2985000 :  0.0\n",
            "2986000 :  0.0\n",
            "2987000 :  0.0\n",
            "2988000 :  0.0\n",
            "2989000 :  0.0\n",
            "2990000 :  0.0\n",
            "2991000 :  0.0\n",
            "2992000 :  0.0\n",
            "2993000 :  0.0\n",
            "2994000 :  0.0\n",
            "2995000 :  0.0\n",
            "2996000 :  0.0\n",
            "2997000 :  0.0\n",
            "2998000 :  0.0\n",
            "2999000 :  0.0\n",
            "3000000 :  0.0\n",
            "Q table\n",
            "\n",
            "[[0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]\n",
            " [0. 0. 0. 0.]]\n"
          ]
        }
      ],
      "source": [
        "rewards = []\n",
        "exploration_rate_tracker = []\n",
        "\n",
        "for episode in range(total_episodes):\n",
        "    state = env.reset()\n",
        "    step = 0\n",
        "    done = False\n",
        "    total_rewards = 0\n",
        "    \n",
        "    for step in range(max_steps):\n",
        "        exp_exp_tradeoff = random.uniform(0, 1)\n",
        "        if exp_exp_tradeoff > epsilon:\n",
        "            action = np.argmax(q_table[state,:])\n",
        "        else:\n",
        "            action = env.action_space.sample()\n",
        "\n",
        "        new_state, reward, done, info = env.step(action)\n",
        "        temp_reward = reward\n",
        "\n",
        "        q_table[state, action] = q_table[state, action] + learning_rate * (temp_reward + gamma * np.max(q_table[new_state, :]) - q_table[state, action])\n",
        "        \n",
        "        total_rewards += reward\n",
        "        state = new_state\n",
        "        \n",
        "        if done == True: \n",
        "            break\n",
        "    epsilon = min_epsilon + (max_epsilon - min_epsilon)*np.exp(-decay_rate*episode) \n",
        "    rewards.append(total_rewards)\n",
        "    exploration_rate_tracker.append(epsilon)\n",
        "\n",
        "print (\"Score over time: \" +  str(sum(rewards)/total_episodes))\n",
        "print(q_table)\n",
        "\n",
        "plot_epsilon(exploration_rate_tracker, total_episodes)\n",
        "plot_rewards(rewards, total_episodes)\n",
        "\n",
        "rewards_per_thousand_episodes = np.split(np.array(rewards), total_episodes/1000)\n",
        "count = 1000\n",
        "print(\"Avg reward per 1000 episode \\n\")\n",
        "for r in rewards_per_thousand_episodes:\n",
        "  print(count, \": \", str(sum(r/1000)))\n",
        "  count += 1000\n",
        "\n",
        "print(\"Q table\\n\")\n",
        "print(q_table)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "# Agent knowing about the reward at the right most corner"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "id": "ubXjMeA8Uymt"
      },
      "outputs": [],
      "source": [
        "# Values of hyperparameters for 10\n",
        "total_episodes = 3000000\n",
        "learning_rate = 0.8  \n",
        "max_steps = 800\n",
        "gamma = 0.95\n",
        "epsilon = 1.0\n",
        "max_epsilon = 1.0\n",
        "min_epsilon = 0.001\n",
        "decay_rate = 0.000005\n",
        "q_table = np.zeros((state_space_size, action_space_size))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nzLLxXRiVOnx",
        "outputId": "db330d20-8ad8-4f9b-d021-a11a3cd0b622"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Score over time: 0.00011766666666666666\n",
            "[[2.59857091e-01 8.81471535e-02 6.79265912e-02 5.53600300e-02]\n",
            " [4.11953143e-02 4.45509544e-02 4.11030370e-02 4.09923071e-02]\n",
            " [7.94761584e-02 1.91213021e-02 2.02596638e-02 1.96593690e-02]\n",
            " [9.00000000e-03 1.50000000e-02 1.20000000e-02 9.00000000e-03]\n",
            " [1.20000000e-02 2.00000000e-02 1.60000000e-02 1.20000000e-02]\n",
            " [1.50000000e-02 2.50000000e-02 2.00000000e-02 1.50000000e-02]\n",
            " [1.80000000e-02 3.00000000e-02 2.40000000e-02 1.80000000e-02]\n",
            " [2.10000000e-02 3.50000000e-02 2.80000000e-02 2.10000000e-02]\n",
            " [2.40000000e-02 4.00000000e-02 3.20000000e-02 2.40000000e-02]\n",
            " [2.70000000e-02 4.50000000e-02 3.60000000e-02 2.70000000e-02]\n",
            " [3.15718284e-01 8.93295887e-02 9.13887058e-02 4.75667777e-02]\n",
            " [3.30000000e-02 5.50000000e-02 4.40000000e-02 3.30000000e-02]\n",
            " [4.22719704e-02 4.27245608e-02 8.14389807e-02 4.13121929e-02]\n",
            " [3.90000000e-02 6.50000000e-02 5.20000000e-02 3.90000000e-02]\n",
            " [4.20000000e-02 7.00000000e-02 5.60000000e-02 4.20000000e-02]\n",
            " [4.50000000e-02 7.50000000e-02 6.00000000e-02 4.50000000e-02]\n",
            " [4.80000000e-02 8.00000000e-02 6.40000000e-02 4.80000000e-02]\n",
            " [5.10000000e-02 8.50000000e-02 6.80000000e-02 5.10000000e-02]\n",
            " [5.40000000e-02 9.00000000e-02 7.20000000e-02 5.40000000e-02]\n",
            " [5.70000000e-02 9.50000000e-02 7.60000000e-02 5.70000000e-02]\n",
            " [2.64481166e-01 1.13745392e-01 1.23185535e-01 1.20948966e-01]\n",
            " [6.30000000e-02 1.05000000e-01 8.40000000e-02 6.30000000e-02]\n",
            " [6.60000000e-02 1.10000000e-01 8.80000000e-02 6.60000000e-02]\n",
            " [6.90000000e-02 1.15000000e-01 9.20000000e-02 6.90000000e-02]\n",
            " [7.20000000e-02 1.20000000e-01 9.60000000e-02 7.20000000e-02]\n",
            " [7.50000000e-02 1.25000000e-01 1.00000000e-01 7.50000000e-02]\n",
            " [7.80000000e-02 1.30000000e-01 1.04000000e-01 7.80000000e-02]\n",
            " [8.10000000e-02 1.35000000e-01 1.08000000e-01 8.10000000e-02]\n",
            " [8.40000000e-02 1.40000000e-01 1.12000000e-01 8.40000000e-02]\n",
            " [8.70000000e-02 1.45000000e-01 1.16000000e-01 8.70000000e-02]\n",
            " [1.74181473e-01 1.50832039e-01 1.63116314e-01 2.95268922e-01]\n",
            " [1.34834772e-01 2.24900908e-01 8.21936806e-02 8.79890261e-02]\n",
            " [1.45179770e-01 1.63062429e-01 8.75947779e-02 9.21355918e-02]\n",
            " [9.17042347e-02 1.64111854e-01 1.20721582e-01 1.23305280e-01]\n",
            " [1.02000000e-01 1.70000000e-01 1.36000000e-01 1.02000000e-01]\n",
            " [1.05000000e-01 1.75000000e-01 1.40000000e-01 1.05000000e-01]\n",
            " [1.08000000e-01 1.80000000e-01 1.44000000e-01 1.08000000e-01]\n",
            " [1.11000000e-01 1.85000000e-01 1.48000000e-01 1.11000000e-01]\n",
            " [1.14000000e-01 1.90000000e-01 1.52000000e-01 1.14000000e-01]\n",
            " [1.17000000e-01 1.95000000e-01 1.56000000e-01 1.17000000e-01]\n",
            " [1.20000000e-01 2.00000000e-01 1.60000000e-01 1.20000000e-01]\n",
            " [1.51966235e-01 1.52295942e-01 2.11745940e-01 1.57199152e-01]\n",
            " [1.26000000e-01 2.10000000e-01 1.68000000e-01 1.26000000e-01]\n",
            " [1.61630968e-01 1.57500098e-01 3.05037768e-01 1.57500076e-01]\n",
            " [1.32000000e-01 2.20000000e-01 1.76000000e-01 1.32000000e-01]\n",
            " [1.35000000e-01 2.25000000e-01 1.80000000e-01 1.35000000e-01]\n",
            " [1.38000000e-01 2.30000000e-01 1.84000000e-01 1.38000000e-01]\n",
            " [1.41000000e-01 2.35000000e-01 1.88000000e-01 1.41000000e-01]\n",
            " [1.44000000e-01 2.40000000e-01 1.92000000e-01 1.44000000e-01]\n",
            " [1.47000000e-01 2.45000000e-01 1.96000000e-01 1.47000000e-01]\n",
            " [1.96651993e-01 5.46847739e-01 2.33267928e-01 2.23504342e-01]\n",
            " [2.87437714e-01 1.95233244e-01 1.96111103e-01 1.95476408e-01]\n",
            " [1.56000000e-01 2.60000000e-01 2.08000000e-01 1.56000000e-01]\n",
            " [1.96434517e-01 1.96492764e-01 2.18695893e-01 1.95288681e-01]\n",
            " [1.62000000e-01 2.70000000e-01 2.16000000e-01 1.62000000e-01]\n",
            " [1.65000000e-01 2.75000000e-01 2.20000000e-01 1.65000000e-01]\n",
            " [1.68000000e-01 2.80000000e-01 2.24000000e-01 1.68000000e-01]\n",
            " [1.71000000e-01 2.85000000e-01 2.28000000e-01 1.71000000e-01]\n",
            " [1.74000000e-01 2.90000000e-01 2.32000000e-01 1.74000000e-01]\n",
            " [1.77000000e-01 2.95000000e-01 2.36000000e-01 1.77000000e-01]\n",
            " [5.55905699e-01 2.24176190e-01 2.50852708e-01 2.30110265e-01]\n",
            " [1.83000000e-01 3.05000000e-01 2.44000000e-01 1.83000000e-01]\n",
            " [1.86000000e-01 3.10000000e-01 2.48000000e-01 1.86000000e-01]\n",
            " [2.73244113e-01 2.52665988e-01 2.26863862e+00 2.58596258e-01]\n",
            " [2.41504815e-01 2.00277654e+00 3.66826029e-01 2.48948872e-01]\n",
            " [2.90657521e-01 1.49170968e+00 2.99590820e-01 2.78433616e-01]\n",
            " [2.26403939e-01 1.80127447e+00 2.38660782e-01 2.12388446e-01]\n",
            " [2.01000000e-01 3.35000000e-01 2.68000000e-01 2.01000000e-01]\n",
            " [2.04000000e-01 3.40000000e-01 2.72000000e-01 2.04000000e-01]\n",
            " [2.07000000e-01 3.45000000e-01 2.76000000e-01 2.07000000e-01]\n",
            " [2.56346307e-01 2.00391452e-01 3.76540265e-01 2.55828675e-01]\n",
            " [2.13000000e-01 3.55000000e-01 2.84000000e-01 2.13000000e-01]\n",
            " [2.16000000e-01 3.60000000e-01 2.88000000e-01 2.16000000e-01]\n",
            " [2.19000000e-01 3.65000000e-01 2.92000000e-01 2.19000000e-01]\n",
            " [3.13631727e-01 3.08123613e-01 6.22700293e-01 3.01331568e-01]\n",
            " [3.35482130e-01 3.68127825e-01 3.88579109e-01 3.51481534e+00]\n",
            " [3.14996550e-01 4.13085073e-01 3.25515315e-01 3.22642015e-01]\n",
            " [3.36300307e-01 1.36187347e+01 5.59020284e-01 4.18658450e-01]\n",
            " [1.03331588e+00 2.25854846e+01 2.66400000e-01 2.34000000e-01]\n",
            " [2.37000000e-01 4.23340071e+01 3.16000000e-01 2.37000000e-01]\n",
            " [2.40000000e-01 4.00000000e-01 3.20000000e-01 2.40000000e-01]\n",
            " [2.43000000e-01 4.05000000e-01 3.24000000e-01 2.43000000e-01]\n",
            " [2.46000000e-01 4.10000000e-01 3.28000000e-01 2.46000000e-01]\n",
            " [2.49000000e-01 4.15000000e-01 3.32000000e-01 2.49000000e-01]\n",
            " [2.52000000e-01 4.20000000e-01 3.36000000e-01 2.52000000e-01]\n",
            " [2.55000000e-01 4.25000000e-01 3.40000000e-01 2.55000000e-01]\n",
            " [2.58000000e-01 4.30000000e-01 3.44000000e-01 2.58000000e-01]\n",
            " [4.31069905e-01 3.26713720e-01 8.42486838e+00 3.27528458e-01]\n",
            " [1.24501921e+00 6.01105625e-01 3.64400000e-01 1.44726779e+01]\n",
            " [2.67000000e-01 2.03477230e+01 3.56000000e-01 2.67000000e-01]\n",
            " [2.70000000e-01 4.50000000e-01 3.60000000e-01 2.70000000e-01]\n",
            " [2.73000000e-01 4.55000000e-01 3.64000000e-01 2.73000000e-01]\n",
            " [2.76000000e-01 4.60000000e-01 3.68000000e-01 2.76000000e-01]\n",
            " [2.79000000e-01 4.65000000e-01 3.72000000e-01 2.79000000e-01]\n",
            " [2.82000000e-01 4.70000000e-01 3.76000000e-01 2.82000000e-01]\n",
            " [2.85000000e-01 4.75000000e-01 3.80000000e-01 2.85000000e-01]\n",
            " [2.88000000e-01 4.80000000e-01 3.84000000e-01 2.88000000e-01]\n",
            " [2.91000000e-01 3.19097314e-01 3.49985800e-01 1.47076876e+00]\n",
            " [2.94000000e-01 4.90000000e-01 3.92000000e-01 2.94000000e-01]\n",
            " [2.97000000e-01 4.95000000e-01 3.96000000e-01 2.97000000e-01]]\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Avg reward per 1000 episode \n",
            "\n",
            "1000 :  0.0\n",
            "2000 :  0.0\n",
            "3000 :  0.0\n",
            "4000 :  0.0\n",
            "5000 :  0.0\n",
            "6000 :  0.0\n",
            "7000 :  0.0\n",
            "8000 :  0.0\n",
            "9000 :  0.0\n",
            "10000 :  0.0\n",
            "11000 :  0.0\n",
            "12000 :  0.0\n",
            "13000 :  0.0\n",
            "14000 :  0.0\n",
            "15000 :  0.0\n",
            "16000 :  0.0\n",
            "17000 :  0.0\n",
            "18000 :  0.0\n",
            "19000 :  0.0\n",
            "20000 :  0.0\n",
            "21000 :  0.0\n",
            "22000 :  0.0\n",
            "23000 :  0.0\n",
            "24000 :  0.0\n",
            "25000 :  0.0\n",
            "26000 :  0.0\n",
            "27000 :  0.0\n",
            "28000 :  0.0\n",
            "29000 :  0.0\n",
            "30000 :  0.0\n",
            "31000 :  0.0\n",
            "32000 :  0.0\n",
            "33000 :  0.0\n",
            "34000 :  0.0\n",
            "35000 :  0.0\n",
            "36000 :  0.0\n",
            "37000 :  0.0\n",
            "38000 :  0.0\n",
            "39000 :  0.0\n",
            "40000 :  0.0\n",
            "41000 :  0.0\n",
            "42000 :  0.0\n",
            "43000 :  0.0\n",
            "44000 :  0.0\n",
            "45000 :  0.0\n",
            "46000 :  0.0\n",
            "47000 :  0.0\n",
            "48000 :  0.0\n",
            "49000 :  0.0\n",
            "50000 :  0.0\n",
            "51000 :  0.0\n",
            "52000 :  0.0\n",
            "53000 :  0.0\n",
            "54000 :  0.0\n",
            "55000 :  0.0\n",
            "56000 :  0.0\n",
            "57000 :  0.0\n",
            "58000 :  0.0\n",
            "59000 :  0.0\n",
            "60000 :  0.0\n",
            "61000 :  0.0\n",
            "62000 :  0.0\n",
            "63000 :  0.0\n",
            "64000 :  0.0\n",
            "65000 :  0.0\n",
            "66000 :  0.0\n",
            "67000 :  0.0\n",
            "68000 :  0.0\n",
            "69000 :  0.0\n",
            "70000 :  0.0\n",
            "71000 :  0.0\n",
            "72000 :  0.0\n",
            "73000 :  0.0\n",
            "74000 :  0.0\n",
            "75000 :  0.0\n",
            "76000 :  0.0\n",
            "77000 :  0.0\n",
            "78000 :  0.0\n",
            "79000 :  0.0\n",
            "80000 :  0.0\n",
            "81000 :  0.0\n",
            "82000 :  0.0\n",
            "83000 :  0.0\n",
            "84000 :  0.0\n",
            "85000 :  0.0\n",
            "86000 :  0.0\n",
            "87000 :  0.0\n",
            "88000 :  0.0\n",
            "89000 :  0.0\n",
            "90000 :  0.0\n",
            "91000 :  0.0\n",
            "92000 :  0.0\n",
            "93000 :  0.0\n",
            "94000 :  0.0\n",
            "95000 :  0.0\n",
            "96000 :  0.0\n",
            "97000 :  0.0\n",
            "98000 :  0.0\n",
            "99000 :  0.0\n",
            "100000 :  0.0\n",
            "101000 :  0.0\n",
            "102000 :  0.0\n",
            "103000 :  0.0\n",
            "104000 :  0.0\n",
            "105000 :  0.0\n",
            "106000 :  0.0\n",
            "107000 :  0.0\n",
            "108000 :  0.0\n",
            "109000 :  0.0\n",
            "110000 :  0.0\n",
            "111000 :  0.0\n",
            "112000 :  0.0\n",
            "113000 :  0.0\n",
            "114000 :  0.0\n",
            "115000 :  0.0\n",
            "116000 :  0.0\n",
            "117000 :  0.0\n",
            "118000 :  0.0\n",
            "119000 :  0.0\n",
            "120000 :  0.0\n",
            "121000 :  0.0\n",
            "122000 :  0.0\n",
            "123000 :  0.0\n",
            "124000 :  0.0\n",
            "125000 :  0.0\n",
            "126000 :  0.0\n",
            "127000 :  0.0\n",
            "128000 :  0.0\n",
            "129000 :  0.0\n",
            "130000 :  0.0\n",
            "131000 :  0.0\n",
            "132000 :  0.0\n",
            "133000 :  0.0\n",
            "134000 :  0.0\n",
            "135000 :  0.0\n",
            "136000 :  0.0\n",
            "137000 :  0.0\n",
            "138000 :  0.0\n",
            "139000 :  0.0\n",
            "140000 :  0.0\n",
            "141000 :  0.0\n",
            "142000 :  0.0\n",
            "143000 :  0.0\n",
            "144000 :  0.0\n",
            "145000 :  0.0\n",
            "146000 :  0.0\n",
            "147000 :  0.0\n",
            "148000 :  0.0\n",
            "149000 :  0.0\n",
            "150000 :  0.0\n",
            "151000 :  0.0\n",
            "152000 :  0.0\n",
            "153000 :  0.0\n",
            "154000 :  0.0\n",
            "155000 :  0.0\n",
            "156000 :  0.0\n",
            "157000 :  0.0\n",
            "158000 :  0.0\n",
            "159000 :  0.0\n",
            "160000 :  0.0\n",
            "161000 :  0.0\n",
            "162000 :  0.0\n",
            "163000 :  0.0\n",
            "164000 :  0.0\n",
            "165000 :  0.0\n",
            "166000 :  0.0\n",
            "167000 :  0.0\n",
            "168000 :  0.0\n",
            "169000 :  0.0\n",
            "170000 :  0.0\n",
            "171000 :  0.0\n",
            "172000 :  0.0\n",
            "173000 :  0.0\n",
            "174000 :  0.0\n",
            "175000 :  0.0\n",
            "176000 :  0.0\n",
            "177000 :  0.0\n",
            "178000 :  0.0\n",
            "179000 :  0.0\n",
            "180000 :  0.0\n",
            "181000 :  0.0\n",
            "182000 :  0.0\n",
            "183000 :  0.0\n",
            "184000 :  0.0\n",
            "185000 :  0.0\n",
            "186000 :  0.0\n",
            "187000 :  0.0\n",
            "188000 :  0.0\n",
            "189000 :  0.0\n",
            "190000 :  0.0\n",
            "191000 :  0.0\n",
            "192000 :  0.0\n",
            "193000 :  0.0\n",
            "194000 :  0.0\n",
            "195000 :  0.0\n",
            "196000 :  0.0\n",
            "197000 :  0.0\n",
            "198000 :  0.0\n",
            "199000 :  0.0\n",
            "200000 :  0.0\n",
            "201000 :  0.0\n",
            "202000 :  0.0\n",
            "203000 :  0.0\n",
            "204000 :  0.0\n",
            "205000 :  0.0\n",
            "206000 :  0.0\n",
            "207000 :  0.0\n",
            "208000 :  0.0\n",
            "209000 :  0.0\n",
            "210000 :  0.0\n",
            "211000 :  0.0\n",
            "212000 :  0.0\n",
            "213000 :  0.0\n",
            "214000 :  0.0\n",
            "215000 :  0.0\n",
            "216000 :  0.0\n",
            "217000 :  0.0\n",
            "218000 :  0.0\n",
            "219000 :  0.0\n",
            "220000 :  0.0\n",
            "221000 :  0.0\n",
            "222000 :  0.0\n",
            "223000 :  0.0\n",
            "224000 :  0.0\n",
            "225000 :  0.0\n",
            "226000 :  0.0\n",
            "227000 :  0.0\n",
            "228000 :  0.0\n",
            "229000 :  0.0\n",
            "230000 :  0.0\n",
            "231000 :  0.0\n",
            "232000 :  0.0\n",
            "233000 :  0.0\n",
            "234000 :  0.0\n",
            "235000 :  0.0\n",
            "236000 :  0.0\n",
            "237000 :  0.0\n",
            "238000 :  0.0\n",
            "239000 :  0.0\n",
            "240000 :  0.0\n",
            "241000 :  0.0\n",
            "242000 :  0.0\n",
            "243000 :  0.0\n",
            "244000 :  0.0\n",
            "245000 :  0.0\n",
            "246000 :  0.0\n",
            "247000 :  0.0\n",
            "248000 :  0.0\n",
            "249000 :  0.0\n",
            "250000 :  0.0\n",
            "251000 :  0.0\n",
            "252000 :  0.0\n",
            "253000 :  0.0\n",
            "254000 :  0.0\n",
            "255000 :  0.0\n",
            "256000 :  0.0\n",
            "257000 :  0.0\n",
            "258000 :  0.0\n",
            "259000 :  0.0\n",
            "260000 :  0.0\n",
            "261000 :  0.0\n",
            "262000 :  0.0\n",
            "263000 :  0.0\n",
            "264000 :  0.0\n",
            "265000 :  0.0\n",
            "266000 :  0.0\n",
            "267000 :  0.0\n",
            "268000 :  0.0\n",
            "269000 :  0.0\n",
            "270000 :  0.0\n",
            "271000 :  0.0\n",
            "272000 :  0.0\n",
            "273000 :  0.0\n",
            "274000 :  0.0\n",
            "275000 :  0.0\n",
            "276000 :  0.0\n",
            "277000 :  0.0\n",
            "278000 :  0.0\n",
            "279000 :  0.0\n",
            "280000 :  0.0\n",
            "281000 :  0.0\n",
            "282000 :  0.0\n",
            "283000 :  0.0\n",
            "284000 :  0.0\n",
            "285000 :  0.0\n",
            "286000 :  0.0\n",
            "287000 :  0.0\n",
            "288000 :  0.0\n",
            "289000 :  0.0\n",
            "290000 :  0.0\n",
            "291000 :  0.0\n",
            "292000 :  0.0\n",
            "293000 :  0.0\n",
            "294000 :  0.0\n",
            "295000 :  0.0\n",
            "296000 :  0.0\n",
            "297000 :  0.0\n",
            "298000 :  0.0\n",
            "299000 :  0.0\n",
            "300000 :  0.0\n",
            "301000 :  0.0\n",
            "302000 :  0.0\n",
            "303000 :  0.0\n",
            "304000 :  0.0\n",
            "305000 :  0.0\n",
            "306000 :  0.0\n",
            "307000 :  0.0\n",
            "308000 :  0.0\n",
            "309000 :  0.0\n",
            "310000 :  0.0\n",
            "311000 :  0.0\n",
            "312000 :  0.0\n",
            "313000 :  0.0\n",
            "314000 :  0.0\n",
            "315000 :  0.0\n",
            "316000 :  0.0\n",
            "317000 :  0.0\n",
            "318000 :  0.0\n",
            "319000 :  0.0\n",
            "320000 :  0.0\n",
            "321000 :  0.0\n",
            "322000 :  0.0\n",
            "323000 :  0.0\n",
            "324000 :  0.0\n",
            "325000 :  0.0\n",
            "326000 :  0.0\n",
            "327000 :  0.0\n",
            "328000 :  0.0\n",
            "329000 :  0.0\n",
            "330000 :  0.0\n",
            "331000 :  0.0\n",
            "332000 :  0.0\n",
            "333000 :  0.0\n",
            "334000 :  0.0\n",
            "335000 :  0.0\n",
            "336000 :  0.0\n",
            "337000 :  0.0\n",
            "338000 :  0.0\n",
            "339000 :  0.0\n",
            "340000 :  0.0\n",
            "341000 :  0.0\n",
            "342000 :  0.0\n",
            "343000 :  0.0\n",
            "344000 :  0.0\n",
            "345000 :  0.0\n",
            "346000 :  0.0\n",
            "347000 :  0.0\n",
            "348000 :  0.0\n",
            "349000 :  0.0\n",
            "350000 :  0.0\n",
            "351000 :  0.0\n",
            "352000 :  0.0\n",
            "353000 :  0.0\n",
            "354000 :  0.0\n",
            "355000 :  0.0\n",
            "356000 :  0.0\n",
            "357000 :  0.0\n",
            "358000 :  0.0\n",
            "359000 :  0.0\n",
            "360000 :  0.0\n",
            "361000 :  0.0\n",
            "362000 :  0.0\n",
            "363000 :  0.0\n",
            "364000 :  0.0\n",
            "365000 :  0.0\n",
            "366000 :  0.0\n",
            "367000 :  0.0\n",
            "368000 :  0.0\n",
            "369000 :  0.0\n",
            "370000 :  0.0\n",
            "371000 :  0.0\n",
            "372000 :  0.0\n",
            "373000 :  0.0\n",
            "374000 :  0.0\n",
            "375000 :  0.0\n",
            "376000 :  0.0\n",
            "377000 :  0.0\n",
            "378000 :  0.0\n",
            "379000 :  0.0\n",
            "380000 :  0.0\n",
            "381000 :  0.0\n",
            "382000 :  0.0\n",
            "383000 :  0.0\n",
            "384000 :  0.0\n",
            "385000 :  0.0\n",
            "386000 :  0.0\n",
            "387000 :  0.0\n",
            "388000 :  0.0\n",
            "389000 :  0.0\n",
            "390000 :  0.0\n",
            "391000 :  0.0\n",
            "392000 :  0.0\n",
            "393000 :  0.0\n",
            "394000 :  0.0\n",
            "395000 :  0.0\n",
            "396000 :  0.0\n",
            "397000 :  0.0\n",
            "398000 :  0.0\n",
            "399000 :  0.0\n",
            "400000 :  0.0\n",
            "401000 :  0.0\n",
            "402000 :  0.0\n",
            "403000 :  0.0\n",
            "404000 :  0.0\n",
            "405000 :  0.0\n",
            "406000 :  0.0\n",
            "407000 :  0.0\n",
            "408000 :  0.0\n",
            "409000 :  0.0\n",
            "410000 :  0.0\n",
            "411000 :  0.0\n",
            "412000 :  0.0\n",
            "413000 :  0.0\n",
            "414000 :  0.0\n",
            "415000 :  0.0\n",
            "416000 :  0.0\n",
            "417000 :  0.0\n",
            "418000 :  0.0\n",
            "419000 :  0.0\n",
            "420000 :  0.0\n",
            "421000 :  0.0\n",
            "422000 :  0.0\n",
            "423000 :  0.0\n",
            "424000 :  0.0\n",
            "425000 :  0.0\n",
            "426000 :  0.0\n",
            "427000 :  0.0\n",
            "428000 :  0.0\n",
            "429000 :  0.0\n",
            "430000 :  0.0\n",
            "431000 :  0.0\n",
            "432000 :  0.0\n",
            "433000 :  0.0\n",
            "434000 :  0.0\n",
            "435000 :  0.0\n",
            "436000 :  0.0\n",
            "437000 :  0.0\n",
            "438000 :  0.0\n",
            "439000 :  0.0\n",
            "440000 :  0.0\n",
            "441000 :  0.0\n",
            "442000 :  0.0\n",
            "443000 :  0.0\n",
            "444000 :  0.0\n",
            "445000 :  0.0\n",
            "446000 :  0.0\n",
            "447000 :  0.0\n",
            "448000 :  0.0\n",
            "449000 :  0.0\n",
            "450000 :  0.001\n",
            "451000 :  0.0\n",
            "452000 :  0.0\n",
            "453000 :  0.0\n",
            "454000 :  0.0\n",
            "455000 :  0.0\n",
            "456000 :  0.0\n",
            "457000 :  0.0\n",
            "458000 :  0.0\n",
            "459000 :  0.0\n",
            "460000 :  0.0\n",
            "461000 :  0.0\n",
            "462000 :  0.0\n",
            "463000 :  0.0\n",
            "464000 :  0.0\n",
            "465000 :  0.0\n",
            "466000 :  0.0\n",
            "467000 :  0.0\n",
            "468000 :  0.0\n",
            "469000 :  0.0\n",
            "470000 :  0.0\n",
            "471000 :  0.0\n",
            "472000 :  0.0\n",
            "473000 :  0.0\n",
            "474000 :  0.0\n",
            "475000 :  0.0\n",
            "476000 :  0.0\n",
            "477000 :  0.0\n",
            "478000 :  0.0\n",
            "479000 :  0.0\n",
            "480000 :  0.0\n",
            "481000 :  0.0\n",
            "482000 :  0.0\n",
            "483000 :  0.0\n",
            "484000 :  0.0\n",
            "485000 :  0.0\n",
            "486000 :  0.0\n",
            "487000 :  0.0\n",
            "488000 :  0.0\n",
            "489000 :  0.0\n",
            "490000 :  0.0\n",
            "491000 :  0.0\n",
            "492000 :  0.0\n",
            "493000 :  0.0\n",
            "494000 :  0.001\n",
            "495000 :  0.0\n",
            "496000 :  0.0\n",
            "497000 :  0.0\n",
            "498000 :  0.0\n",
            "499000 :  0.0\n",
            "500000 :  0.0\n",
            "501000 :  0.0\n",
            "502000 :  0.0\n",
            "503000 :  0.0\n",
            "504000 :  0.0\n",
            "505000 :  0.0\n",
            "506000 :  0.0\n",
            "507000 :  0.0\n",
            "508000 :  0.0\n",
            "509000 :  0.0\n",
            "510000 :  0.0\n",
            "511000 :  0.0\n",
            "512000 :  0.0\n",
            "513000 :  0.0\n",
            "514000 :  0.0\n",
            "515000 :  0.0\n",
            "516000 :  0.0\n",
            "517000 :  0.0\n",
            "518000 :  0.0\n",
            "519000 :  0.0\n",
            "520000 :  0.0\n",
            "521000 :  0.0\n",
            "522000 :  0.0\n",
            "523000 :  0.0\n",
            "524000 :  0.0\n",
            "525000 :  0.0\n",
            "526000 :  0.0\n",
            "527000 :  0.0\n",
            "528000 :  0.0\n",
            "529000 :  0.0\n",
            "530000 :  0.0\n",
            "531000 :  0.0\n",
            "532000 :  0.0\n",
            "533000 :  0.0\n",
            "534000 :  0.0\n",
            "535000 :  0.0\n",
            "536000 :  0.0\n",
            "537000 :  0.0\n",
            "538000 :  0.0\n",
            "539000 :  0.0\n",
            "540000 :  0.0\n",
            "541000 :  0.0\n",
            "542000 :  0.0\n",
            "543000 :  0.0\n",
            "544000 :  0.0\n",
            "545000 :  0.0\n",
            "546000 :  0.0\n",
            "547000 :  0.0\n",
            "548000 :  0.0\n",
            "549000 :  0.0\n",
            "550000 :  0.0\n",
            "551000 :  0.0\n",
            "552000 :  0.0\n",
            "553000 :  0.0\n",
            "554000 :  0.0\n",
            "555000 :  0.0\n",
            "556000 :  0.0\n",
            "557000 :  0.0\n",
            "558000 :  0.0\n",
            "559000 :  0.0\n",
            "560000 :  0.0\n",
            "561000 :  0.0\n",
            "562000 :  0.0\n",
            "563000 :  0.0\n",
            "564000 :  0.0\n",
            "565000 :  0.0\n",
            "566000 :  0.0\n",
            "567000 :  0.0\n",
            "568000 :  0.0\n",
            "569000 :  0.0\n",
            "570000 :  0.0\n",
            "571000 :  0.0\n",
            "572000 :  0.001\n",
            "573000 :  0.0\n",
            "574000 :  0.0\n",
            "575000 :  0.0\n",
            "576000 :  0.0\n",
            "577000 :  0.0\n",
            "578000 :  0.0\n",
            "579000 :  0.0\n",
            "580000 :  0.0\n",
            "581000 :  0.0\n",
            "582000 :  0.0\n",
            "583000 :  0.0\n",
            "584000 :  0.0\n",
            "585000 :  0.0\n",
            "586000 :  0.0\n",
            "587000 :  0.0\n",
            "588000 :  0.0\n",
            "589000 :  0.0\n",
            "590000 :  0.0\n",
            "591000 :  0.0\n",
            "592000 :  0.0\n",
            "593000 :  0.0\n",
            "594000 :  0.0\n",
            "595000 :  0.0\n",
            "596000 :  0.0\n",
            "597000 :  0.0\n",
            "598000 :  0.0\n",
            "599000 :  0.0\n",
            "600000 :  0.0\n",
            "601000 :  0.0\n",
            "602000 :  0.0\n",
            "603000 :  0.0\n",
            "604000 :  0.0\n",
            "605000 :  0.0\n",
            "606000 :  0.0\n",
            "607000 :  0.0\n",
            "608000 :  0.0\n",
            "609000 :  0.0\n",
            "610000 :  0.0\n",
            "611000 :  0.0\n",
            "612000 :  0.0\n",
            "613000 :  0.0\n",
            "614000 :  0.0\n",
            "615000 :  0.0\n",
            "616000 :  0.0\n",
            "617000 :  0.0\n",
            "618000 :  0.0\n",
            "619000 :  0.0\n",
            "620000 :  0.0\n",
            "621000 :  0.0\n",
            "622000 :  0.0\n",
            "623000 :  0.0\n",
            "624000 :  0.0\n",
            "625000 :  0.0\n",
            "626000 :  0.0\n",
            "627000 :  0.0\n",
            "628000 :  0.0\n",
            "629000 :  0.0\n",
            "630000 :  0.0\n",
            "631000 :  0.0\n",
            "632000 :  0.0\n",
            "633000 :  0.0\n",
            "634000 :  0.0\n",
            "635000 :  0.0\n",
            "636000 :  0.0\n",
            "637000 :  0.0\n",
            "638000 :  0.0\n",
            "639000 :  0.0\n",
            "640000 :  0.0\n",
            "641000 :  0.0\n",
            "642000 :  0.0\n",
            "643000 :  0.0\n",
            "644000 :  0.0\n",
            "645000 :  0.0\n",
            "646000 :  0.0\n",
            "647000 :  0.0\n",
            "648000 :  0.0\n",
            "649000 :  0.0\n",
            "650000 :  0.0\n",
            "651000 :  0.0\n",
            "652000 :  0.0\n",
            "653000 :  0.0\n",
            "654000 :  0.0\n",
            "655000 :  0.0\n",
            "656000 :  0.0\n",
            "657000 :  0.0\n",
            "658000 :  0.0\n",
            "659000 :  0.0\n",
            "660000 :  0.0\n",
            "661000 :  0.0\n",
            "662000 :  0.0\n",
            "663000 :  0.0\n",
            "664000 :  0.0\n",
            "665000 :  0.0\n",
            "666000 :  0.0\n",
            "667000 :  0.0\n",
            "668000 :  0.0\n",
            "669000 :  0.0\n",
            "670000 :  0.0\n",
            "671000 :  0.0\n",
            "672000 :  0.0\n",
            "673000 :  0.0\n",
            "674000 :  0.0\n",
            "675000 :  0.0\n",
            "676000 :  0.0\n",
            "677000 :  0.0\n",
            "678000 :  0.0\n",
            "679000 :  0.0\n",
            "680000 :  0.0\n",
            "681000 :  0.0\n",
            "682000 :  0.0\n",
            "683000 :  0.0\n",
            "684000 :  0.0\n",
            "685000 :  0.0\n",
            "686000 :  0.001\n",
            "687000 :  0.0\n",
            "688000 :  0.0\n",
            "689000 :  0.0\n",
            "690000 :  0.0\n",
            "691000 :  0.0\n",
            "692000 :  0.0\n",
            "693000 :  0.0\n",
            "694000 :  0.0\n",
            "695000 :  0.0\n",
            "696000 :  0.0\n",
            "697000 :  0.0\n",
            "698000 :  0.0\n",
            "699000 :  0.0\n",
            "700000 :  0.0\n",
            "701000 :  0.0\n",
            "702000 :  0.0\n",
            "703000 :  0.0\n",
            "704000 :  0.0\n",
            "705000 :  0.0\n",
            "706000 :  0.0\n",
            "707000 :  0.0\n",
            "708000 :  0.0\n",
            "709000 :  0.0\n",
            "710000 :  0.0\n",
            "711000 :  0.0\n",
            "712000 :  0.0\n",
            "713000 :  0.0\n",
            "714000 :  0.0\n",
            "715000 :  0.0\n",
            "716000 :  0.0\n",
            "717000 :  0.0\n",
            "718000 :  0.0\n",
            "719000 :  0.0\n",
            "720000 :  0.0\n",
            "721000 :  0.0\n",
            "722000 :  0.0\n",
            "723000 :  0.0\n",
            "724000 :  0.0\n",
            "725000 :  0.0\n",
            "726000 :  0.0\n",
            "727000 :  0.0\n",
            "728000 :  0.0\n",
            "729000 :  0.0\n",
            "730000 :  0.0\n",
            "731000 :  0.0\n",
            "732000 :  0.0\n",
            "733000 :  0.0\n",
            "734000 :  0.0\n",
            "735000 :  0.0\n",
            "736000 :  0.0\n",
            "737000 :  0.0\n",
            "738000 :  0.0\n",
            "739000 :  0.0\n",
            "740000 :  0.0\n",
            "741000 :  0.0\n",
            "742000 :  0.0\n",
            "743000 :  0.0\n",
            "744000 :  0.0\n",
            "745000 :  0.0\n",
            "746000 :  0.0\n",
            "747000 :  0.0\n",
            "748000 :  0.0\n",
            "749000 :  0.0\n",
            "750000 :  0.0\n",
            "751000 :  0.0\n",
            "752000 :  0.0\n",
            "753000 :  0.0\n",
            "754000 :  0.0\n",
            "755000 :  0.0\n",
            "756000 :  0.0\n",
            "757000 :  0.0\n",
            "758000 :  0.0\n",
            "759000 :  0.0\n",
            "760000 :  0.0\n",
            "761000 :  0.0\n",
            "762000 :  0.0\n",
            "763000 :  0.0\n",
            "764000 :  0.0\n",
            "765000 :  0.0\n",
            "766000 :  0.0\n",
            "767000 :  0.0\n",
            "768000 :  0.0\n",
            "769000 :  0.0\n",
            "770000 :  0.0\n",
            "771000 :  0.0\n",
            "772000 :  0.0\n",
            "773000 :  0.0\n",
            "774000 :  0.0\n",
            "775000 :  0.0\n",
            "776000 :  0.0\n",
            "777000 :  0.0\n",
            "778000 :  0.0\n",
            "779000 :  0.0\n",
            "780000 :  0.0\n",
            "781000 :  0.0\n",
            "782000 :  0.0\n",
            "783000 :  0.0\n",
            "784000 :  0.0\n",
            "785000 :  0.0\n",
            "786000 :  0.0\n",
            "787000 :  0.0\n",
            "788000 :  0.0\n",
            "789000 :  0.0\n",
            "790000 :  0.0\n",
            "791000 :  0.0\n",
            "792000 :  0.0\n",
            "793000 :  0.0\n",
            "794000 :  0.0\n",
            "795000 :  0.0\n",
            "796000 :  0.0\n",
            "797000 :  0.0\n",
            "798000 :  0.0\n",
            "799000 :  0.0\n",
            "800000 :  0.0\n",
            "801000 :  0.0\n",
            "802000 :  0.0\n",
            "803000 :  0.0\n",
            "804000 :  0.0\n",
            "805000 :  0.0\n",
            "806000 :  0.0\n",
            "807000 :  0.0\n",
            "808000 :  0.0\n",
            "809000 :  0.0\n",
            "810000 :  0.0\n",
            "811000 :  0.0\n",
            "812000 :  0.0\n",
            "813000 :  0.0\n",
            "814000 :  0.0\n",
            "815000 :  0.0\n",
            "816000 :  0.0\n",
            "817000 :  0.0\n",
            "818000 :  0.0\n",
            "819000 :  0.0\n",
            "820000 :  0.0\n",
            "821000 :  0.0\n",
            "822000 :  0.0\n",
            "823000 :  0.0\n",
            "824000 :  0.0\n",
            "825000 :  0.0\n",
            "826000 :  0.0\n",
            "827000 :  0.0\n",
            "828000 :  0.0\n",
            "829000 :  0.0\n",
            "830000 :  0.0\n",
            "831000 :  0.0\n",
            "832000 :  0.0\n",
            "833000 :  0.0\n",
            "834000 :  0.001\n",
            "835000 :  0.0\n",
            "836000 :  0.0\n",
            "837000 :  0.0\n",
            "838000 :  0.0\n",
            "839000 :  0.0\n",
            "840000 :  0.0\n",
            "841000 :  0.0\n",
            "842000 :  0.0\n",
            "843000 :  0.0\n",
            "844000 :  0.0\n",
            "845000 :  0.0\n",
            "846000 :  0.0\n",
            "847000 :  0.0\n",
            "848000 :  0.0\n",
            "849000 :  0.0\n",
            "850000 :  0.0\n",
            "851000 :  0.0\n",
            "852000 :  0.0\n",
            "853000 :  0.0\n",
            "854000 :  0.0\n",
            "855000 :  0.0\n",
            "856000 :  0.0\n",
            "857000 :  0.0\n",
            "858000 :  0.0\n",
            "859000 :  0.0\n",
            "860000 :  0.0\n",
            "861000 :  0.0\n",
            "862000 :  0.0\n",
            "863000 :  0.0\n",
            "864000 :  0.0\n",
            "865000 :  0.0\n",
            "866000 :  0.0\n",
            "867000 :  0.001\n",
            "868000 :  0.0\n",
            "869000 :  0.0\n",
            "870000 :  0.0\n",
            "871000 :  0.0\n",
            "872000 :  0.0\n",
            "873000 :  0.0\n",
            "874000 :  0.0\n",
            "875000 :  0.0\n",
            "876000 :  0.0\n",
            "877000 :  0.0\n",
            "878000 :  0.0\n",
            "879000 :  0.0\n",
            "880000 :  0.0\n",
            "881000 :  0.0\n",
            "882000 :  0.0\n",
            "883000 :  0.001\n",
            "884000 :  0.0\n",
            "885000 :  0.0\n",
            "886000 :  0.0\n",
            "887000 :  0.0\n",
            "888000 :  0.0\n",
            "889000 :  0.0\n",
            "890000 :  0.0\n",
            "891000 :  0.0\n",
            "892000 :  0.0\n",
            "893000 :  0.001\n",
            "894000 :  0.001\n",
            "895000 :  0.0\n",
            "896000 :  0.001\n",
            "897000 :  0.0\n",
            "898000 :  0.0\n",
            "899000 :  0.0\n",
            "900000 :  0.001\n",
            "901000 :  0.0\n",
            "902000 :  0.0\n",
            "903000 :  0.0\n",
            "904000 :  0.0\n",
            "905000 :  0.0\n",
            "906000 :  0.0\n",
            "907000 :  0.0\n",
            "908000 :  0.0\n",
            "909000 :  0.0\n",
            "910000 :  0.0\n",
            "911000 :  0.001\n",
            "912000 :  0.0\n",
            "913000 :  0.0\n",
            "914000 :  0.0\n",
            "915000 :  0.0\n",
            "916000 :  0.001\n",
            "917000 :  0.002\n",
            "918000 :  0.0\n",
            "919000 :  0.0\n",
            "920000 :  0.0\n",
            "921000 :  0.0\n",
            "922000 :  0.0\n",
            "923000 :  0.0\n",
            "924000 :  0.0\n",
            "925000 :  0.0\n",
            "926000 :  0.0\n",
            "927000 :  0.0\n",
            "928000 :  0.0\n",
            "929000 :  0.001\n",
            "930000 :  0.0\n",
            "931000 :  0.0\n",
            "932000 :  0.0\n",
            "933000 :  0.0\n",
            "934000 :  0.001\n",
            "935000 :  0.0\n",
            "936000 :  0.0\n",
            "937000 :  0.0\n",
            "938000 :  0.0\n",
            "939000 :  0.0\n",
            "940000 :  0.001\n",
            "941000 :  0.0\n",
            "942000 :  0.0\n",
            "943000 :  0.0\n",
            "944000 :  0.0\n",
            "945000 :  0.0\n",
            "946000 :  0.0\n",
            "947000 :  0.0\n",
            "948000 :  0.0\n",
            "949000 :  0.0\n",
            "950000 :  0.0\n",
            "951000 :  0.0\n",
            "952000 :  0.0\n",
            "953000 :  0.0\n",
            "954000 :  0.0\n",
            "955000 :  0.0\n",
            "956000 :  0.0\n",
            "957000 :  0.0\n",
            "958000 :  0.0\n",
            "959000 :  0.0\n",
            "960000 :  0.0\n",
            "961000 :  0.0\n",
            "962000 :  0.0\n",
            "963000 :  0.0\n",
            "964000 :  0.0\n",
            "965000 :  0.0\n",
            "966000 :  0.0\n",
            "967000 :  0.0\n",
            "968000 :  0.0\n",
            "969000 :  0.0\n",
            "970000 :  0.0\n",
            "971000 :  0.0\n",
            "972000 :  0.0\n",
            "973000 :  0.0\n",
            "974000 :  0.0\n",
            "975000 :  0.0\n",
            "976000 :  0.0\n",
            "977000 :  0.0\n",
            "978000 :  0.0\n",
            "979000 :  0.0\n",
            "980000 :  0.0\n",
            "981000 :  0.0\n",
            "982000 :  0.0\n",
            "983000 :  0.0\n",
            "984000 :  0.0\n",
            "985000 :  0.0\n",
            "986000 :  0.0\n",
            "987000 :  0.0\n",
            "988000 :  0.001\n",
            "989000 :  0.0\n",
            "990000 :  0.0\n",
            "991000 :  0.0\n",
            "992000 :  0.0\n",
            "993000 :  0.0\n",
            "994000 :  0.0\n",
            "995000 :  0.0\n",
            "996000 :  0.0\n",
            "997000 :  0.0\n",
            "998000 :  0.0\n",
            "999000 :  0.0\n",
            "1000000 :  0.0\n",
            "1001000 :  0.001\n",
            "1002000 :  0.0\n",
            "1003000 :  0.0\n",
            "1004000 :  0.0\n",
            "1005000 :  0.0\n",
            "1006000 :  0.001\n",
            "1007000 :  0.0\n",
            "1008000 :  0.0\n",
            "1009000 :  0.0\n",
            "1010000 :  0.0\n",
            "1011000 :  0.0\n",
            "1012000 :  0.0\n",
            "1013000 :  0.0\n",
            "1014000 :  0.0\n",
            "1015000 :  0.0\n",
            "1016000 :  0.0\n",
            "1017000 :  0.0\n",
            "1018000 :  0.0\n",
            "1019000 :  0.001\n",
            "1020000 :  0.0\n",
            "1021000 :  0.0\n",
            "1022000 :  0.0\n",
            "1023000 :  0.0\n",
            "1024000 :  0.0\n",
            "1025000 :  0.0\n",
            "1026000 :  0.001\n",
            "1027000 :  0.0\n",
            "1028000 :  0.0\n",
            "1029000 :  0.0\n",
            "1030000 :  0.0\n",
            "1031000 :  0.0\n",
            "1032000 :  0.0\n",
            "1033000 :  0.0\n",
            "1034000 :  0.0\n",
            "1035000 :  0.0\n",
            "1036000 :  0.0\n",
            "1037000 :  0.001\n",
            "1038000 :  0.0\n",
            "1039000 :  0.0\n",
            "1040000 :  0.0\n",
            "1041000 :  0.0\n",
            "1042000 :  0.0\n",
            "1043000 :  0.0\n",
            "1044000 :  0.001\n",
            "1045000 :  0.0\n",
            "1046000 :  0.0\n",
            "1047000 :  0.0\n",
            "1048000 :  0.0\n",
            "1049000 :  0.001\n",
            "1050000 :  0.0\n",
            "1051000 :  0.0\n",
            "1052000 :  0.0\n",
            "1053000 :  0.0\n",
            "1054000 :  0.0\n",
            "1055000 :  0.001\n",
            "1056000 :  0.0\n",
            "1057000 :  0.0\n",
            "1058000 :  0.0\n",
            "1059000 :  0.0\n",
            "1060000 :  0.0\n",
            "1061000 :  0.0\n",
            "1062000 :  0.0\n",
            "1063000 :  0.001\n",
            "1064000 :  0.0\n",
            "1065000 :  0.0\n",
            "1066000 :  0.0\n",
            "1067000 :  0.0\n",
            "1068000 :  0.0\n",
            "1069000 :  0.0\n",
            "1070000 :  0.0\n",
            "1071000 :  0.0\n",
            "1072000 :  0.0\n",
            "1073000 :  0.0\n",
            "1074000 :  0.0\n",
            "1075000 :  0.0\n",
            "1076000 :  0.0\n",
            "1077000 :  0.0\n",
            "1078000 :  0.0\n",
            "1079000 :  0.002\n",
            "1080000 :  0.0\n",
            "1081000 :  0.0\n",
            "1082000 :  0.001\n",
            "1083000 :  0.001\n",
            "1084000 :  0.0\n",
            "1085000 :  0.0\n",
            "1086000 :  0.0\n",
            "1087000 :  0.0\n",
            "1088000 :  0.0\n",
            "1089000 :  0.0\n",
            "1090000 :  0.0\n",
            "1091000 :  0.0\n",
            "1092000 :  0.0\n",
            "1093000 :  0.001\n",
            "1094000 :  0.0\n",
            "1095000 :  0.0\n",
            "1096000 :  0.001\n",
            "1097000 :  0.0\n",
            "1098000 :  0.0\n",
            "1099000 :  0.0\n",
            "1100000 :  0.0\n",
            "1101000 :  0.0\n",
            "1102000 :  0.001\n",
            "1103000 :  0.0\n",
            "1104000 :  0.001\n",
            "1105000 :  0.0\n",
            "1106000 :  0.0\n",
            "1107000 :  0.0\n",
            "1108000 :  0.0\n",
            "1109000 :  0.0\n",
            "1110000 :  0.0\n",
            "1111000 :  0.002\n",
            "1112000 :  0.0\n",
            "1113000 :  0.0\n",
            "1114000 :  0.0\n",
            "1115000 :  0.0\n",
            "1116000 :  0.0\n",
            "1117000 :  0.0\n",
            "1118000 :  0.002\n",
            "1119000 :  0.0\n",
            "1120000 :  0.0\n",
            "1121000 :  0.0\n",
            "1122000 :  0.001\n",
            "1123000 :  0.0\n",
            "1124000 :  0.0\n",
            "1125000 :  0.0\n",
            "1126000 :  0.0\n",
            "1127000 :  0.0\n",
            "1128000 :  0.0\n",
            "1129000 :  0.001\n",
            "1130000 :  0.0\n",
            "1131000 :  0.0\n",
            "1132000 :  0.0\n",
            "1133000 :  0.0\n",
            "1134000 :  0.0\n",
            "1135000 :  0.0\n",
            "1136000 :  0.0\n",
            "1137000 :  0.0\n",
            "1138000 :  0.0\n",
            "1139000 :  0.0\n",
            "1140000 :  0.0\n",
            "1141000 :  0.0\n",
            "1142000 :  0.0\n",
            "1143000 :  0.001\n",
            "1144000 :  0.0\n",
            "1145000 :  0.001\n",
            "1146000 :  0.0\n",
            "1147000 :  0.001\n",
            "1148000 :  0.001\n",
            "1149000 :  0.0\n",
            "1150000 :  0.0\n",
            "1151000 :  0.0\n",
            "1152000 :  0.0\n",
            "1153000 :  0.0\n",
            "1154000 :  0.001\n",
            "1155000 :  0.0\n",
            "1156000 :  0.0\n",
            "1157000 :  0.0\n",
            "1158000 :  0.0\n",
            "1159000 :  0.0\n",
            "1160000 :  0.0\n",
            "1161000 :  0.0\n",
            "1162000 :  0.0\n",
            "1163000 :  0.002\n",
            "1164000 :  0.0\n",
            "1165000 :  0.0\n",
            "1166000 :  0.0\n",
            "1167000 :  0.0\n",
            "1168000 :  0.0\n",
            "1169000 :  0.0\n",
            "1170000 :  0.0\n",
            "1171000 :  0.0\n",
            "1172000 :  0.0\n",
            "1173000 :  0.0\n",
            "1174000 :  0.0\n",
            "1175000 :  0.0\n",
            "1176000 :  0.0\n",
            "1177000 :  0.0\n",
            "1178000 :  0.0\n",
            "1179000 :  0.0\n",
            "1180000 :  0.0\n",
            "1181000 :  0.0\n",
            "1182000 :  0.001\n",
            "1183000 :  0.0\n",
            "1184000 :  0.0\n",
            "1185000 :  0.0\n",
            "1186000 :  0.0\n",
            "1187000 :  0.001\n",
            "1188000 :  0.0\n",
            "1189000 :  0.0\n",
            "1190000 :  0.001\n",
            "1191000 :  0.0\n",
            "1192000 :  0.0\n",
            "1193000 :  0.0\n",
            "1194000 :  0.0\n",
            "1195000 :  0.0\n",
            "1196000 :  0.0\n",
            "1197000 :  0.0\n",
            "1198000 :  0.0\n",
            "1199000 :  0.0\n",
            "1200000 :  0.0\n",
            "1201000 :  0.0\n",
            "1202000 :  0.0\n",
            "1203000 :  0.0\n",
            "1204000 :  0.0\n",
            "1205000 :  0.0\n",
            "1206000 :  0.0\n",
            "1207000 :  0.0\n",
            "1208000 :  0.0\n",
            "1209000 :  0.001\n",
            "1210000 :  0.0\n",
            "1211000 :  0.0\n",
            "1212000 :  0.001\n",
            "1213000 :  0.0\n",
            "1214000 :  0.001\n",
            "1215000 :  0.0\n",
            "1216000 :  0.0\n",
            "1217000 :  0.0\n",
            "1218000 :  0.0\n",
            "1219000 :  0.0\n",
            "1220000 :  0.0\n",
            "1221000 :  0.0\n",
            "1222000 :  0.0\n",
            "1223000 :  0.0\n",
            "1224000 :  0.0\n",
            "1225000 :  0.0\n",
            "1226000 :  0.0\n",
            "1227000 :  0.0\n",
            "1228000 :  0.0\n",
            "1229000 :  0.0\n",
            "1230000 :  0.0\n",
            "1231000 :  0.0\n",
            "1232000 :  0.0\n",
            "1233000 :  0.0\n",
            "1234000 :  0.0\n",
            "1235000 :  0.0\n",
            "1236000 :  0.0\n",
            "1237000 :  0.0\n",
            "1238000 :  0.0\n",
            "1239000 :  0.0\n",
            "1240000 :  0.0\n",
            "1241000 :  0.0\n",
            "1242000 :  0.0\n",
            "1243000 :  0.0\n",
            "1244000 :  0.0\n",
            "1245000 :  0.0\n",
            "1246000 :  0.001\n",
            "1247000 :  0.0\n",
            "1248000 :  0.0\n",
            "1249000 :  0.001\n",
            "1250000 :  0.0\n",
            "1251000 :  0.001\n",
            "1252000 :  0.0\n",
            "1253000 :  0.0\n",
            "1254000 :  0.0\n",
            "1255000 :  0.0\n",
            "1256000 :  0.001\n",
            "1257000 :  0.0\n",
            "1258000 :  0.001\n",
            "1259000 :  0.0\n",
            "1260000 :  0.0\n",
            "1261000 :  0.0\n",
            "1262000 :  0.0\n",
            "1263000 :  0.001\n",
            "1264000 :  0.0\n",
            "1265000 :  0.0\n",
            "1266000 :  0.0\n",
            "1267000 :  0.001\n",
            "1268000 :  0.0\n",
            "1269000 :  0.002\n",
            "1270000 :  0.001\n",
            "1271000 :  0.0\n",
            "1272000 :  0.0\n",
            "1273000 :  0.0\n",
            "1274000 :  0.0\n",
            "1275000 :  0.0\n",
            "1276000 :  0.0\n",
            "1277000 :  0.0\n",
            "1278000 :  0.0\n",
            "1279000 :  0.0\n",
            "1280000 :  0.0\n",
            "1281000 :  0.0\n",
            "1282000 :  0.0\n",
            "1283000 :  0.0\n",
            "1284000 :  0.0\n",
            "1285000 :  0.0\n",
            "1286000 :  0.0\n",
            "1287000 :  0.0\n",
            "1288000 :  0.0\n",
            "1289000 :  0.0\n",
            "1290000 :  0.0\n",
            "1291000 :  0.0\n",
            "1292000 :  0.0\n",
            "1293000 :  0.001\n",
            "1294000 :  0.001\n",
            "1295000 :  0.0\n",
            "1296000 :  0.0\n",
            "1297000 :  0.0\n",
            "1298000 :  0.0\n",
            "1299000 :  0.0\n",
            "1300000 :  0.0\n",
            "1301000 :  0.0\n",
            "1302000 :  0.001\n",
            "1303000 :  0.0\n",
            "1304000 :  0.0\n",
            "1305000 :  0.0\n",
            "1306000 :  0.0\n",
            "1307000 :  0.0\n",
            "1308000 :  0.0\n",
            "1309000 :  0.001\n",
            "1310000 :  0.0\n",
            "1311000 :  0.0\n",
            "1312000 :  0.0\n",
            "1313000 :  0.002\n",
            "1314000 :  0.0\n",
            "1315000 :  0.0\n",
            "1316000 :  0.0\n",
            "1317000 :  0.0\n",
            "1318000 :  0.001\n",
            "1319000 :  0.0\n",
            "1320000 :  0.0\n",
            "1321000 :  0.0\n",
            "1322000 :  0.0\n",
            "1323000 :  0.001\n",
            "1324000 :  0.0\n",
            "1325000 :  0.0\n",
            "1326000 :  0.0\n",
            "1327000 :  0.0\n",
            "1328000 :  0.0\n",
            "1329000 :  0.0\n",
            "1330000 :  0.0\n",
            "1331000 :  0.0\n",
            "1332000 :  0.0\n",
            "1333000 :  0.0\n",
            "1334000 :  0.0\n",
            "1335000 :  0.0\n",
            "1336000 :  0.0\n",
            "1337000 :  0.0\n",
            "1338000 :  0.001\n",
            "1339000 :  0.0\n",
            "1340000 :  0.0\n",
            "1341000 :  0.0\n",
            "1342000 :  0.0\n",
            "1343000 :  0.0\n",
            "1344000 :  0.0\n",
            "1345000 :  0.001\n",
            "1346000 :  0.0\n",
            "1347000 :  0.0\n",
            "1348000 :  0.0\n",
            "1349000 :  0.001\n",
            "1350000 :  0.001\n",
            "1351000 :  0.0\n",
            "1352000 :  0.0\n",
            "1353000 :  0.0\n",
            "1354000 :  0.001\n",
            "1355000 :  0.0\n",
            "1356000 :  0.0\n",
            "1357000 :  0.0\n",
            "1358000 :  0.0\n",
            "1359000 :  0.0\n",
            "1360000 :  0.0\n",
            "1361000 :  0.0\n",
            "1362000 :  0.0\n",
            "1363000 :  0.0\n",
            "1364000 :  0.001\n",
            "1365000 :  0.0\n",
            "1366000 :  0.0\n",
            "1367000 :  0.0\n",
            "1368000 :  0.0\n",
            "1369000 :  0.0\n",
            "1370000 :  0.0\n",
            "1371000 :  0.0\n",
            "1372000 :  0.0\n",
            "1373000 :  0.0\n",
            "1374000 :  0.001\n",
            "1375000 :  0.0\n",
            "1376000 :  0.0\n",
            "1377000 :  0.0\n",
            "1378000 :  0.0\n",
            "1379000 :  0.0\n",
            "1380000 :  0.0\n",
            "1381000 :  0.0\n",
            "1382000 :  0.001\n",
            "1383000 :  0.0\n",
            "1384000 :  0.0\n",
            "1385000 :  0.001\n",
            "1386000 :  0.0\n",
            "1387000 :  0.0\n",
            "1388000 :  0.0\n",
            "1389000 :  0.0\n",
            "1390000 :  0.0\n",
            "1391000 :  0.0\n",
            "1392000 :  0.0\n",
            "1393000 :  0.0\n",
            "1394000 :  0.0\n",
            "1395000 :  0.001\n",
            "1396000 :  0.0\n",
            "1397000 :  0.0\n",
            "1398000 :  0.0\n",
            "1399000 :  0.0\n",
            "1400000 :  0.002\n",
            "1401000 :  0.0\n",
            "1402000 :  0.0\n",
            "1403000 :  0.0\n",
            "1404000 :  0.0\n",
            "1405000 :  0.0\n",
            "1406000 :  0.0\n",
            "1407000 :  0.0\n",
            "1408000 :  0.001\n",
            "1409000 :  0.0\n",
            "1410000 :  0.0\n",
            "1411000 :  0.0\n",
            "1412000 :  0.0\n",
            "1413000 :  0.0\n",
            "1414000 :  0.0\n",
            "1415000 :  0.0\n",
            "1416000 :  0.0\n",
            "1417000 :  0.0\n",
            "1418000 :  0.0\n",
            "1419000 :  0.0\n",
            "1420000 :  0.0\n",
            "1421000 :  0.0\n",
            "1422000 :  0.001\n",
            "1423000 :  0.001\n",
            "1424000 :  0.002\n",
            "1425000 :  0.0\n",
            "1426000 :  0.0\n",
            "1427000 :  0.0\n",
            "1428000 :  0.001\n",
            "1429000 :  0.0\n",
            "1430000 :  0.0\n",
            "1431000 :  0.0\n",
            "1432000 :  0.0\n",
            "1433000 :  0.001\n",
            "1434000 :  0.0\n",
            "1435000 :  0.0\n",
            "1436000 :  0.0\n",
            "1437000 :  0.0\n",
            "1438000 :  0.0\n",
            "1439000 :  0.0\n",
            "1440000 :  0.0\n",
            "1441000 :  0.0\n",
            "1442000 :  0.001\n",
            "1443000 :  0.0\n",
            "1444000 :  0.0\n",
            "1445000 :  0.002\n",
            "1446000 :  0.0\n",
            "1447000 :  0.0\n",
            "1448000 :  0.0\n",
            "1449000 :  0.0\n",
            "1450000 :  0.001\n",
            "1451000 :  0.001\n",
            "1452000 :  0.002\n",
            "1453000 :  0.0\n",
            "1454000 :  0.001\n",
            "1455000 :  0.0\n",
            "1456000 :  0.0\n",
            "1457000 :  0.0\n",
            "1458000 :  0.0\n",
            "1459000 :  0.0\n",
            "1460000 :  0.001\n",
            "1461000 :  0.0\n",
            "1462000 :  0.001\n",
            "1463000 :  0.0\n",
            "1464000 :  0.001\n",
            "1465000 :  0.001\n",
            "1466000 :  0.0\n",
            "1467000 :  0.0\n",
            "1468000 :  0.0\n",
            "1469000 :  0.0\n",
            "1470000 :  0.0\n",
            "1471000 :  0.0\n",
            "1472000 :  0.0\n",
            "1473000 :  0.0\n",
            "1474000 :  0.001\n",
            "1475000 :  0.0\n",
            "1476000 :  0.001\n",
            "1477000 :  0.0\n",
            "1478000 :  0.0\n",
            "1479000 :  0.0\n",
            "1480000 :  0.001\n",
            "1481000 :  0.0\n",
            "1482000 :  0.0\n",
            "1483000 :  0.0\n",
            "1484000 :  0.0\n",
            "1485000 :  0.0\n",
            "1486000 :  0.0\n",
            "1487000 :  0.001\n",
            "1488000 :  0.001\n",
            "1489000 :  0.0\n",
            "1490000 :  0.0\n",
            "1491000 :  0.0\n",
            "1492000 :  0.0\n",
            "1493000 :  0.0\n",
            "1494000 :  0.0\n",
            "1495000 :  0.0\n",
            "1496000 :  0.0\n",
            "1497000 :  0.0\n",
            "1498000 :  0.0\n",
            "1499000 :  0.0\n",
            "1500000 :  0.0\n",
            "1501000 :  0.0\n",
            "1502000 :  0.002\n",
            "1503000 :  0.0\n",
            "1504000 :  0.0\n",
            "1505000 :  0.001\n",
            "1506000 :  0.0\n",
            "1507000 :  0.0\n",
            "1508000 :  0.0\n",
            "1509000 :  0.001\n",
            "1510000 :  0.0\n",
            "1511000 :  0.003\n",
            "1512000 :  0.001\n",
            "1513000 :  0.0\n",
            "1514000 :  0.0\n",
            "1515000 :  0.0\n",
            "1516000 :  0.0\n",
            "1517000 :  0.0\n",
            "1518000 :  0.0\n",
            "1519000 :  0.0\n",
            "1520000 :  0.0\n",
            "1521000 :  0.0\n",
            "1522000 :  0.0\n",
            "1523000 :  0.0\n",
            "1524000 :  0.0\n",
            "1525000 :  0.0\n",
            "1526000 :  0.001\n",
            "1527000 :  0.0\n",
            "1528000 :  0.0\n",
            "1529000 :  0.0\n",
            "1530000 :  0.0\n",
            "1531000 :  0.0\n",
            "1532000 :  0.0\n",
            "1533000 :  0.0\n",
            "1534000 :  0.0\n",
            "1535000 :  0.0\n",
            "1536000 :  0.0\n",
            "1537000 :  0.002\n",
            "1538000 :  0.0\n",
            "1539000 :  0.0\n",
            "1540000 :  0.0\n",
            "1541000 :  0.001\n",
            "1542000 :  0.0\n",
            "1543000 :  0.002\n",
            "1544000 :  0.0\n",
            "1545000 :  0.0\n",
            "1546000 :  0.001\n",
            "1547000 :  0.001\n",
            "1548000 :  0.0\n",
            "1549000 :  0.0\n",
            "1550000 :  0.0\n",
            "1551000 :  0.001\n",
            "1552000 :  0.0\n",
            "1553000 :  0.0\n",
            "1554000 :  0.0\n",
            "1555000 :  0.002\n",
            "1556000 :  0.0\n",
            "1557000 :  0.0\n",
            "1558000 :  0.0\n",
            "1559000 :  0.0\n",
            "1560000 :  0.0\n",
            "1561000 :  0.0\n",
            "1562000 :  0.0\n",
            "1563000 :  0.0\n",
            "1564000 :  0.0\n",
            "1565000 :  0.0\n",
            "1566000 :  0.0\n",
            "1567000 :  0.0\n",
            "1568000 :  0.0\n",
            "1569000 :  0.001\n",
            "1570000 :  0.0\n",
            "1571000 :  0.0\n",
            "1572000 :  0.0\n",
            "1573000 :  0.0\n",
            "1574000 :  0.0\n",
            "1575000 :  0.0\n",
            "1576000 :  0.0\n",
            "1577000 :  0.0\n",
            "1578000 :  0.0\n",
            "1579000 :  0.0\n",
            "1580000 :  0.001\n",
            "1581000 :  0.001\n",
            "1582000 :  0.0\n",
            "1583000 :  0.0\n",
            "1584000 :  0.0\n",
            "1585000 :  0.001\n",
            "1586000 :  0.0\n",
            "1587000 :  0.0\n",
            "1588000 :  0.0\n",
            "1589000 :  0.0\n",
            "1590000 :  0.0\n",
            "1591000 :  0.0\n",
            "1592000 :  0.0\n",
            "1593000 :  0.0\n",
            "1594000 :  0.0\n",
            "1595000 :  0.0\n",
            "1596000 :  0.001\n",
            "1597000 :  0.0\n",
            "1598000 :  0.0\n",
            "1599000 :  0.0\n",
            "1600000 :  0.0\n",
            "1601000 :  0.001\n",
            "1602000 :  0.0\n",
            "1603000 :  0.0\n",
            "1604000 :  0.0\n",
            "1605000 :  0.0\n",
            "1606000 :  0.001\n",
            "1607000 :  0.0\n",
            "1608000 :  0.0\n",
            "1609000 :  0.0\n",
            "1610000 :  0.0\n",
            "1611000 :  0.0\n",
            "1612000 :  0.0\n",
            "1613000 :  0.0\n",
            "1614000 :  0.0\n",
            "1615000 :  0.0\n",
            "1616000 :  0.0\n",
            "1617000 :  0.0\n",
            "1618000 :  0.001\n",
            "1619000 :  0.0\n",
            "1620000 :  0.0\n",
            "1621000 :  0.0\n",
            "1622000 :  0.0\n",
            "1623000 :  0.0\n",
            "1624000 :  0.0\n",
            "1625000 :  0.001\n",
            "1626000 :  0.001\n",
            "1627000 :  0.0\n",
            "1628000 :  0.0\n",
            "1629000 :  0.0\n",
            "1630000 :  0.0\n",
            "1631000 :  0.0\n",
            "1632000 :  0.0\n",
            "1633000 :  0.0\n",
            "1634000 :  0.0\n",
            "1635000 :  0.001\n",
            "1636000 :  0.0\n",
            "1637000 :  0.001\n",
            "1638000 :  0.001\n",
            "1639000 :  0.0\n",
            "1640000 :  0.001\n",
            "1641000 :  0.002\n",
            "1642000 :  0.0\n",
            "1643000 :  0.0\n",
            "1644000 :  0.0\n",
            "1645000 :  0.0\n",
            "1646000 :  0.0\n",
            "1647000 :  0.0\n",
            "1648000 :  0.001\n",
            "1649000 :  0.0\n",
            "1650000 :  0.0\n",
            "1651000 :  0.0\n",
            "1652000 :  0.0\n",
            "1653000 :  0.001\n",
            "1654000 :  0.0\n",
            "1655000 :  0.0\n",
            "1656000 :  0.0\n",
            "1657000 :  0.0\n",
            "1658000 :  0.0\n",
            "1659000 :  0.0\n",
            "1660000 :  0.0\n",
            "1661000 :  0.0\n",
            "1662000 :  0.0\n",
            "1663000 :  0.0\n",
            "1664000 :  0.0\n",
            "1665000 :  0.0\n",
            "1666000 :  0.001\n",
            "1667000 :  0.0\n",
            "1668000 :  0.001\n",
            "1669000 :  0.0\n",
            "1670000 :  0.0\n",
            "1671000 :  0.002\n",
            "1672000 :  0.0\n",
            "1673000 :  0.0\n",
            "1674000 :  0.0\n",
            "1675000 :  0.0\n",
            "1676000 :  0.0\n",
            "1677000 :  0.0\n",
            "1678000 :  0.0\n",
            "1679000 :  0.0\n",
            "1680000 :  0.0\n",
            "1681000 :  0.0\n",
            "1682000 :  0.0\n",
            "1683000 :  0.0\n",
            "1684000 :  0.0\n",
            "1685000 :  0.0\n",
            "1686000 :  0.0\n",
            "1687000 :  0.0\n",
            "1688000 :  0.0\n",
            "1689000 :  0.0\n",
            "1690000 :  0.0\n",
            "1691000 :  0.0\n",
            "1692000 :  0.0\n",
            "1693000 :  0.0\n",
            "1694000 :  0.0\n",
            "1695000 :  0.0\n",
            "1696000 :  0.0\n",
            "1697000 :  0.0\n",
            "1698000 :  0.0\n",
            "1699000 :  0.0\n",
            "1700000 :  0.0\n",
            "1701000 :  0.0\n",
            "1702000 :  0.0\n",
            "1703000 :  0.0\n",
            "1704000 :  0.0\n",
            "1705000 :  0.0\n",
            "1706000 :  0.0\n",
            "1707000 :  0.0\n",
            "1708000 :  0.0\n",
            "1709000 :  0.0\n",
            "1710000 :  0.0\n",
            "1711000 :  0.0\n",
            "1712000 :  0.0\n",
            "1713000 :  0.0\n",
            "1714000 :  0.0\n",
            "1715000 :  0.0\n",
            "1716000 :  0.0\n",
            "1717000 :  0.003\n",
            "1718000 :  0.001\n",
            "1719000 :  0.0\n",
            "1720000 :  0.0\n",
            "1721000 :  0.001\n",
            "1722000 :  0.0\n",
            "1723000 :  0.0\n",
            "1724000 :  0.0\n",
            "1725000 :  0.0\n",
            "1726000 :  0.001\n",
            "1727000 :  0.001\n",
            "1728000 :  0.0\n",
            "1729000 :  0.0\n",
            "1730000 :  0.0\n",
            "1731000 :  0.0\n",
            "1732000 :  0.0\n",
            "1733000 :  0.0\n",
            "1734000 :  0.0\n",
            "1735000 :  0.0\n",
            "1736000 :  0.0\n",
            "1737000 :  0.0\n",
            "1738000 :  0.0\n",
            "1739000 :  0.0\n",
            "1740000 :  0.0\n",
            "1741000 :  0.001\n",
            "1742000 :  0.0\n",
            "1743000 :  0.0\n",
            "1744000 :  0.0\n",
            "1745000 :  0.0\n",
            "1746000 :  0.0\n",
            "1747000 :  0.0\n",
            "1748000 :  0.0\n",
            "1749000 :  0.001\n",
            "1750000 :  0.0\n",
            "1751000 :  0.0\n",
            "1752000 :  0.0\n",
            "1753000 :  0.0\n",
            "1754000 :  0.0\n",
            "1755000 :  0.0\n",
            "1756000 :  0.0\n",
            "1757000 :  0.0\n",
            "1758000 :  0.0\n",
            "1759000 :  0.0\n",
            "1760000 :  0.0\n",
            "1761000 :  0.001\n",
            "1762000 :  0.0\n",
            "1763000 :  0.0\n",
            "1764000 :  0.001\n",
            "1765000 :  0.0\n",
            "1766000 :  0.0\n",
            "1767000 :  0.0\n",
            "1768000 :  0.001\n",
            "1769000 :  0.001\n",
            "1770000 :  0.0\n",
            "1771000 :  0.0\n",
            "1772000 :  0.0\n",
            "1773000 :  0.0\n",
            "1774000 :  0.0\n",
            "1775000 :  0.0\n",
            "1776000 :  0.0\n",
            "1777000 :  0.0\n",
            "1778000 :  0.0\n",
            "1779000 :  0.0\n",
            "1780000 :  0.001\n",
            "1781000 :  0.0\n",
            "1782000 :  0.001\n",
            "1783000 :  0.0\n",
            "1784000 :  0.0\n",
            "1785000 :  0.0\n",
            "1786000 :  0.001\n",
            "1787000 :  0.0\n",
            "1788000 :  0.0\n",
            "1789000 :  0.0\n",
            "1790000 :  0.0\n",
            "1791000 :  0.0\n",
            "1792000 :  0.0\n",
            "1793000 :  0.0\n",
            "1794000 :  0.0\n",
            "1795000 :  0.0\n",
            "1796000 :  0.0\n",
            "1797000 :  0.0\n",
            "1798000 :  0.0\n",
            "1799000 :  0.0\n",
            "1800000 :  0.001\n",
            "1801000 :  0.001\n",
            "1802000 :  0.0\n",
            "1803000 :  0.0\n",
            "1804000 :  0.0\n",
            "1805000 :  0.0\n",
            "1806000 :  0.0\n",
            "1807000 :  0.0\n",
            "1808000 :  0.001\n",
            "1809000 :  0.0\n",
            "1810000 :  0.0\n",
            "1811000 :  0.0\n",
            "1812000 :  0.0\n",
            "1813000 :  0.0\n",
            "1814000 :  0.0\n",
            "1815000 :  0.0\n",
            "1816000 :  0.0\n",
            "1817000 :  0.0\n",
            "1818000 :  0.0\n",
            "1819000 :  0.0\n",
            "1820000 :  0.0\n",
            "1821000 :  0.0\n",
            "1822000 :  0.0\n",
            "1823000 :  0.0\n",
            "1824000 :  0.0\n",
            "1825000 :  0.0\n",
            "1826000 :  0.0\n",
            "1827000 :  0.001\n",
            "1828000 :  0.0\n",
            "1829000 :  0.0\n",
            "1830000 :  0.0\n",
            "1831000 :  0.0\n",
            "1832000 :  0.0\n",
            "1833000 :  0.0\n",
            "1834000 :  0.001\n",
            "1835000 :  0.0\n",
            "1836000 :  0.0\n",
            "1837000 :  0.0\n",
            "1838000 :  0.001\n",
            "1839000 :  0.0\n",
            "1840000 :  0.0\n",
            "1841000 :  0.0\n",
            "1842000 :  0.0\n",
            "1843000 :  0.0\n",
            "1844000 :  0.0\n",
            "1845000 :  0.0\n",
            "1846000 :  0.0\n",
            "1847000 :  0.0\n",
            "1848000 :  0.0\n",
            "1849000 :  0.0\n",
            "1850000 :  0.0\n",
            "1851000 :  0.0\n",
            "1852000 :  0.0\n",
            "1853000 :  0.0\n",
            "1854000 :  0.0\n",
            "1855000 :  0.0\n",
            "1856000 :  0.0\n",
            "1857000 :  0.0\n",
            "1858000 :  0.0\n",
            "1859000 :  0.0\n",
            "1860000 :  0.0\n",
            "1861000 :  0.0\n",
            "1862000 :  0.0\n",
            "1863000 :  0.0\n",
            "1864000 :  0.0\n",
            "1865000 :  0.0\n",
            "1866000 :  0.0\n",
            "1867000 :  0.0\n",
            "1868000 :  0.0\n",
            "1869000 :  0.0\n",
            "1870000 :  0.0\n",
            "1871000 :  0.001\n",
            "1872000 :  0.0\n",
            "1873000 :  0.001\n",
            "1874000 :  0.0\n",
            "1875000 :  0.0\n",
            "1876000 :  0.0\n",
            "1877000 :  0.0\n",
            "1878000 :  0.0\n",
            "1879000 :  0.0\n",
            "1880000 :  0.0\n",
            "1881000 :  0.0\n",
            "1882000 :  0.0\n",
            "1883000 :  0.0\n",
            "1884000 :  0.0\n",
            "1885000 :  0.0\n",
            "1886000 :  0.0\n",
            "1887000 :  0.0\n",
            "1888000 :  0.0\n",
            "1889000 :  0.0\n",
            "1890000 :  0.0\n",
            "1891000 :  0.0\n",
            "1892000 :  0.0\n",
            "1893000 :  0.001\n",
            "1894000 :  0.0\n",
            "1895000 :  0.0\n",
            "1896000 :  0.0\n",
            "1897000 :  0.001\n",
            "1898000 :  0.0\n",
            "1899000 :  0.0\n",
            "1900000 :  0.0\n",
            "1901000 :  0.001\n",
            "1902000 :  0.0\n",
            "1903000 :  0.0\n",
            "1904000 :  0.001\n",
            "1905000 :  0.001\n",
            "1906000 :  0.0\n",
            "1907000 :  0.0\n",
            "1908000 :  0.0\n",
            "1909000 :  0.0\n",
            "1910000 :  0.0\n",
            "1911000 :  0.001\n",
            "1912000 :  0.0\n",
            "1913000 :  0.0\n",
            "1914000 :  0.0\n",
            "1915000 :  0.0\n",
            "1916000 :  0.0\n",
            "1917000 :  0.0\n",
            "1918000 :  0.0\n",
            "1919000 :  0.0\n",
            "1920000 :  0.0\n",
            "1921000 :  0.001\n",
            "1922000 :  0.002\n",
            "1923000 :  0.0\n",
            "1924000 :  0.0\n",
            "1925000 :  0.0\n",
            "1926000 :  0.0\n",
            "1927000 :  0.002\n",
            "1928000 :  0.0\n",
            "1929000 :  0.0\n",
            "1930000 :  0.0\n",
            "1931000 :  0.001\n",
            "1932000 :  0.001\n",
            "1933000 :  0.0\n",
            "1934000 :  0.0\n",
            "1935000 :  0.0\n",
            "1936000 :  0.001\n",
            "1937000 :  0.001\n",
            "1938000 :  0.0\n",
            "1939000 :  0.0\n",
            "1940000 :  0.0\n",
            "1941000 :  0.0\n",
            "1942000 :  0.0\n",
            "1943000 :  0.0\n",
            "1944000 :  0.0\n",
            "1945000 :  0.0\n",
            "1946000 :  0.0\n",
            "1947000 :  0.0\n",
            "1948000 :  0.0\n",
            "1949000 :  0.0\n",
            "1950000 :  0.0\n",
            "1951000 :  0.0\n",
            "1952000 :  0.0\n",
            "1953000 :  0.001\n",
            "1954000 :  0.0\n",
            "1955000 :  0.0\n",
            "1956000 :  0.0\n",
            "1957000 :  0.0\n",
            "1958000 :  0.0\n",
            "1959000 :  0.0\n",
            "1960000 :  0.001\n",
            "1961000 :  0.001\n",
            "1962000 :  0.0\n",
            "1963000 :  0.0\n",
            "1964000 :  0.0\n",
            "1965000 :  0.0\n",
            "1966000 :  0.0\n",
            "1967000 :  0.0\n",
            "1968000 :  0.001\n",
            "1969000 :  0.0\n",
            "1970000 :  0.001\n",
            "1971000 :  0.0\n",
            "1972000 :  0.0\n",
            "1973000 :  0.001\n",
            "1974000 :  0.0\n",
            "1975000 :  0.0\n",
            "1976000 :  0.0\n",
            "1977000 :  0.0\n",
            "1978000 :  0.0\n",
            "1979000 :  0.0\n",
            "1980000 :  0.0\n",
            "1981000 :  0.0\n",
            "1982000 :  0.0\n",
            "1983000 :  0.0\n",
            "1984000 :  0.0\n",
            "1985000 :  0.0\n",
            "1986000 :  0.001\n",
            "1987000 :  0.0\n",
            "1988000 :  0.0\n",
            "1989000 :  0.0\n",
            "1990000 :  0.001\n",
            "1991000 :  0.0\n",
            "1992000 :  0.001\n",
            "1993000 :  0.0\n",
            "1994000 :  0.0\n",
            "1995000 :  0.001\n",
            "1996000 :  0.0\n",
            "1997000 :  0.0\n",
            "1998000 :  0.001\n",
            "1999000 :  0.002\n",
            "2000000 :  0.0\n",
            "2001000 :  0.0\n",
            "2002000 :  0.0\n",
            "2003000 :  0.0\n",
            "2004000 :  0.0\n",
            "2005000 :  0.0\n",
            "2006000 :  0.0\n",
            "2007000 :  0.001\n",
            "2008000 :  0.0\n",
            "2009000 :  0.0\n",
            "2010000 :  0.0\n",
            "2011000 :  0.0\n",
            "2012000 :  0.0\n",
            "2013000 :  0.0\n",
            "2014000 :  0.0\n",
            "2015000 :  0.0\n",
            "2016000 :  0.0\n",
            "2017000 :  0.0\n",
            "2018000 :  0.002\n",
            "2019000 :  0.0\n",
            "2020000 :  0.0\n",
            "2021000 :  0.0\n",
            "2022000 :  0.001\n",
            "2023000 :  0.0\n",
            "2024000 :  0.0\n",
            "2025000 :  0.001\n",
            "2026000 :  0.0\n",
            "2027000 :  0.0\n",
            "2028000 :  0.0\n",
            "2029000 :  0.0\n",
            "2030000 :  0.0\n",
            "2031000 :  0.0\n",
            "2032000 :  0.0\n",
            "2033000 :  0.0\n",
            "2034000 :  0.0\n",
            "2035000 :  0.001\n",
            "2036000 :  0.0\n",
            "2037000 :  0.0\n",
            "2038000 :  0.0\n",
            "2039000 :  0.001\n",
            "2040000 :  0.0\n",
            "2041000 :  0.0\n",
            "2042000 :  0.001\n",
            "2043000 :  0.0\n",
            "2044000 :  0.0\n",
            "2045000 :  0.0\n",
            "2046000 :  0.0\n",
            "2047000 :  0.0\n",
            "2048000 :  0.0\n",
            "2049000 :  0.0\n",
            "2050000 :  0.001\n",
            "2051000 :  0.001\n",
            "2052000 :  0.0\n",
            "2053000 :  0.0\n",
            "2054000 :  0.001\n",
            "2055000 :  0.0\n",
            "2056000 :  0.0\n",
            "2057000 :  0.0\n",
            "2058000 :  0.001\n",
            "2059000 :  0.0\n",
            "2060000 :  0.0\n",
            "2061000 :  0.0\n",
            "2062000 :  0.0\n",
            "2063000 :  0.0\n",
            "2064000 :  0.0\n",
            "2065000 :  0.0\n",
            "2066000 :  0.0\n",
            "2067000 :  0.0\n",
            "2068000 :  0.0\n",
            "2069000 :  0.0\n",
            "2070000 :  0.0\n",
            "2071000 :  0.0\n",
            "2072000 :  0.0\n",
            "2073000 :  0.0\n",
            "2074000 :  0.001\n",
            "2075000 :  0.0\n",
            "2076000 :  0.0\n",
            "2077000 :  0.0\n",
            "2078000 :  0.0\n",
            "2079000 :  0.002\n",
            "2080000 :  0.0\n",
            "2081000 :  0.001\n",
            "2082000 :  0.0\n",
            "2083000 :  0.0\n",
            "2084000 :  0.0\n",
            "2085000 :  0.0\n",
            "2086000 :  0.0\n",
            "2087000 :  0.0\n",
            "2088000 :  0.001\n",
            "2089000 :  0.0\n",
            "2090000 :  0.0\n",
            "2091000 :  0.001\n",
            "2092000 :  0.0\n",
            "2093000 :  0.0\n",
            "2094000 :  0.0\n",
            "2095000 :  0.0\n",
            "2096000 :  0.0\n",
            "2097000 :  0.0\n",
            "2098000 :  0.0\n",
            "2099000 :  0.0\n",
            "2100000 :  0.001\n",
            "2101000 :  0.0\n",
            "2102000 :  0.0\n",
            "2103000 :  0.0\n",
            "2104000 :  0.0\n",
            "2105000 :  0.0\n",
            "2106000 :  0.0\n",
            "2107000 :  0.0\n",
            "2108000 :  0.0\n",
            "2109000 :  0.0\n",
            "2110000 :  0.001\n",
            "2111000 :  0.001\n",
            "2112000 :  0.0\n",
            "2113000 :  0.0\n",
            "2114000 :  0.001\n",
            "2115000 :  0.0\n",
            "2116000 :  0.0\n",
            "2117000 :  0.0\n",
            "2118000 :  0.0\n",
            "2119000 :  0.0\n",
            "2120000 :  0.0\n",
            "2121000 :  0.0\n",
            "2122000 :  0.0\n",
            "2123000 :  0.0\n",
            "2124000 :  0.0\n",
            "2125000 :  0.001\n",
            "2126000 :  0.0\n",
            "2127000 :  0.0\n",
            "2128000 :  0.0\n",
            "2129000 :  0.001\n",
            "2130000 :  0.0\n",
            "2131000 :  0.0\n",
            "2132000 :  0.0\n",
            "2133000 :  0.0\n",
            "2134000 :  0.0\n",
            "2135000 :  0.0\n",
            "2136000 :  0.0\n",
            "2137000 :  0.0\n",
            "2138000 :  0.0\n",
            "2139000 :  0.0\n",
            "2140000 :  0.0\n",
            "2141000 :  0.0\n",
            "2142000 :  0.0\n",
            "2143000 :  0.0\n",
            "2144000 :  0.001\n",
            "2145000 :  0.001\n",
            "2146000 :  0.0\n",
            "2147000 :  0.0\n",
            "2148000 :  0.0\n",
            "2149000 :  0.001\n",
            "2150000 :  0.001\n",
            "2151000 :  0.0\n",
            "2152000 :  0.0\n",
            "2153000 :  0.0\n",
            "2154000 :  0.0\n",
            "2155000 :  0.0\n",
            "2156000 :  0.0\n",
            "2157000 :  0.0\n",
            "2158000 :  0.0\n",
            "2159000 :  0.0\n",
            "2160000 :  0.0\n",
            "2161000 :  0.001\n",
            "2162000 :  0.0\n",
            "2163000 :  0.001\n",
            "2164000 :  0.001\n",
            "2165000 :  0.001\n",
            "2166000 :  0.0\n",
            "2167000 :  0.002\n",
            "2168000 :  0.002\n",
            "2169000 :  0.0\n",
            "2170000 :  0.0\n",
            "2171000 :  0.0\n",
            "2172000 :  0.001\n",
            "2173000 :  0.0\n",
            "2174000 :  0.0\n",
            "2175000 :  0.0\n",
            "2176000 :  0.001\n",
            "2177000 :  0.0\n",
            "2178000 :  0.0\n",
            "2179000 :  0.0\n",
            "2180000 :  0.0\n",
            "2181000 :  0.0\n",
            "2182000 :  0.0\n",
            "2183000 :  0.0\n",
            "2184000 :  0.0\n",
            "2185000 :  0.0\n",
            "2186000 :  0.001\n",
            "2187000 :  0.0\n",
            "2188000 :  0.0\n",
            "2189000 :  0.001\n",
            "2190000 :  0.0\n",
            "2191000 :  0.0\n",
            "2192000 :  0.0\n",
            "2193000 :  0.0\n",
            "2194000 :  0.0\n",
            "2195000 :  0.0\n",
            "2196000 :  0.0\n",
            "2197000 :  0.0\n",
            "2198000 :  0.0\n",
            "2199000 :  0.0\n",
            "2200000 :  0.0\n",
            "2201000 :  0.0\n",
            "2202000 :  0.0\n",
            "2203000 :  0.001\n",
            "2204000 :  0.0\n",
            "2205000 :  0.0\n",
            "2206000 :  0.0\n",
            "2207000 :  0.0\n",
            "2208000 :  0.0\n",
            "2209000 :  0.0\n",
            "2210000 :  0.0\n",
            "2211000 :  0.0\n",
            "2212000 :  0.0\n",
            "2213000 :  0.0\n",
            "2214000 :  0.0\n",
            "2215000 :  0.0\n",
            "2216000 :  0.0\n",
            "2217000 :  0.001\n",
            "2218000 :  0.0\n",
            "2219000 :  0.0\n",
            "2220000 :  0.0\n",
            "2221000 :  0.0\n",
            "2222000 :  0.0\n",
            "2223000 :  0.0\n",
            "2224000 :  0.0\n",
            "2225000 :  0.0\n",
            "2226000 :  0.0\n",
            "2227000 :  0.0\n",
            "2228000 :  0.0\n",
            "2229000 :  0.001\n",
            "2230000 :  0.0\n",
            "2231000 :  0.0\n",
            "2232000 :  0.001\n",
            "2233000 :  0.0\n",
            "2234000 :  0.0\n",
            "2235000 :  0.001\n",
            "2236000 :  0.0\n",
            "2237000 :  0.0\n",
            "2238000 :  0.001\n",
            "2239000 :  0.0\n",
            "2240000 :  0.0\n",
            "2241000 :  0.0\n",
            "2242000 :  0.0\n",
            "2243000 :  0.0\n",
            "2244000 :  0.0\n",
            "2245000 :  0.0\n",
            "2246000 :  0.0\n",
            "2247000 :  0.0\n",
            "2248000 :  0.001\n",
            "2249000 :  0.0\n",
            "2250000 :  0.0\n",
            "2251000 :  0.0\n",
            "2252000 :  0.001\n",
            "2253000 :  0.0\n",
            "2254000 :  0.001\n",
            "2255000 :  0.0\n",
            "2256000 :  0.0\n",
            "2257000 :  0.0\n",
            "2258000 :  0.001\n",
            "2259000 :  0.001\n",
            "2260000 :  0.0\n",
            "2261000 :  0.002\n",
            "2262000 :  0.0\n",
            "2263000 :  0.0\n",
            "2264000 :  0.0\n",
            "2265000 :  0.0\n",
            "2266000 :  0.0\n",
            "2267000 :  0.0\n",
            "2268000 :  0.0\n",
            "2269000 :  0.0\n",
            "2270000 :  0.0\n",
            "2271000 :  0.0\n",
            "2272000 :  0.0\n",
            "2273000 :  0.001\n",
            "2274000 :  0.0\n",
            "2275000 :  0.0\n",
            "2276000 :  0.001\n",
            "2277000 :  0.0\n",
            "2278000 :  0.0\n",
            "2279000 :  0.0\n",
            "2280000 :  0.0\n",
            "2281000 :  0.0\n",
            "2282000 :  0.0\n",
            "2283000 :  0.001\n",
            "2284000 :  0.0\n",
            "2285000 :  0.0\n",
            "2286000 :  0.0\n",
            "2287000 :  0.0\n",
            "2288000 :  0.0\n",
            "2289000 :  0.0\n",
            "2290000 :  0.0\n",
            "2291000 :  0.0\n",
            "2292000 :  0.0\n",
            "2293000 :  0.0\n",
            "2294000 :  0.0\n",
            "2295000 :  0.0\n",
            "2296000 :  0.001\n",
            "2297000 :  0.001\n",
            "2298000 :  0.0\n",
            "2299000 :  0.0\n",
            "2300000 :  0.0\n",
            "2301000 :  0.0\n",
            "2302000 :  0.0\n",
            "2303000 :  0.0\n",
            "2304000 :  0.0\n",
            "2305000 :  0.0\n",
            "2306000 :  0.0\n",
            "2307000 :  0.0\n",
            "2308000 :  0.0\n",
            "2309000 :  0.0\n",
            "2310000 :  0.0\n",
            "2311000 :  0.0\n",
            "2312000 :  0.0\n",
            "2313000 :  0.0\n",
            "2314000 :  0.0\n",
            "2315000 :  0.0\n",
            "2316000 :  0.0\n",
            "2317000 :  0.0\n",
            "2318000 :  0.0\n",
            "2319000 :  0.001\n",
            "2320000 :  0.0\n",
            "2321000 :  0.0\n",
            "2322000 :  0.0\n",
            "2323000 :  0.0\n",
            "2324000 :  0.0\n",
            "2325000 :  0.0\n",
            "2326000 :  0.0\n",
            "2327000 :  0.0\n",
            "2328000 :  0.0\n",
            "2329000 :  0.0\n",
            "2330000 :  0.0\n",
            "2331000 :  0.0\n",
            "2332000 :  0.0\n",
            "2333000 :  0.0\n",
            "2334000 :  0.0\n",
            "2335000 :  0.0\n",
            "2336000 :  0.0\n",
            "2337000 :  0.0\n",
            "2338000 :  0.001\n",
            "2339000 :  0.0\n",
            "2340000 :  0.0\n",
            "2341000 :  0.0\n",
            "2342000 :  0.0\n",
            "2343000 :  0.0\n",
            "2344000 :  0.0\n",
            "2345000 :  0.001\n",
            "2346000 :  0.0\n",
            "2347000 :  0.0\n",
            "2348000 :  0.0\n",
            "2349000 :  0.0\n",
            "2350000 :  0.0\n",
            "2351000 :  0.0\n",
            "2352000 :  0.0\n",
            "2353000 :  0.0\n",
            "2354000 :  0.0\n",
            "2355000 :  0.0\n",
            "2356000 :  0.0\n",
            "2357000 :  0.0\n",
            "2358000 :  0.0\n",
            "2359000 :  0.0\n",
            "2360000 :  0.0\n",
            "2361000 :  0.0\n",
            "2362000 :  0.0\n",
            "2363000 :  0.0\n",
            "2364000 :  0.001\n",
            "2365000 :  0.0\n",
            "2366000 :  0.0\n",
            "2367000 :  0.001\n",
            "2368000 :  0.0\n",
            "2369000 :  0.0\n",
            "2370000 :  0.0\n",
            "2371000 :  0.0\n",
            "2372000 :  0.0\n",
            "2373000 :  0.0\n",
            "2374000 :  0.001\n",
            "2375000 :  0.002\n",
            "2376000 :  0.0\n",
            "2377000 :  0.0\n",
            "2378000 :  0.0\n",
            "2379000 :  0.0\n",
            "2380000 :  0.002\n",
            "2381000 :  0.0\n",
            "2382000 :  0.0\n",
            "2383000 :  0.0\n",
            "2384000 :  0.0\n",
            "2385000 :  0.0\n",
            "2386000 :  0.001\n",
            "2387000 :  0.001\n",
            "2388000 :  0.0\n",
            "2389000 :  0.001\n",
            "2390000 :  0.0\n",
            "2391000 :  0.0\n",
            "2392000 :  0.001\n",
            "2393000 :  0.0\n",
            "2394000 :  0.0\n",
            "2395000 :  0.0\n",
            "2396000 :  0.0\n",
            "2397000 :  0.0\n",
            "2398000 :  0.0\n",
            "2399000 :  0.0\n",
            "2400000 :  0.001\n",
            "2401000 :  0.0\n",
            "2402000 :  0.0\n",
            "2403000 :  0.0\n",
            "2404000 :  0.0\n",
            "2405000 :  0.001\n",
            "2406000 :  0.0\n",
            "2407000 :  0.0\n",
            "2408000 :  0.0\n",
            "2409000 :  0.0\n",
            "2410000 :  0.0\n",
            "2411000 :  0.001\n",
            "2412000 :  0.0\n",
            "2413000 :  0.0\n",
            "2414000 :  0.0\n",
            "2415000 :  0.0\n",
            "2416000 :  0.0\n",
            "2417000 :  0.0\n",
            "2418000 :  0.0\n",
            "2419000 :  0.0\n",
            "2420000 :  0.001\n",
            "2421000 :  0.001\n",
            "2422000 :  0.0\n",
            "2423000 :  0.0\n",
            "2424000 :  0.0\n",
            "2425000 :  0.0\n",
            "2426000 :  0.0\n",
            "2427000 :  0.0\n",
            "2428000 :  0.0\n",
            "2429000 :  0.0\n",
            "2430000 :  0.0\n",
            "2431000 :  0.0\n",
            "2432000 :  0.0\n",
            "2433000 :  0.0\n",
            "2434000 :  0.0\n",
            "2435000 :  0.0\n",
            "2436000 :  0.0\n",
            "2437000 :  0.0\n",
            "2438000 :  0.0\n",
            "2439000 :  0.0\n",
            "2440000 :  0.0\n",
            "2441000 :  0.0\n",
            "2442000 :  0.0\n",
            "2443000 :  0.0\n",
            "2444000 :  0.0\n",
            "2445000 :  0.002\n",
            "2446000 :  0.0\n",
            "2447000 :  0.0\n",
            "2448000 :  0.001\n",
            "2449000 :  0.001\n",
            "2450000 :  0.0\n",
            "2451000 :  0.0\n",
            "2452000 :  0.0\n",
            "2453000 :  0.0\n",
            "2454000 :  0.0\n",
            "2455000 :  0.0\n",
            "2456000 :  0.0\n",
            "2457000 :  0.001\n",
            "2458000 :  0.0\n",
            "2459000 :  0.0\n",
            "2460000 :  0.0\n",
            "2461000 :  0.0\n",
            "2462000 :  0.0\n",
            "2463000 :  0.0\n",
            "2464000 :  0.0\n",
            "2465000 :  0.0\n",
            "2466000 :  0.0\n",
            "2467000 :  0.0\n",
            "2468000 :  0.0\n",
            "2469000 :  0.0\n",
            "2470000 :  0.0\n",
            "2471000 :  0.0\n",
            "2472000 :  0.0\n",
            "2473000 :  0.0\n",
            "2474000 :  0.0\n",
            "2475000 :  0.0\n",
            "2476000 :  0.0\n",
            "2477000 :  0.0\n",
            "2478000 :  0.001\n",
            "2479000 :  0.0\n",
            "2480000 :  0.0\n",
            "2481000 :  0.002\n",
            "2482000 :  0.0\n",
            "2483000 :  0.0\n",
            "2484000 :  0.0\n",
            "2485000 :  0.0\n",
            "2486000 :  0.0\n",
            "2487000 :  0.0\n",
            "2488000 :  0.0\n",
            "2489000 :  0.0\n",
            "2490000 :  0.0\n",
            "2491000 :  0.001\n",
            "2492000 :  0.0\n",
            "2493000 :  0.0\n",
            "2494000 :  0.0\n",
            "2495000 :  0.0\n",
            "2496000 :  0.0\n",
            "2497000 :  0.0\n",
            "2498000 :  0.0\n",
            "2499000 :  0.0\n",
            "2500000 :  0.0\n",
            "2501000 :  0.0\n",
            "2502000 :  0.0\n",
            "2503000 :  0.0\n",
            "2504000 :  0.0\n",
            "2505000 :  0.0\n",
            "2506000 :  0.0\n",
            "2507000 :  0.0\n",
            "2508000 :  0.0\n",
            "2509000 :  0.0\n",
            "2510000 :  0.0\n",
            "2511000 :  0.0\n",
            "2512000 :  0.0\n",
            "2513000 :  0.0\n",
            "2514000 :  0.0\n",
            "2515000 :  0.0\n",
            "2516000 :  0.0\n",
            "2517000 :  0.0\n",
            "2518000 :  0.0\n",
            "2519000 :  0.0\n",
            "2520000 :  0.0\n",
            "2521000 :  0.0\n",
            "2522000 :  0.0\n",
            "2523000 :  0.0\n",
            "2524000 :  0.0\n",
            "2525000 :  0.0\n",
            "2526000 :  0.0\n",
            "2527000 :  0.0\n",
            "2528000 :  0.0\n",
            "2529000 :  0.001\n",
            "2530000 :  0.001\n",
            "2531000 :  0.0\n",
            "2532000 :  0.0\n",
            "2533000 :  0.0\n",
            "2534000 :  0.0\n",
            "2535000 :  0.001\n",
            "2536000 :  0.0\n",
            "2537000 :  0.0\n",
            "2538000 :  0.0\n",
            "2539000 :  0.0\n",
            "2540000 :  0.0\n",
            "2541000 :  0.001\n",
            "2542000 :  0.0\n",
            "2543000 :  0.002\n",
            "2544000 :  0.0\n",
            "2545000 :  0.0\n",
            "2546000 :  0.0\n",
            "2547000 :  0.0\n",
            "2548000 :  0.0\n",
            "2549000 :  0.0\n",
            "2550000 :  0.0\n",
            "2551000 :  0.0\n",
            "2552000 :  0.0\n",
            "2553000 :  0.0\n",
            "2554000 :  0.0\n",
            "2555000 :  0.0\n",
            "2556000 :  0.0\n",
            "2557000 :  0.0\n",
            "2558000 :  0.0\n",
            "2559000 :  0.0\n",
            "2560000 :  0.0\n",
            "2561000 :  0.0\n",
            "2562000 :  0.001\n",
            "2563000 :  0.0\n",
            "2564000 :  0.0\n",
            "2565000 :  0.0\n",
            "2566000 :  0.0\n",
            "2567000 :  0.0\n",
            "2568000 :  0.0\n",
            "2569000 :  0.0\n",
            "2570000 :  0.0\n",
            "2571000 :  0.0\n",
            "2572000 :  0.0\n",
            "2573000 :  0.0\n",
            "2574000 :  0.0\n",
            "2575000 :  0.0\n",
            "2576000 :  0.0\n",
            "2577000 :  0.0\n",
            "2578000 :  0.0\n",
            "2579000 :  0.0\n",
            "2580000 :  0.0\n",
            "2581000 :  0.0\n",
            "2582000 :  0.0\n",
            "2583000 :  0.0\n",
            "2584000 :  0.0\n",
            "2585000 :  0.0\n",
            "2586000 :  0.0\n",
            "2587000 :  0.0\n",
            "2588000 :  0.0\n",
            "2589000 :  0.0\n",
            "2590000 :  0.0\n",
            "2591000 :  0.0\n",
            "2592000 :  0.0\n",
            "2593000 :  0.0\n",
            "2594000 :  0.0\n",
            "2595000 :  0.0\n",
            "2596000 :  0.0\n",
            "2597000 :  0.0\n",
            "2598000 :  0.0\n",
            "2599000 :  0.0\n",
            "2600000 :  0.0\n",
            "2601000 :  0.0\n",
            "2602000 :  0.0\n",
            "2603000 :  0.001\n",
            "2604000 :  0.0\n",
            "2605000 :  0.0\n",
            "2606000 :  0.0\n",
            "2607000 :  0.0\n",
            "2608000 :  0.0\n",
            "2609000 :  0.0\n",
            "2610000 :  0.001\n",
            "2611000 :  0.0\n",
            "2612000 :  0.0\n",
            "2613000 :  0.0\n",
            "2614000 :  0.001\n",
            "2615000 :  0.0\n",
            "2616000 :  0.001\n",
            "2617000 :  0.001\n",
            "2618000 :  0.0\n",
            "2619000 :  0.0\n",
            "2620000 :  0.001\n",
            "2621000 :  0.0\n",
            "2622000 :  0.0\n",
            "2623000 :  0.0\n",
            "2624000 :  0.0\n",
            "2625000 :  0.0\n",
            "2626000 :  0.001\n",
            "2627000 :  0.0\n",
            "2628000 :  0.0\n",
            "2629000 :  0.0\n",
            "2630000 :  0.0\n",
            "2631000 :  0.0\n",
            "2632000 :  0.0\n",
            "2633000 :  0.0\n",
            "2634000 :  0.0\n",
            "2635000 :  0.002\n",
            "2636000 :  0.0\n",
            "2637000 :  0.0\n",
            "2638000 :  0.0\n",
            "2639000 :  0.0\n",
            "2640000 :  0.0\n",
            "2641000 :  0.0\n",
            "2642000 :  0.0\n",
            "2643000 :  0.0\n",
            "2644000 :  0.0\n",
            "2645000 :  0.0\n",
            "2646000 :  0.0\n",
            "2647000 :  0.0\n",
            "2648000 :  0.0\n",
            "2649000 :  0.0\n",
            "2650000 :  0.0\n",
            "2651000 :  0.001\n",
            "2652000 :  0.0\n",
            "2653000 :  0.0\n",
            "2654000 :  0.0\n",
            "2655000 :  0.0\n",
            "2656000 :  0.0\n",
            "2657000 :  0.001\n",
            "2658000 :  0.001\n",
            "2659000 :  0.0\n",
            "2660000 :  0.0\n",
            "2661000 :  0.0\n",
            "2662000 :  0.0\n",
            "2663000 :  0.0\n",
            "2664000 :  0.0\n",
            "2665000 :  0.0\n",
            "2666000 :  0.0\n",
            "2667000 :  0.0\n",
            "2668000 :  0.0\n",
            "2669000 :  0.0\n",
            "2670000 :  0.0\n",
            "2671000 :  0.001\n",
            "2672000 :  0.0\n",
            "2673000 :  0.0\n",
            "2674000 :  0.0\n",
            "2675000 :  0.0\n",
            "2676000 :  0.0\n",
            "2677000 :  0.0\n",
            "2678000 :  0.0\n",
            "2679000 :  0.0\n",
            "2680000 :  0.0\n",
            "2681000 :  0.0\n",
            "2682000 :  0.001\n",
            "2683000 :  0.0\n",
            "2684000 :  0.001\n",
            "2685000 :  0.001\n",
            "2686000 :  0.0\n",
            "2687000 :  0.0\n",
            "2688000 :  0.0\n",
            "2689000 :  0.0\n",
            "2690000 :  0.0\n",
            "2691000 :  0.0\n",
            "2692000 :  0.0\n",
            "2693000 :  0.0\n",
            "2694000 :  0.0\n",
            "2695000 :  0.0\n",
            "2696000 :  0.0\n",
            "2697000 :  0.0\n",
            "2698000 :  0.001\n",
            "2699000 :  0.001\n",
            "2700000 :  0.0\n",
            "2701000 :  0.0\n",
            "2702000 :  0.0\n",
            "2703000 :  0.0\n",
            "2704000 :  0.0\n",
            "2705000 :  0.001\n",
            "2706000 :  0.0\n",
            "2707000 :  0.0\n",
            "2708000 :  0.0\n",
            "2709000 :  0.0\n",
            "2710000 :  0.0\n",
            "2711000 :  0.0\n",
            "2712000 :  0.0\n",
            "2713000 :  0.0\n",
            "2714000 :  0.0\n",
            "2715000 :  0.0\n",
            "2716000 :  0.001\n",
            "2717000 :  0.0\n",
            "2718000 :  0.0\n",
            "2719000 :  0.0\n",
            "2720000 :  0.001\n",
            "2721000 :  0.001\n",
            "2722000 :  0.0\n",
            "2723000 :  0.0\n",
            "2724000 :  0.001\n",
            "2725000 :  0.001\n",
            "2726000 :  0.0\n",
            "2727000 :  0.001\n",
            "2728000 :  0.0\n",
            "2729000 :  0.0\n",
            "2730000 :  0.0\n",
            "2731000 :  0.0\n",
            "2732000 :  0.0\n",
            "2733000 :  0.0\n",
            "2734000 :  0.0\n",
            "2735000 :  0.001\n",
            "2736000 :  0.0\n",
            "2737000 :  0.001\n",
            "2738000 :  0.0\n",
            "2739000 :  0.0\n",
            "2740000 :  0.0\n",
            "2741000 :  0.0\n",
            "2742000 :  0.0\n",
            "2743000 :  0.0\n",
            "2744000 :  0.0\n",
            "2745000 :  0.0\n",
            "2746000 :  0.0\n",
            "2747000 :  0.0\n",
            "2748000 :  0.0\n",
            "2749000 :  0.0\n",
            "2750000 :  0.0\n",
            "2751000 :  0.0\n",
            "2752000 :  0.0\n",
            "2753000 :  0.0\n",
            "2754000 :  0.0\n",
            "2755000 :  0.0\n",
            "2756000 :  0.0\n",
            "2757000 :  0.0\n",
            "2758000 :  0.0\n",
            "2759000 :  0.0\n",
            "2760000 :  0.0\n",
            "2761000 :  0.0\n",
            "2762000 :  0.0\n",
            "2763000 :  0.0\n",
            "2764000 :  0.0\n",
            "2765000 :  0.0\n",
            "2766000 :  0.0\n",
            "2767000 :  0.0\n",
            "2768000 :  0.0\n",
            "2769000 :  0.0\n",
            "2770000 :  0.0\n",
            "2771000 :  0.0\n",
            "2772000 :  0.0\n",
            "2773000 :  0.0\n",
            "2774000 :  0.0\n",
            "2775000 :  0.0\n",
            "2776000 :  0.0\n",
            "2777000 :  0.0\n",
            "2778000 :  0.0\n",
            "2779000 :  0.0\n",
            "2780000 :  0.0\n",
            "2781000 :  0.0\n",
            "2782000 :  0.0\n",
            "2783000 :  0.0\n",
            "2784000 :  0.0\n",
            "2785000 :  0.0\n",
            "2786000 :  0.0\n",
            "2787000 :  0.0\n",
            "2788000 :  0.001\n",
            "2789000 :  0.0\n",
            "2790000 :  0.0\n",
            "2791000 :  0.0\n",
            "2792000 :  0.0\n",
            "2793000 :  0.0\n",
            "2794000 :  0.001\n",
            "2795000 :  0.0\n",
            "2796000 :  0.0\n",
            "2797000 :  0.0\n",
            "2798000 :  0.0\n",
            "2799000 :  0.0\n",
            "2800000 :  0.0\n",
            "2801000 :  0.001\n",
            "2802000 :  0.0\n",
            "2803000 :  0.0\n",
            "2804000 :  0.0\n",
            "2805000 :  0.0\n",
            "2806000 :  0.0\n",
            "2807000 :  0.0\n",
            "2808000 :  0.0\n",
            "2809000 :  0.0\n",
            "2810000 :  0.0\n",
            "2811000 :  0.0\n",
            "2812000 :  0.0\n",
            "2813000 :  0.0\n",
            "2814000 :  0.001\n",
            "2815000 :  0.0\n",
            "2816000 :  0.0\n",
            "2817000 :  0.0\n",
            "2818000 :  0.0\n",
            "2819000 :  0.0\n",
            "2820000 :  0.0\n",
            "2821000 :  0.0\n",
            "2822000 :  0.0\n",
            "2823000 :  0.0\n",
            "2824000 :  0.001\n",
            "2825000 :  0.0\n",
            "2826000 :  0.0\n",
            "2827000 :  0.0\n",
            "2828000 :  0.0\n",
            "2829000 :  0.0\n",
            "2830000 :  0.0\n",
            "2831000 :  0.0\n",
            "2832000 :  0.0\n",
            "2833000 :  0.0\n",
            "2834000 :  0.001\n",
            "2835000 :  0.0\n",
            "2836000 :  0.0\n",
            "2837000 :  0.001\n",
            "2838000 :  0.0\n",
            "2839000 :  0.001\n",
            "2840000 :  0.0\n",
            "2841000 :  0.0\n",
            "2842000 :  0.001\n",
            "2843000 :  0.0\n",
            "2844000 :  0.001\n",
            "2845000 :  0.001\n",
            "2846000 :  0.001\n",
            "2847000 :  0.0\n",
            "2848000 :  0.0\n",
            "2849000 :  0.0\n",
            "2850000 :  0.0\n",
            "2851000 :  0.0\n",
            "2852000 :  0.0\n",
            "2853000 :  0.0\n",
            "2854000 :  0.0\n",
            "2855000 :  0.0\n",
            "2856000 :  0.0\n",
            "2857000 :  0.0\n",
            "2858000 :  0.0\n",
            "2859000 :  0.0\n",
            "2860000 :  0.0\n",
            "2861000 :  0.002\n",
            "2862000 :  0.0\n",
            "2863000 :  0.0\n",
            "2864000 :  0.001\n",
            "2865000 :  0.0\n",
            "2866000 :  0.0\n",
            "2867000 :  0.0\n",
            "2868000 :  0.0\n",
            "2869000 :  0.0\n",
            "2870000 :  0.0\n",
            "2871000 :  0.0\n",
            "2872000 :  0.0\n",
            "2873000 :  0.001\n",
            "2874000 :  0.0\n",
            "2875000 :  0.0\n",
            "2876000 :  0.001\n",
            "2877000 :  0.0\n",
            "2878000 :  0.0\n",
            "2879000 :  0.0\n",
            "2880000 :  0.0\n",
            "2881000 :  0.0\n",
            "2882000 :  0.0\n",
            "2883000 :  0.0\n",
            "2884000 :  0.0\n",
            "2885000 :  0.0\n",
            "2886000 :  0.0\n",
            "2887000 :  0.001\n",
            "2888000 :  0.0\n",
            "2889000 :  0.0\n",
            "2890000 :  0.0\n",
            "2891000 :  0.0\n",
            "2892000 :  0.0\n",
            "2893000 :  0.0\n",
            "2894000 :  0.0\n",
            "2895000 :  0.0\n",
            "2896000 :  0.0\n",
            "2897000 :  0.0\n",
            "2898000 :  0.0\n",
            "2899000 :  0.0\n",
            "2900000 :  0.0\n",
            "2901000 :  0.0\n",
            "2902000 :  0.001\n",
            "2903000 :  0.0\n",
            "2904000 :  0.0\n",
            "2905000 :  0.0\n",
            "2906000 :  0.0\n",
            "2907000 :  0.0\n",
            "2908000 :  0.001\n",
            "2909000 :  0.0\n",
            "2910000 :  0.0\n",
            "2911000 :  0.001\n",
            "2912000 :  0.0\n",
            "2913000 :  0.0\n",
            "2914000 :  0.001\n",
            "2915000 :  0.0\n",
            "2916000 :  0.001\n",
            "2917000 :  0.0\n",
            "2918000 :  0.001\n",
            "2919000 :  0.001\n",
            "2920000 :  0.0\n",
            "2921000 :  0.0\n",
            "2922000 :  0.0\n",
            "2923000 :  0.0\n",
            "2924000 :  0.0\n",
            "2925000 :  0.001\n",
            "2926000 :  0.0\n",
            "2927000 :  0.0\n",
            "2928000 :  0.0\n",
            "2929000 :  0.0\n",
            "2930000 :  0.001\n",
            "2931000 :  0.0\n",
            "2932000 :  0.0\n",
            "2933000 :  0.0\n",
            "2934000 :  0.0\n",
            "2935000 :  0.0\n",
            "2936000 :  0.0\n",
            "2937000 :  0.0\n",
            "2938000 :  0.0\n",
            "2939000 :  0.0\n",
            "2940000 :  0.0\n",
            "2941000 :  0.0\n",
            "2942000 :  0.0\n",
            "2943000 :  0.0\n",
            "2944000 :  0.001\n",
            "2945000 :  0.0\n",
            "2946000 :  0.0\n",
            "2947000 :  0.0\n",
            "2948000 :  0.0\n",
            "2949000 :  0.0\n",
            "2950000 :  0.0\n",
            "2951000 :  0.0\n",
            "2952000 :  0.0\n",
            "2953000 :  0.0\n",
            "2954000 :  0.0\n",
            "2955000 :  0.0\n",
            "2956000 :  0.0\n",
            "2957000 :  0.0\n",
            "2958000 :  0.0\n",
            "2959000 :  0.0\n",
            "2960000 :  0.0\n",
            "2961000 :  0.0\n",
            "2962000 :  0.0\n",
            "2963000 :  0.0\n",
            "2964000 :  0.0\n",
            "2965000 :  0.0\n",
            "2966000 :  0.0\n",
            "2967000 :  0.0\n",
            "2968000 :  0.0\n",
            "2969000 :  0.0\n",
            "2970000 :  0.0\n",
            "2971000 :  0.0\n",
            "2972000 :  0.0\n",
            "2973000 :  0.0\n",
            "2974000 :  0.0\n",
            "2975000 :  0.0\n",
            "2976000 :  0.0\n",
            "2977000 :  0.0\n",
            "2978000 :  0.0\n",
            "2979000 :  0.0\n",
            "2980000 :  0.0\n",
            "2981000 :  0.0\n",
            "2982000 :  0.0\n",
            "2983000 :  0.001\n",
            "2984000 :  0.0\n",
            "2985000 :  0.0\n",
            "2986000 :  0.0\n",
            "2987000 :  0.0\n",
            "2988000 :  0.0\n",
            "2989000 :  0.0\n",
            "2990000 :  0.0\n",
            "2991000 :  0.001\n",
            "2992000 :  0.001\n",
            "2993000 :  0.0\n",
            "2994000 :  0.0\n",
            "2995000 :  0.0\n",
            "2996000 :  0.0\n",
            "2997000 :  0.0\n",
            "2998000 :  0.0\n",
            "2999000 :  0.0\n",
            "3000000 :  0.0\n",
            "Q table\n",
            "\n",
            "[[2.59857091e-01 8.81471535e-02 6.79265912e-02 5.53600300e-02]\n",
            " [4.11953143e-02 4.45509544e-02 4.11030370e-02 4.09923071e-02]\n",
            " [7.94761584e-02 1.91213021e-02 2.02596638e-02 1.96593690e-02]\n",
            " [9.00000000e-03 1.50000000e-02 1.20000000e-02 9.00000000e-03]\n",
            " [1.20000000e-02 2.00000000e-02 1.60000000e-02 1.20000000e-02]\n",
            " [1.50000000e-02 2.50000000e-02 2.00000000e-02 1.50000000e-02]\n",
            " [1.80000000e-02 3.00000000e-02 2.40000000e-02 1.80000000e-02]\n",
            " [2.10000000e-02 3.50000000e-02 2.80000000e-02 2.10000000e-02]\n",
            " [2.40000000e-02 4.00000000e-02 3.20000000e-02 2.40000000e-02]\n",
            " [2.70000000e-02 4.50000000e-02 3.60000000e-02 2.70000000e-02]\n",
            " [3.15718284e-01 8.93295887e-02 9.13887058e-02 4.75667777e-02]\n",
            " [3.30000000e-02 5.50000000e-02 4.40000000e-02 3.30000000e-02]\n",
            " [4.22719704e-02 4.27245608e-02 8.14389807e-02 4.13121929e-02]\n",
            " [3.90000000e-02 6.50000000e-02 5.20000000e-02 3.90000000e-02]\n",
            " [4.20000000e-02 7.00000000e-02 5.60000000e-02 4.20000000e-02]\n",
            " [4.50000000e-02 7.50000000e-02 6.00000000e-02 4.50000000e-02]\n",
            " [4.80000000e-02 8.00000000e-02 6.40000000e-02 4.80000000e-02]\n",
            " [5.10000000e-02 8.50000000e-02 6.80000000e-02 5.10000000e-02]\n",
            " [5.40000000e-02 9.00000000e-02 7.20000000e-02 5.40000000e-02]\n",
            " [5.70000000e-02 9.50000000e-02 7.60000000e-02 5.70000000e-02]\n",
            " [2.64481166e-01 1.13745392e-01 1.23185535e-01 1.20948966e-01]\n",
            " [6.30000000e-02 1.05000000e-01 8.40000000e-02 6.30000000e-02]\n",
            " [6.60000000e-02 1.10000000e-01 8.80000000e-02 6.60000000e-02]\n",
            " [6.90000000e-02 1.15000000e-01 9.20000000e-02 6.90000000e-02]\n",
            " [7.20000000e-02 1.20000000e-01 9.60000000e-02 7.20000000e-02]\n",
            " [7.50000000e-02 1.25000000e-01 1.00000000e-01 7.50000000e-02]\n",
            " [7.80000000e-02 1.30000000e-01 1.04000000e-01 7.80000000e-02]\n",
            " [8.10000000e-02 1.35000000e-01 1.08000000e-01 8.10000000e-02]\n",
            " [8.40000000e-02 1.40000000e-01 1.12000000e-01 8.40000000e-02]\n",
            " [8.70000000e-02 1.45000000e-01 1.16000000e-01 8.70000000e-02]\n",
            " [1.74181473e-01 1.50832039e-01 1.63116314e-01 2.95268922e-01]\n",
            " [1.34834772e-01 2.24900908e-01 8.21936806e-02 8.79890261e-02]\n",
            " [1.45179770e-01 1.63062429e-01 8.75947779e-02 9.21355918e-02]\n",
            " [9.17042347e-02 1.64111854e-01 1.20721582e-01 1.23305280e-01]\n",
            " [1.02000000e-01 1.70000000e-01 1.36000000e-01 1.02000000e-01]\n",
            " [1.05000000e-01 1.75000000e-01 1.40000000e-01 1.05000000e-01]\n",
            " [1.08000000e-01 1.80000000e-01 1.44000000e-01 1.08000000e-01]\n",
            " [1.11000000e-01 1.85000000e-01 1.48000000e-01 1.11000000e-01]\n",
            " [1.14000000e-01 1.90000000e-01 1.52000000e-01 1.14000000e-01]\n",
            " [1.17000000e-01 1.95000000e-01 1.56000000e-01 1.17000000e-01]\n",
            " [1.20000000e-01 2.00000000e-01 1.60000000e-01 1.20000000e-01]\n",
            " [1.51966235e-01 1.52295942e-01 2.11745940e-01 1.57199152e-01]\n",
            " [1.26000000e-01 2.10000000e-01 1.68000000e-01 1.26000000e-01]\n",
            " [1.61630968e-01 1.57500098e-01 3.05037768e-01 1.57500076e-01]\n",
            " [1.32000000e-01 2.20000000e-01 1.76000000e-01 1.32000000e-01]\n",
            " [1.35000000e-01 2.25000000e-01 1.80000000e-01 1.35000000e-01]\n",
            " [1.38000000e-01 2.30000000e-01 1.84000000e-01 1.38000000e-01]\n",
            " [1.41000000e-01 2.35000000e-01 1.88000000e-01 1.41000000e-01]\n",
            " [1.44000000e-01 2.40000000e-01 1.92000000e-01 1.44000000e-01]\n",
            " [1.47000000e-01 2.45000000e-01 1.96000000e-01 1.47000000e-01]\n",
            " [1.96651993e-01 5.46847739e-01 2.33267928e-01 2.23504342e-01]\n",
            " [2.87437714e-01 1.95233244e-01 1.96111103e-01 1.95476408e-01]\n",
            " [1.56000000e-01 2.60000000e-01 2.08000000e-01 1.56000000e-01]\n",
            " [1.96434517e-01 1.96492764e-01 2.18695893e-01 1.95288681e-01]\n",
            " [1.62000000e-01 2.70000000e-01 2.16000000e-01 1.62000000e-01]\n",
            " [1.65000000e-01 2.75000000e-01 2.20000000e-01 1.65000000e-01]\n",
            " [1.68000000e-01 2.80000000e-01 2.24000000e-01 1.68000000e-01]\n",
            " [1.71000000e-01 2.85000000e-01 2.28000000e-01 1.71000000e-01]\n",
            " [1.74000000e-01 2.90000000e-01 2.32000000e-01 1.74000000e-01]\n",
            " [1.77000000e-01 2.95000000e-01 2.36000000e-01 1.77000000e-01]\n",
            " [5.55905699e-01 2.24176190e-01 2.50852708e-01 2.30110265e-01]\n",
            " [1.83000000e-01 3.05000000e-01 2.44000000e-01 1.83000000e-01]\n",
            " [1.86000000e-01 3.10000000e-01 2.48000000e-01 1.86000000e-01]\n",
            " [2.73244113e-01 2.52665988e-01 2.26863862e+00 2.58596258e-01]\n",
            " [2.41504815e-01 2.00277654e+00 3.66826029e-01 2.48948872e-01]\n",
            " [2.90657521e-01 1.49170968e+00 2.99590820e-01 2.78433616e-01]\n",
            " [2.26403939e-01 1.80127447e+00 2.38660782e-01 2.12388446e-01]\n",
            " [2.01000000e-01 3.35000000e-01 2.68000000e-01 2.01000000e-01]\n",
            " [2.04000000e-01 3.40000000e-01 2.72000000e-01 2.04000000e-01]\n",
            " [2.07000000e-01 3.45000000e-01 2.76000000e-01 2.07000000e-01]\n",
            " [2.56346307e-01 2.00391452e-01 3.76540265e-01 2.55828675e-01]\n",
            " [2.13000000e-01 3.55000000e-01 2.84000000e-01 2.13000000e-01]\n",
            " [2.16000000e-01 3.60000000e-01 2.88000000e-01 2.16000000e-01]\n",
            " [2.19000000e-01 3.65000000e-01 2.92000000e-01 2.19000000e-01]\n",
            " [3.13631727e-01 3.08123613e-01 6.22700293e-01 3.01331568e-01]\n",
            " [3.35482130e-01 3.68127825e-01 3.88579109e-01 3.51481534e+00]\n",
            " [3.14996550e-01 4.13085073e-01 3.25515315e-01 3.22642015e-01]\n",
            " [3.36300307e-01 1.36187347e+01 5.59020284e-01 4.18658450e-01]\n",
            " [1.03331588e+00 2.25854846e+01 2.66400000e-01 2.34000000e-01]\n",
            " [2.37000000e-01 4.23340071e+01 3.16000000e-01 2.37000000e-01]\n",
            " [2.40000000e-01 4.00000000e-01 3.20000000e-01 2.40000000e-01]\n",
            " [2.43000000e-01 4.05000000e-01 3.24000000e-01 2.43000000e-01]\n",
            " [2.46000000e-01 4.10000000e-01 3.28000000e-01 2.46000000e-01]\n",
            " [2.49000000e-01 4.15000000e-01 3.32000000e-01 2.49000000e-01]\n",
            " [2.52000000e-01 4.20000000e-01 3.36000000e-01 2.52000000e-01]\n",
            " [2.55000000e-01 4.25000000e-01 3.40000000e-01 2.55000000e-01]\n",
            " [2.58000000e-01 4.30000000e-01 3.44000000e-01 2.58000000e-01]\n",
            " [4.31069905e-01 3.26713720e-01 8.42486838e+00 3.27528458e-01]\n",
            " [1.24501921e+00 6.01105625e-01 3.64400000e-01 1.44726779e+01]\n",
            " [2.67000000e-01 2.03477230e+01 3.56000000e-01 2.67000000e-01]\n",
            " [2.70000000e-01 4.50000000e-01 3.60000000e-01 2.70000000e-01]\n",
            " [2.73000000e-01 4.55000000e-01 3.64000000e-01 2.73000000e-01]\n",
            " [2.76000000e-01 4.60000000e-01 3.68000000e-01 2.76000000e-01]\n",
            " [2.79000000e-01 4.65000000e-01 3.72000000e-01 2.79000000e-01]\n",
            " [2.82000000e-01 4.70000000e-01 3.76000000e-01 2.82000000e-01]\n",
            " [2.85000000e-01 4.75000000e-01 3.80000000e-01 2.85000000e-01]\n",
            " [2.88000000e-01 4.80000000e-01 3.84000000e-01 2.88000000e-01]\n",
            " [2.91000000e-01 3.19097314e-01 3.49985800e-01 1.47076876e+00]\n",
            " [2.94000000e-01 4.90000000e-01 3.92000000e-01 2.94000000e-01]\n",
            " [2.97000000e-01 4.95000000e-01 3.96000000e-01 2.97000000e-01]]\n"
          ]
        }
      ],
      "source": [
        "rewards = []\n",
        "exploration_rate_tracker = []\n",
        "# Let agent know that goal is in the right most corner\n",
        "# Populate q table with ascending values to the goal\n",
        "for i in range(0, state_space_size):\n",
        "    for j in range(0, action_space_size):\n",
        "        # give second index the highest value to encourage agent to go down\n",
        "        if j == 1:\n",
        "            q_table[i][j] = 0.5 * i / 100\n",
        "        # give third index the second highest value to encourage agent to go right\n",
        "        if j == 2:\n",
        "            q_table[i][j] = 0.4 * i / 100\n",
        "        # give the other two actions the lowest value to discourage agent from going up or left (This will be overwritten by the random initialization)\n",
        "        if j == 0:\n",
        "            q_table[i][j] = 0.3 * i / 100\n",
        "        if j == 3:\n",
        "            q_table[i][j] = 0.3 * i / 100\n",
        "\n",
        "for episode in range(total_episodes):\n",
        "    state = env.reset()\n",
        "    step = 0\n",
        "    done = False\n",
        "    total_rewards = 0\n",
        "    \n",
        "    for step in range(max_steps):\n",
        "        exp_exp_tradeoff = random.uniform(0, 1)\n",
        "        if exp_exp_tradeoff > epsilon:\n",
        "            # action = np.argmax(q_table[state,:])\n",
        "            action = np.argmax(q_table[state,:])\n",
        "        else:\n",
        "            action = env.action_space.sample()\n",
        "\n",
        "        new_state, reward, done, info = env.step(action)\n",
        "        temp_reward = reward\n",
        "        if not done and reward == 0.0 and new_state != state:\n",
        "            temp_reward = 0.1 * new_state/100\n",
        "        if done and reward == 0.0:\n",
        "            # negative reward for falling in a hole in early states\n",
        "            temp_reward = -0.1 * new_state/100\n",
        "        if done and reward == 1.0:\n",
        "            temp_reward = 50\n",
        "\n",
        "        q_table[state, action] = q_table[state, action] + learning_rate * (temp_reward + gamma * np.max(q_table[new_state, :]) - q_table[state, action])\n",
        "        \n",
        "        total_rewards += reward\n",
        "        state = new_state\n",
        "        \n",
        "        if done == True: \n",
        "            break\n",
        "    epsilon = min_epsilon + (max_epsilon - min_epsilon)*np.exp(-decay_rate*episode) \n",
        "    rewards.append(total_rewards)\n",
        "    exploration_rate_tracker.append(epsilon)\n",
        "\n",
        "print (\"Score over time: \" +  str(sum(rewards)/total_episodes))\n",
        "print(q_table)\n",
        "\n",
        "plot_epsilon(exploration_rate_tracker, total_episodes)\n",
        "plot_rewards(rewards, total_episodes)\n",
        "\n",
        "rewards_per_thousand_episodes = np.split(np.array(rewards), total_episodes/1000)\n",
        "count = 1000\n",
        "print(\"Avg reward per 1000 episode \\n\")\n",
        "for r in rewards_per_thousand_episodes:\n",
        "  print(count, \": \", str(sum(r/1000)))\n",
        "  count += 1000\n",
        "\n",
        "print(\"Q table\\n\")\n",
        "print(q_table)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {},
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/var/folders/g2/_dzws83d5_x2sjj38b2qb5mw0000gn/T/ipykernel_13340/4250246634.py:6: DeprecationWarning: Importing display from IPython.core.display is deprecated since IPython 7.14, please import from IPython display\n",
            "  from IPython.core.display import display, HTML\n"
          ]
        }
      ],
      "source": [
        "import imageio\n",
        "import base64\n",
        "import pygame\n",
        "import os\n",
        "import matplotlib.pyplot as plt\n",
        "from IPython.core.display import display, HTML\n",
        "import os\n",
        "os.environ[\"IMAGEIO_FFMPEG_EXE\"] = \"/usr/local/bin/ffmpeg\"\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {},
      "outputs": [],
      "source": [
        "def embed_mp4(filename):\n",
        "    \"\"\"Embeds an mp4 file in the notebook.\"\"\"\n",
        "    video = open(filename,'rb').read()\n",
        "    b64 = base64.b64encode(video)\n",
        "    tag = '''\n",
        "    <video width=\"640\" height=\"480\" controls>\n",
        "      <source src=\"data:video/mp4;base64,{0}\" type=\"video/mp4\">\n",
        "    Your browser does not support the video tag.\n",
        "    </video>'''.format(b64.decode())\n",
        "    \n",
        "    return HTML(tag)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "def create_policy_eval_video(filename, num_episodes=100, fps=5):\n",
        "    filename = filename + \".mp4\"\n",
        "    wins = 0.0\n",
        "    # eval_py_env = make_env()\n",
        "    with imageio.get_writer(filename, fps=fps) as video:\n",
        "        for i in range(num_episodes):\n",
        "            print(\"Game:\", i, \"Wins:\", wins)\n",
        "            done = False\n",
        "            env.reset()\n",
        "            start = np.array(env.render(\"rgb_array\"))\n",
        "            start = np.array(start).reshape((512,512, 3))\n",
        "            video.append_data(start)\n",
        "            steps = 0\n",
        "            while not done:\n",
        "                if steps > 800:\n",
        "                    raise Exception(\"Too many steps\")\n",
        "                steps += 1\n",
        "\n",
        "                action = np.argmax(q_table[state,:])\n",
        "                next_state, reward, te, tr = env.step(action)\n",
        "                wins += reward\n",
        "                done = te\n",
        "                end = np.array(env.render(\"rgb_array\"))\n",
        "                end = np.array(end).reshape((512,512, 3))\n",
        "                video.append_data(end)\n",
        "    return embed_mp4(filename)\n",
        "\n",
        "create_policy_eval_video(\"trained_agent_q_learning\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [
        {
          "ename": "",
          "evalue": "",
          "output_type": "error",
          "traceback": [
            "\u001b[1;31mFailed to start the Kernel. \n",
            "\u001b[1;31mKernel Python 3.9.6 64-bit is not usable. Check the Jupyter output tab for more information. \n",
            "\u001b[1;31mView Jupyter <a href='command:jupyter.viewOutput'>log</a> for further details."
          ]
        }
      ],
      "source": [
        "count = 0\n",
        "\n",
        "map_number_to_alphabet = {}\n",
        "\n",
        "for i in range(0, 26):\n",
        "    map_number_to_alphabet[i] = chr(i + 65)\n",
        "map_number_to_alphabet"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "# Proteins are composed of small organic molecules called amino acids, which are chained together. There exist 20 natural amino acids, which can be designated by a one-letter code: ARNDCEQGHILKMFPSTWYV.\n",
        "\n",
        "# The linear sequence of amino acids within a protein is called its \"primary structure\".\n",
        "\n",
        " \n",
        "\n",
        "# For instance, the subunit alpha of the human Hemoglobin protein has the following sequence: \n",
        "\n",
        "# MVLSPADKTNVKAAWGKVGAHAGEYGAEALERMFLSFPTTKTYFPHFDLSHGSAQVKGHGKKVADALTNAVAHVDDMPNALSALSDLHAHKLRVDPVNFKLLSHCLLVTLAAHLPAEFTPAVHASLDKFLASVSTVLTSKYR\n",
        "\n",
        " \n",
        "\n",
        " \n",
        "\n",
        "# While valuable information can be inferred from the primary structure only, e.g. with Protein Language Models, protein design heavily relies on the 3-dimensional arrangement of the atoms of the protein in space, called the tertiary structure.\n",
        "\n",
        "# Such data is usually obtained experimentally using X-ray crystallography, or more recently by cryo-electron microscopy.\n",
        "\n",
        "# However, for a variety of reasons, it can happen that the protein is only partially observed: atoms or even strands of amino acids end up missing in the 3D model. As a consequence, the sequence of amino acids that are present in the 3D model will differ from the actual protein's primary structure.\n",
        "\n",
        " \n",
        "\n",
        "# In terms of storing 3D structural protein data, the PDB (Protein Data Bank) file format is ubiquitous, and uniquely identifies amino acids with the following triplet:\n",
        "\n",
        "# - chain ID: a letter identify which entity the amino acid belongs to, as there could be more than 1 protein or several copies of the same protein represented in the experimental file. Example: \"A\"\n",
        "\n",
        "# - position: a number identifying the amino acid, which does not necessarily start with 1 (because part of the protein might be missing for instance)\n",
        "\n",
        "# - an insertion code: for comparability concerns, structural biologists may consider that a given amino acid of importance should always bear the same position in different structures, even if the lengths of the sequences differ. So that some positions may be comparable, optional insertion codes allow to consider some amino acids as \"inserted\", and they'll then bear the same position, but be distinguished by one additional letter such as \"A\", \"B\", \"C\".\n",
        "\n",
        " \n",
        "\n",
        "# In this exercise, position triplets will be written as strings with the following format: \"<chain ID>.<position>.<Optional insertion code>\". Here's an example:\n",
        "\n",
        " \n",
        "\n",
        "# \"A.12.\" # chain A, position 12, no insertion code\n",
        "# \"A.14.A\" # chain A, position 14, insertion code A\n",
        "\n",
        " \n",
        "\n",
        "# In this problem, you'll be tasked with finding the correct mapping from PDB position identifiers of 3D structures of proteins to the corresponding primary structure sequence indexes (indexing from 0). Only a single chain will be considered at a time: all the positions will have the same chain ID and refer to the same protein.\n",
        "\n",
        "# In addition, we make the following key assumptions:\n",
        "\n",
        "# - 1. that there are no missing \"inserted\" amino acids between two consecutive position numbers (no missing insertion codes), e.g. there cannot be a missing \"H.4.A\", or \"H.5.C\" in \"H.4. H.5. H.5.A H.5.B H.6\";\n",
        "\n",
        "# - 2. on the other hand a gap between position X.N. and X.N+K. where K > 1 does not mean there are K−1 missing amino acids, but rather at least K−1, taking into account the potential presence of insertion codes within the \"missing positions\";\n",
        "\n",
        "# - 3. insertion codes always follow in alphabetic order, starting with none: \"\", \"A\", \"B\", \"C\", …;\n",
        "\n",
        "# - 4. the first position starting at number N does not mean the first index is N−1;\n",
        "# - 5. all present insertion codes in the 3D model sequence shall me mapped (they are not optional).\n",
        "\n",
        " \n",
        "\n",
        "# For instance, if the primary structure is MVLSP, and the following amino acids can be found in a 3D model (position and amino acid are separated by a space):\n",
        "\n",
        " \n",
        "\n",
        "# A.1. M\n",
        "# A.2. V \n",
        "# A.3. L \n",
        "# A.4. S\n",
        "# A.5. P\n",
        " \n",
        "\n",
        "# A mapping satisfying our assumptions would be:\n",
        "\n",
        " \n",
        "\n",
        "# A.1. 0 \n",
        "# A.2. 1\n",
        "# A.3. 2\n",
        "# A.4. 3\n",
        "# A.5. 4\n",
        " \n",
        "\n",
        "# Let's now consider the following example, with primary structure MFVFLVLLPLVSSQCVN and 3D Model sequence:\n",
        "\n",
        " \n",
        "\n",
        "# C.4. F\n",
        "# C.5. L\n",
        "# C.6. V\n",
        "# C.6.A L\n",
        "# C.6.B L\n",
        "# C.7. P \n",
        "# C.12. Q\n",
        "# C.13. C\n",
        " \n",
        "\n",
        "# The expected mapping would be:\n",
        "\n",
        " \n",
        "\n",
        "# C.4. 3\n",
        "# C.5. 4\n",
        "# C.6. 5\n",
        "# C.6.A 6\n",
        "# C.6.B 7\n",
        "# C.7. 8\n",
        "# C.12. 13\n",
        "# C.13. 14\n",
        "\n",
        "def find_pdb_mapping(primary_structure, pdb_sequence):\n",
        "    pdb_amino_acids = []\n",
        "    "
      ]
    }
  ],
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3.9.6 64-bit",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.9.6"
    },
    "vscode": {
      "interpreter": {
        "hash": "31f2aee4e71d21fbe5cf8b01ff0e069b9275f58929596ceb00d14d90e3e16cd6"
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}