notebooks/plot_mnist.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pylab import *\n",
    "import numpy as np\n",
    "import matplotlib, pylab\n",
    "from dcll.npamlib import plotLIF\n",
    "from dcll.pytorch_libdcll import *\n",
    "import os.path as path\n",
    "\n",
    "matplotlib.rcParams['text.usetex']=False\n",
    "matplotlib.rcParams['savefig.dpi']=400.\n",
    "matplotlib.rcParams['font.size']=14.0\n",
    "matplotlib.rcParams['figure.figsize']=(5.0,3.5)\n",
    "matplotlib.rcParams['axes.formatter.limits']=[-10,10]\n",
    "matplotlib.rcParams['axes.labelsize']= 14.\n",
    "matplotlib.rcParams['figure.subplot.bottom'] = .2\n",
    "matplotlib.rcParams['figure.subplot.left'] = .2       "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1]\n"
     ]
    }
   ],
   "source": [
    "# all_layers_folder_id = range(1, 5, 1)\n",
    "all_layers_folder_id = [1]\n",
    "base_folder = \"../Paper_results/mnist_long/\"\n",
    "base_folder_begin = \"../Paper_results/mnist_short/\"\n",
    "all_layers_folders = [\n",
    "    base_folder.format(n) for n in all_layers_folder_id\n",
    "]\n",
    "all_layers_begin_folders = [\n",
    "    base_folder_begin.format(n) for n in all_layers_folder_id\n",
    "]\n",
    "print(list(all_layers_folder_id))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "    \n",
    "def plot_accuracy(directories, n_updates_tick=1, comment=\"\", discard_last_epoch=0, vline=None, xmax=None, ymax=60, n_layers=None):\n",
    "    args = np.load(path.join(directories[0],\"args.pkl\"))\n",
    "    batch_size = args['batch_size']\n",
    "    n_test_interval = args['n_test_interval']\n",
    " \n",
    "    acc_test = [ np.load(path.join(d, \"acc_test.npy\")) for d in directories ]\n",
    "    acc_test = np.array(acc_test)\n",
    "    \n",
    "    acc_test_ref = [ np.load(path.join(d, \"acc_test_ref.npy\")) for d in directories ]\n",
    "    acc_test_ref = np.array(acc_test_ref)\n",
    "    \n",
    "    if discard_last_epoch > 0:\n",
    "        acc_test = acc_test[:, :-discard_last_epoch]\n",
    "        acc_test_ref = acc_test_ref[:, :-discard_last_epoch]\n",
    "    \n",
    "    print(\"Acc test last layer last epoch: {}\".format(acc_test[0,-1,:,-1]))\n",
    "    print(\"(n_runs, n_epoch/test_interval, n_batch_test, n_layers): {}\".format(acc_test.shape))\n",
    "    n_epoch = (acc_test.shape[1] - 1) * n_test_interval + 1\n",
    "    if n_layers is None:\n",
    "        n_layers = acc_test.shape[3]\n",
    "    \n",
    "    test_epochs = np.arange(0, n_epoch, n_test_interval, dtype='int') + 1\n",
    "    # mean over the multiple test batches\n",
    "    weighted_acc = np.mean(acc_test, axis=2)\n",
    "    weighted_acc_ref = np.mean(acc_test_ref, axis=2)\n",
    "    # mean over the multiple runs with different seeds\n",
    "    mean_acc = np.mean(weighted_acc, axis=0)\n",
    "    std_acc = np.std(weighted_acc, axis=0)\n",
    "    mean_acc_ref = np.mean(weighted_acc_ref, axis=0)\n",
    "    std_acc_ref = np.std(weighted_acc_ref, axis=0)\n",
    "    \n",
    "    print(\"Accuracy: {}\".format(mean_acc[:, -1]))\n",
    "    print(\"Accuracy ref: {}\".format(mean_acc_ref))\n",
    "    \n",
    "    fig = figure()\n",
    "    ax = subplot(111)\n",
    "    x = test_epochs*batch_size//n_updates_tick\n",
    "    \n",
    "    for l in range(n_layers):\n",
    "        ax.plot(x, 100.*(1-mean_acc[:,l]),\n",
    "                linewidth=3,\n",
    "                alpha=.75, label = 'Layer {}'.format(l+1))\n",
    "        ax.fill_between(x, 1-mean_acc[:,l] - std_acc[:,l], 1-mean_acc[:,l] + std_acc[:,l],\n",
    "                       alpha=.2)\n",
    "    ax.plot(x, 100.*(1-mean_acc_ref),\n",
    "                linewidth=3, linestyle='--',\n",
    "                alpha=.75, label = 'Reference')\n",
    "    ax.fill_between(x, 1-mean_acc_ref - std_acc_ref, 1-mean_acc_ref + std_acc_ref,\n",
    "                alpha=.2)\n",
    "    \n",
    "\n",
    "        \n",
    "    ax.grid()\n",
    "    legend()\n",
    "    ax.set_ylabel('Error[%]')\n",
    "    ax.set_xlabel('Training Samples [{}]'.format(n_updates_tick))\n",
    "    tight_layout()\n",
    "    ax.set_ylim([0, ymax])\n",
    "    ax.set_xlim(0, n_epoch*batch_size // n_updates_tick)\n",
    "    #ax.set_xlim(batch_size, 40000)\n",
    "    if vline is not None:\n",
    "        ax.axvline(x=vline, linestyle=':', color='black')  \n",
    "    [prev_xmin, prev_xmax] = ax.get_xlim()\n",
    "    if xmax is not None:\n",
    "        ax.set_xlim(prev_xmin, xmax)\n",
    "    savefig(path.join(path.dirname(directories[0]), 'convergence_mnist{}.png'.format(comment)), dpi=300, frameon=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Accuracy of the network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Acc test last layer last epoch: [0.98828125 0.98828125]\n",
      "(n_runs, n_epoch/test_interval, n_batch_test, n_layers): (1, 500, 2, 3)\n",
      "Accuracy: [0.51757812 0.89648438 0.93359375 0.93652344 0.94726562 0.94433594\n",
      " 0.95507812 0.94921875 0.95507812 0.96582031 0.9453125  0.95703125\n",
      " 0.95898438 0.9609375  0.95996094 0.95996094 0.94824219 0.96582031\n",
      " 0.96777344 0.95898438 0.96972656 0.96484375 0.97167969 0.96289062\n",
      " 0.95117188 0.96679688 0.96679688 0.97460938 0.95605469 0.97167969\n",
      " 0.9609375  0.97558594 0.97265625 0.96972656 0.97558594 0.97460938\n",
      " 0.96484375 0.97167969 0.97460938 0.97070312 0.97558594 0.96875\n",
      " 0.96582031 0.97070312 0.97460938 0.96679688 0.9609375  0.97460938\n",
      " 0.96386719 0.97460938 0.97265625 0.96972656 0.97460938 0.97460938\n",
      " 0.98144531 0.9765625  0.97949219 0.98046875 0.97167969 0.98046875\n",
      " 0.96972656 0.97949219 0.9765625  0.9765625  0.97753906 0.984375\n",
      " 0.9765625  0.97363281 0.9765625  0.98339844 0.9765625  0.97363281\n",
      " 0.98046875 0.97363281 0.97851562 0.97070312 0.97558594 0.9765625\n",
      " 0.98046875 0.9765625  0.97460938 0.97363281 0.97753906 0.97851562\n",
      " 0.97851562 0.97558594 0.97167969 0.9765625  0.97460938 0.98144531\n",
      " 0.97851562 0.97949219 0.97363281 0.97363281 0.97265625 0.97460938\n",
      " 0.97753906 0.97363281 0.97949219 0.98144531 0.97851562 0.97949219\n",
      " 0.98242188 0.98046875 0.9765625  0.98144531 0.98242188 0.98242188\n",
      " 0.98144531 0.97851562 0.9765625  0.98144531 0.97851562 0.98046875\n",
      " 0.97363281 0.98242188 0.98339844 0.97753906 0.97949219 0.98046875\n",
      " 0.97949219 0.98046875 0.98242188 0.98242188 0.98144531 0.98535156\n",
      " 0.97753906 0.97363281 0.97851562 0.98144531 0.98046875 0.98144531\n",
      " 0.98242188 0.98730469 0.98242188 0.98339844 0.97851562 0.98339844\n",
      " 0.98339844 0.98144531 0.98339844 0.97851562 0.98144531 0.98632812\n",
      " 0.98339844 0.98046875 0.98242188 0.98242188 0.98242188 0.97851562\n",
      " 0.97851562 0.98730469 0.98339844 0.984375   0.98046875 0.984375\n",
      " 0.98144531 0.98144531 0.98535156 0.97851562 0.97949219 0.97265625\n",
      " 0.97949219 0.98046875 0.98632812 0.98535156 0.98046875 0.97753906\n",
      " 0.98144531 0.984375   0.97949219 0.98242188 0.98339844 0.98242188\n",
      " 0.98828125 0.98144531 0.98632812 0.98339844 0.98242188 0.984375\n",
      " 0.97851562 0.98242188 0.984375   0.98144531 0.984375   0.98339844\n",
      " 0.98046875 0.98535156 0.97851562 0.98242188 0.98632812 0.98242188\n",
      " 0.97949219 0.97949219 0.98046875 0.97851562 0.98046875 0.98242188\n",
      " 0.97949219 0.98144531 0.98828125 0.984375   0.98339844 0.98535156\n",
      " 0.984375   0.98828125 0.98535156 0.98339844 0.98339844 0.98535156\n",
      " 0.98046875 0.98339844 0.98046875 0.98632812 0.98339844 0.98730469\n",
      " 0.98925781 0.984375   0.98632812 0.98632812 0.98925781 0.98535156\n",
      " 0.98339844 0.98730469 0.984375   0.98339844 0.98730469 0.98144531\n",
      " 0.98144531 0.98339844 0.984375   0.98535156 0.98339844 0.984375\n",
      " 0.98730469 0.984375   0.98632812 0.984375   0.98339844 0.98046875\n",
      " 0.98242188 0.98339844 0.984375   0.98535156 0.98632812 0.984375\n",
      " 0.98632812 0.984375   0.97949219 0.98339844 0.984375   0.98242188\n",
      " 0.98144531 0.984375   0.98535156 0.98144531 0.98632812 0.98730469\n",
      " 0.984375   0.98339844 0.98730469 0.98632812 0.984375   0.98925781\n",
      " 0.98925781 0.98242188 0.98632812 0.98339844 0.98632812 0.98535156\n",
      " 0.98242188 0.98242188 0.98339844 0.98144531 0.984375   0.984375\n",
      " 0.98339844 0.98632812 0.98828125 0.98144531 0.98632812 0.984375\n",
      " 0.98632812 0.98046875 0.98339844 0.98632812 0.98339844 0.98535156\n",
      " 0.98242188 0.98828125 0.98339844 0.98339844 0.98339844 0.98632812\n",
      " 0.98144531 0.984375   0.98339844 0.98632812 0.98730469 0.98925781\n",
      " 0.98339844 0.98730469 0.98535156 0.98632812 0.98925781 0.98730469\n",
      " 0.98339844 0.98730469 0.98632812 0.98730469 0.98535156 0.98535156\n",
      " 0.98535156 0.98730469 0.984375   0.98535156 0.98730469 0.98242188\n",
      " 0.98339844 0.98632812 0.98535156 0.98632812 0.98535156 0.984375\n",
      " 0.984375   0.98632812 0.984375   0.98535156 0.98632812 0.98730469\n",
      " 0.98730469 0.98632812 0.98632812 0.98339844 0.98632812 0.98144531\n",
      " 0.984375   0.98339844 0.98339844 0.98535156 0.98730469 0.98730469\n",
      " 0.98828125 0.98535156 0.98339844 0.98242188 0.984375   0.98535156\n",
      " 0.984375   0.98535156 0.98632812 0.984375   0.98632812 0.98828125\n",
      " 0.984375   0.98632812 0.98632812 0.98535156 0.98730469 0.984375\n",
      " 0.98339844 0.984375   0.98828125 0.98828125 0.984375   0.98339844\n",
      " 0.984375   0.98632812 0.98632812 0.98828125 0.98632812 0.98632812\n",
      " 0.984375   0.98730469 0.98730469 0.984375   0.98828125 0.98144531\n",
      " 0.98730469 0.98730469 0.98828125 0.98730469 0.98632812 0.98730469\n",
      " 0.98242188 0.98535156 0.98730469 0.984375   0.98632812 0.984375\n",
      " 0.97949219 0.984375   0.98828125 0.98828125 0.98339844 0.98535156\n",
      " 0.99023438 0.99023438 0.984375   0.98632812 0.98730469 0.98730469\n",
      " 0.984375   0.98730469 0.98730469 0.98730469 0.98632812 0.98632812\n",
      " 0.98632812 0.98925781 0.98632812 0.98339844 0.98828125 0.98242188\n",
      " 0.98535156 0.98632812 0.98535156 0.98632812 0.98925781 0.984375\n",
      " 0.98730469 0.98632812 0.98925781 0.98632812 0.98828125 0.98242188\n",
      " 0.98632812 0.984375   0.98632812 0.98535156 0.98535156 0.98632812\n",
      " 0.984375   0.98535156 0.98535156 0.984375   0.98730469 0.98828125\n",
      " 0.98828125 0.98535156 0.98632812 0.98730469 0.98730469 0.98828125\n",
      " 0.98632812 0.98828125 0.98730469 0.98730469 0.984375   0.98339844\n",
      " 0.984375   0.98632812 0.98730469 0.98535156 0.98632812 0.98632812\n",
      " 0.98535156 0.98144531 0.98632812 0.98730469 0.98632812 0.98242188\n",
      " 0.984375   0.98632812 0.98730469 0.98632812 0.984375   0.98632812\n",
      " 0.98242188 0.98535156 0.98730469 0.98535156 0.98535156 0.98535156\n",
      " 0.984375   0.98632812 0.98925781 0.984375   0.98730469 0.984375\n",
      " 0.98632812 0.98535156 0.98632812 0.98730469 0.98242188 0.98730469\n",
      " 0.98632812 0.98632812 0.984375   0.98632812 0.98925781 0.984375\n",
      " 0.98632812 0.98730469 0.98730469 0.984375   0.98730469 0.98632812\n",
      " 0.98632812 0.98828125]\n",
      "Accuracy ref: [0.10449219 0.86621094 0.9140625  0.92480469 0.95019531 0.95507812\n",
      " 0.96484375 0.96484375 0.96777344 0.97167969 0.97265625 0.97949219\n",
      " 0.97265625 0.97558594 0.97753906 0.97363281 0.97363281 0.97460938\n",
      " 0.97851562 0.98242188 0.97949219 0.98144531 0.98144531 0.97753906\n",
      " 0.98242188 0.98242188 0.98144531 0.98339844 0.98339844 0.984375\n",
      " 0.98046875 0.98339844 0.98339844 0.97851562 0.98339844 0.984375\n",
      " 0.98339844 0.98242188 0.98242188 0.984375   0.98632812 0.984375\n",
      " 0.98632812 0.98535156 0.98339844 0.984375   0.98144531 0.984375\n",
      " 0.98242188 0.98828125 0.98535156 0.98242188 0.98339844 0.984375\n",
      " 0.98339844 0.98535156 0.98730469 0.98535156 0.98535156 0.984375\n",
      " 0.98339844 0.98339844 0.98632812 0.98632812 0.984375   0.98535156\n",
      " 0.98535156 0.98828125 0.98339844 0.98730469 0.98632812 0.98535156\n",
      " 0.98828125 0.99121094 0.98632812 0.98632812 0.98046875 0.98730469\n",
      " 0.98632812 0.98339844 0.98535156 0.98730469 0.98535156 0.98242188\n",
      " 0.98535156 0.98730469 0.98632812 0.98632812 0.98632812 0.98339844\n",
      " 0.98730469 0.98632812 0.98632812 0.98925781 0.984375   0.98535156\n",
      " 0.98828125 0.98730469 0.98632812 0.98730469 0.98730469 0.98925781\n",
      " 0.98632812 0.984375   0.98925781 0.98535156 0.98828125 0.98632812\n",
      " 0.98339844 0.98730469 0.98535156 0.98828125 0.98632812 0.984375\n",
      " 0.98925781 0.98828125 0.98535156 0.98535156 0.984375   0.98632812\n",
      " 0.99023438 0.98828125 0.98730469 0.98730469 0.98632812 0.98828125\n",
      " 0.98730469 0.98828125 0.98632812 0.98632812 0.98730469 0.98730469\n",
      " 0.98828125 0.98632812 0.98632812 0.98632812 0.98535156 0.98828125\n",
      " 0.98730469 0.98730469 0.98828125 0.98535156 0.98730469 0.98535156\n",
      " 0.98535156 0.98632812 0.98925781 0.98730469 0.98730469 0.98828125\n",
      " 0.984375   0.98535156 0.98730469 0.98925781 0.98730469 0.98828125\n",
      " 0.98632812 0.98535156 0.98730469 0.98242188 0.98535156 0.98535156\n",
      " 0.984375   0.98828125 0.98730469 0.98828125 0.98828125 0.98632812\n",
      " 0.98828125 0.98828125 0.98535156 0.98925781 0.98535156 0.98828125\n",
      " 0.98925781 0.98632812 0.98632812 0.98632812 0.98632812 0.98828125\n",
      " 0.99023438 0.99023438 0.99023438 0.98925781 0.98925781 0.98730469\n",
      " 0.98632812 0.98632812 0.98535156 0.98339844 0.98730469 0.98730469\n",
      " 0.98632812 0.98925781 0.98632812 0.984375   0.98535156 0.98730469\n",
      " 0.98535156 0.984375   0.98535156 0.984375   0.98632812 0.98730469\n",
      " 0.98730469 0.98535156 0.98535156 0.98730469 0.98632812 0.98730469\n",
      " 0.98632812 0.98828125 0.98828125 0.98730469 0.98925781 0.98730469\n",
      " 0.98730469 0.98925781 0.98828125 0.98730469 0.98730469 0.98925781\n",
      " 0.98828125 0.98730469 0.98535156 0.99023438 0.98632812 0.98632812\n",
      " 0.98632812 0.98730469 0.99121094 0.98632812 0.98925781 0.984375\n",
      " 0.98925781 0.98828125 0.99023438 0.98925781 0.98632812 0.98730469\n",
      " 0.984375   0.98535156 0.98925781 0.98730469 0.98535156 0.98828125\n",
      " 0.99023438 0.99023438 0.98828125 0.98925781 0.98828125 0.98828125\n",
      " 0.98632812 0.984375   0.98730469 0.98925781 0.98632812 0.98828125\n",
      " 0.98730469 0.98730469 0.98730469 0.98730469 0.98828125 0.98632812\n",
      " 0.98535156 0.98730469 0.98925781 0.98828125 0.98730469 0.99023438\n",
      " 0.984375   0.98828125 0.98535156 0.98730469 0.98730469 0.98828125\n",
      " 0.98925781 0.99023438 0.98730469 0.98535156 0.984375   0.98730469\n",
      " 0.98730469 0.98828125 0.98632812 0.98632812 0.98828125 0.98730469\n",
      " 0.98925781 0.98828125 0.98535156 0.98632812 0.98828125 0.98730469\n",
      " 0.98535156 0.98828125 0.98535156 0.98730469 0.98730469 0.98925781\n",
      " 0.98925781 0.98828125 0.99121094 0.98730469 0.98925781 0.98925781\n",
      " 0.98730469 0.98730469 0.984375   0.98925781 0.98535156 0.98828125\n",
      " 0.98535156 0.98632812 0.99023438 0.984375   0.98632812 0.984375\n",
      " 0.99121094 0.98925781 0.99023438 0.9921875  0.98925781 0.98925781\n",
      " 0.98730469 0.98730469 0.98730469 0.98730469 0.98242188 0.98828125\n",
      " 0.98828125 0.984375   0.98925781 0.98925781 0.9921875  0.98828125\n",
      " 0.98730469 0.98632812 0.98632812 0.98632812 0.98925781 0.99023438\n",
      " 0.99023438 0.98925781 0.98632812 0.98730469 0.98730469 0.98828125\n",
      " 0.99121094 0.98828125 0.99023438 0.98925781 0.98730469 0.99121094\n",
      " 0.98925781 0.98632812 0.98730469 0.98828125 0.99023438 0.98925781\n",
      " 0.98632812 0.98730469 0.98925781 0.98925781 0.98730469 0.98925781\n",
      " 0.98925781 0.98828125 0.98730469 0.9921875  0.98925781 0.98632812\n",
      " 0.98925781 0.98828125 0.98730469 0.98828125 0.98730469 0.98828125\n",
      " 0.99023438 0.98730469 0.99023438 0.98828125 0.99023438 0.98632812\n",
      " 0.98632812 0.98925781 0.98828125 0.98925781 0.98828125 0.98925781\n",
      " 0.98925781 0.99121094 0.99023438 0.99023438 0.984375   0.98730469\n",
      " 0.98828125 0.98925781 0.98535156 0.98730469 0.98828125 0.99023438\n",
      " 0.98828125 0.99023438 0.98632812 0.98730469 0.98828125 0.98828125\n",
      " 0.99121094 0.98925781 0.98535156 0.98925781 0.98730469 0.98828125\n",
      " 0.98730469 0.99023438 0.98730469 0.99121094 0.98925781 0.98828125\n",
      " 0.99121094 0.98828125 0.98925781 0.99121094 0.98828125 0.98925781\n",
      " 0.98828125 0.98828125 0.99023438 0.99023438 0.99023438 0.99023438\n",
      " 0.99023438 0.98632812 0.98632812 0.98535156 0.98730469 0.98730469\n",
      " 0.98828125 0.99023438 0.98828125 0.98925781 0.98339844 0.98828125\n",
      " 0.984375   0.98730469 0.98925781 0.98632812 0.98925781 0.9921875\n",
      " 0.98925781 0.98535156 0.98339844 0.98730469 0.98925781 0.98828125\n",
      " 0.98828125 0.98535156 0.98925781 0.98730469 0.98925781 0.98925781\n",
      " 0.98730469 0.98925781 0.98925781 0.98730469 0.98632812 0.98925781\n",
      " 0.98730469 0.99023438 0.99023438 0.98730469 0.98828125 0.98730469\n",
      " 0.98925781 0.98535156 0.98828125 0.98730469 0.98730469 0.98632812\n",
      " 0.98828125 0.98730469 0.98632812 0.98925781 0.98730469 0.98730469\n",
      " 0.98730469 0.98828125 0.984375   0.98925781 0.99023438 0.98828125\n",
      " 0.98925781 0.98632812 0.98828125 0.98730469 0.98925781 0.98925781\n",
      " 0.99023438 0.98925781]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "n_updates_tick = 1000\n",
    "plot_accuracy(all_layers_folders, n_updates_tick=n_updates_tick, ymax=60)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Accuracy of the network at the beginning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Acc test last layer last epoch: [0.94335938 0.92382812]\n",
      "(n_runs, n_epoch/test_interval, n_batch_test, n_layers): (1, 50, 2, 3)\n",
      "Accuracy: [0.51953125 0.78027344 0.81738281 0.83300781 0.85839844 0.8359375\n",
      " 0.86621094 0.875      0.89257812 0.87011719 0.86914062 0.88964844\n",
      " 0.90625    0.89648438 0.90039062 0.91113281 0.91015625 0.92382812\n",
      " 0.90527344 0.91210938 0.91601562 0.93359375 0.9296875  0.91113281\n",
      " 0.91503906 0.921875   0.9375     0.93554688 0.94921875 0.94042969\n",
      " 0.92871094 0.93359375 0.92871094 0.91503906 0.91992188 0.93554688\n",
      " 0.9375     0.92871094 0.93847656 0.94238281 0.93457031 0.9140625\n",
      " 0.93847656 0.9296875  0.93847656 0.92578125 0.93066406 0.92871094\n",
      " 0.93847656 0.93359375]\n",
      "Accuracy ref: [0.10449219 0.14257812 0.25390625 0.35546875 0.51171875 0.65625\n",
      " 0.69628906 0.72070312 0.73535156 0.70507812 0.71386719 0.76953125\n",
      " 0.78808594 0.79296875 0.83886719 0.84667969 0.86328125 0.82421875\n",
      " 0.85546875 0.84472656 0.86621094 0.85644531 0.86816406 0.875\n",
      " 0.8671875  0.87988281 0.87792969 0.8828125  0.8828125  0.88378906\n",
      " 0.88378906 0.88964844 0.89160156 0.89648438 0.890625   0.89453125\n",
      " 0.91015625 0.90820312 0.90917969 0.90722656 0.9140625  0.90917969\n",
      " 0.92382812 0.91992188 0.91992188 0.91894531 0.91894531 0.91015625\n",
      " 0.92382812 0.92382812]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_accuracy(all_layers_begin_folders, comment=\"_begin\", xmax=3000, n_layers=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}