read_dataset.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Read MNIST dataset\n",
    "\n",
    "Dataset downloaded from *https://www.garrickorchard.com/datasets/n-mnist*\n",
    "\n",
    "event-Python repo downloaded from github: *https://github.com/gorchard/event-Python*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Event-based vision module imported\n"
     ]
    }
   ],
   "source": [
    "import sys\n",
    "from matplotlib import pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "sys.path.append('event-Python')\n",
    "\n",
    "import eventvision\n",
    "from spatio_temporal_feature import Feature"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "ev = eventvision.read_dataset(r'datasets\\mnist\\Test\\0\\00004.bin')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "ev.show_td(100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "rec.array([(18, 11, False,    581), (21, 13, False,   1703),\n",
       "           (12, 24, False,   2625), ..., (21, 19,  True, 307025),\n",
       "           (32, 10,  True, 307397), (10, 19,  True, 308529)],\n",
       "          dtype=[('x', '<u2'), ('y', '<u2'), ('p', '?'), ('ts', '<u8')])"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ev.data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7512be0>]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "plt.plot([v.ts for v in ev.data])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "False\n",
      "False\n",
      "False\n",
      "False\n",
      "True\n",
      "False\n",
      "True\n",
      "True\n",
      "False\n",
      "False\n",
      "True\n",
      "False\n",
      "True\n",
      "False\n",
      "False\n",
      "False\n",
      "False\n",
      "False\n",
      "False\n",
      "True\n",
      "False\n",
      "True\n",
      "False\n",
      "True\n",
      "False\n",
      "False\n",
      "False\n",
      "False\n",
      "True\n",
      "True\n",
      "False\n",
      "False\n",
      "False\n",
      "False\n",
      "False\n",
      "False\n",
      "True\n",
      "False\n",
      "False\n",
      "False\n",
      "False\n",
      "False\n",
      "True\n",
      "True\n",
      "False\n",
      "False\n",
      "False\n",
      "False\n",
      "True\n",
      "True\n",
      "False\n",
      "False\n",
      "True\n",
      "False\n",
      "False\n",
      "True\n",
      "True\n",
      "False\n",
      "False\n",
      "True\n",
      "False\n",
      "False\n",
      "False\n",
      "False\n",
      "True\n",
      "False\n",
      "False\n",
      "True\n",
      "False\n",
      "True\n",
      "False\n",
      "False\n",
      "True\n",
      "True\n",
      "True\n",
      "True\n",
      "True\n",
      "False\n",
      "False\n",
      "True\n",
      "True\n",
      "True\n",
      "True\n",
      "False\n",
      "False\n",
      "False\n",
      "True\n",
      "False\n",
      "True\n",
      "True\n",
      "True\n",
      "False\n",
      "True\n",
      "False\n",
      "False\n",
      "True\n",
      "False\n",
      "False\n",
      "False\n",
      "False\n"
     ]
    }
   ],
   "source": [
    "# plt.plot([v.p * 1 for v in ev.data])\n",
    "for i in range(100):\n",
    "    print ev.data[i].p"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2704"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len([1 for v in ev.data if not v.p])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plot time context $\\mathcal{T}_i(\\mathbf{u},p)$ and time surface $\\mathcal{S}_i(\\mathbf{u},p)$\n",
    "\n",
    "The time context is a 3D (2 spatial axes and 1 time) surface that indicates the most recent activations/events for each pixel at a given moment in time $i$ , within a small spatial region $\\mathbf{u}$:\n",
    "\n",
    "$$\\mathcal{T}_i(\\mathbf{u},p) = max_{j <= i}\\{t_j | \\mathbf{x}_j = (\\mathbf{x}_i + \\mathbf{u}), p_j = p\\}$$\n",
    "\n",
    "Note that we choose to either plot ON ($p_j = 1$) or OFF ($p_j = -1$) events, not both at the same time, just to aid visualisation.\n",
    "\n",
    "This time-context operation is essentially a form of spatio-temporal max-pooling.\n",
    "\n",
    "\n",
    "The time surface is the time context with an exponential decay applied:\n",
    "\n",
    "$$\\mathcal{S}_i(\\mathbf{u},p) = e^{-(t_i - \\mathcal{T}_i(\\mathbf{u},p) )/\\tau} $$\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,'Time surface')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# %%timeit\n",
    "\n",
    "# plot time context\n",
    "\n",
    "feat = Feature(ev.height, ev.width, region_size=1, time_constant=10000*2)\n",
    "\n",
    "# set time to pause at\n",
    "t_pause = 70000\n",
    "\n",
    "for e in ev.data:\n",
    "    if e.ts <= t_pause:\n",
    "        feat.process_event(e)\n",
    "            \n",
    "fig, ax = plt.subplots(1, 3, figsize=(10, 5))\n",
    "ax[0].imshow(feat.latest_times_on)\n",
    "ax[1].imshow(feat.time_context_on)\n",
    "ax[2].imshow(feat.time_surface_on)\n",
    "ax[0].set_title('Latest times')\n",
    "ax[1].set_title('Time context')\n",
    "ax[2].set_title('Time surface')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Initialise time surface prototypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x360 with 20 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Choose number of prototypes for layer 1\n",
    "N_1 = 10\n",
    "\n",
    "C_1 = [Feature(ev.height, ev.width, region_size=1, time_constant=10000*2)  for _ in range(N_1)]\n",
    "\n",
    "# initialise and plot each of the time surface prototypes\n",
    "\n",
    "fig, ax = plt.subplots(2, N_1, figsize=(25, 5))\n",
    "\n",
    "for i in range(N_1):\n",
    "    C_1[i].process_event(ev.data[i])\n",
    "\n",
    "    ax[0, i].imshow(C_1[i].time_surface_on)\n",
    "    ax[1, i].imshow(C_1[i].time_surface_off)\n",
    "    ax[0, i].set_title('Time surface {}'.format(i))\n",
    "    \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train time surface prototypes for layer 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[2, 1, 1, 1, 1, 1, 1, 1, 1, 1]"
      ]
     },
     "execution_count": 136,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# S = Feature(ev.height, ev.width, region_size=1, time_constant=10000*2)\n",
    "# p = [1] * N_1\n",
    "\n",
    "e = ev.data[1]\n",
    "S.process_event(e)\n",
    "\n",
    "\n",
    "\n",
    "# S.time_surface_off\n",
    "\n",
    "dists = [np.linalg.norm(c_k.time_surface_off - S.time_surface_off) for c_k in C_1]\n",
    "\n",
    "dists\n",
    "\n",
    "k = np.argmin(dists)\n",
    "\n",
    "k\n",
    "\n",
    "alpha = 0.01 / (1 + p[k] / 2000.)\n",
    "\n",
    "alpha\n",
    "\n",
    "beta = np.dot(C_1[k].time_surface_off, S.time_surface_off) / (np.linalg.norm(C_1[k].time_surface_off) * np.linalg.norm(S.time_surface_off))\n",
    "\n",
    "beta\n",
    "\n",
    "C_1[k].time_surface_off += alpha * (S.time_surface_off - beta * C_1[k].time_surface_off)\n",
    "\n",
    "C_1[k].time_surface_off\n",
    "\n",
    "p[k] += 1\n",
    "\n",
    "p"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "rec.array([(21, 12,  True, 19379), (20, 20, False, 19416),\n",
       "           (12, 20,  True, 19450), (21, 19, False, 19450),\n",
       "           (21, 18, False, 19655), (19,  9,  True, 19680),\n",
       "           (11, 13, False, 19903), ( 8, 19, False, 20072),\n",
       "           ( 8, 22, False, 20093), (17, 22, False, 20098),\n",
       "           (14, 20,  True, 20116), (18, 13, False, 20143),\n",
       "           (11, 14, False, 20198), (12, 24, False, 20206),\n",
       "           (15, 19,  True, 20278), (16,  4,  True, 20314),\n",
       "           ( 7, 15,  True, 20693), (10, 23, False, 20709),\n",
       "           (13,  7,  True, 20795), (12,  8,  True, 20924)],\n",
       "          dtype=[('x', '<u2'), ('y', '<u2'), ('p', '?'), ('ts', '<u8')])"
      ]
     },
     "execution_count": 171,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ev.data[90:110]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4\n",
      "((9, 14, False, 19146), [19.888186005268366, 18.091249703000905, 16.69005826690113, 14.370526362993957, 13.972092680664598, 14.471138436353641, 13.972092680664598, 13.972092680664598, 14.454904645463346, 14.452410694150084], 4, 0.009975062344139652, array([[nan, nan, nan, ..., nan, nan, nan],\n",
      "       [nan, nan, nan, ..., nan, nan, nan],\n",
      "       [nan, nan, nan, ..., nan, nan, nan],\n",
      "       ...,\n",
      "       [nan, nan, nan, ..., nan, nan, nan],\n",
      "       [nan, nan, nan, ..., nan, nan, nan],\n",
      "       [nan, nan, nan, ..., nan, nan, nan]]), [2, 2, 2, 3, 6, 13, 15, 15, 19, 21])\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Shaun\\Miniconda3\\envs\\neuro27_env\\lib\\site-packages\\ipykernel_launcher.py:34: RuntimeWarning: invalid value encountered in divide\n"
     ]
    }
   ],
   "source": [
    "# initialise time surface\n",
    "S = Feature(ev.height, ev.width, region_size=1, time_constant=10000*2)\n",
    "\n",
    "p = [1] * N_1\n",
    "\n",
    "for e in ev.data[:88]:\n",
    "    S.process_event(e)\n",
    "    \n",
    "    if e.p:\n",
    "        # find closest cluster center (i.e. closest time surface prototype, according to euclidean distance)\n",
    "    \n",
    "        dists = [np.linalg.norm(c_k.time_surface_on - S.time_surface_on) for c_k in C_1]\n",
    "\n",
    "        k = np.argmin(dists)\n",
    "\n",
    "        # update prototype that is closest to\n",
    "\n",
    "        alpha = 0.01 / (1 + p[k] / 2000.)\n",
    "        beta = np.dot(C_1[k].time_surface_on, S.time_surface_on) / (np.linalg.norm(C_1[k].time_surface_on) * np.linalg.norm(S.time_surface_on))\n",
    "\n",
    "        C_1[k].time_surface_on += alpha * (S.time_surface_on - beta * C_1[k].time_surface_on)\n",
    "        \n",
    "        p[k] += 1\n",
    "    else:\n",
    "        # find closest cluster center (i.e. closest time surface prototype, according to euclidean distance)\n",
    "    \n",
    "        dists = [np.linalg.norm(c_k.time_surface_off - S.time_surface_off) for c_k in C_1]\n",
    "\n",
    "        k = np.argmin(dists)\n",
    "\n",
    "        # update prototype that is closest to\n",
    "\n",
    "        alpha = 0.01 / (1 + p[k] / 2000.)\n",
    "        beta = np.dot(C_1[k].time_surface_off, S.time_surface_off) / (np.linalg.norm(C_1[k].time_surface_off) * np.linalg.norm(S.time_surface_off))\n",
    "\n",
    "        C_1[k].time_surface_off += alpha * (S.time_surface_off - beta * C_1[k].time_surface_off)\n",
    "        \n",
    "        p[k] += 1\n",
    "    \n",
    "print k\n",
    "print(e, dists, k, alpha, beta, p)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 193,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4"
      ]
     },
     "execution_count": 193,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "k"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABZgAAAE5CAYAAAAdokWcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAIABJREFUeJzt3X+w7Xdd3/vX2/xUIpKUHw1JBKbkdvBHCTYit3q9UzEDesdJuCMzUFuOFZp2qle4WseUTjvo9Qd6R9SxY20s9KRKCQj0Jm1pY0yxxdYJiRjAcAYSUMhpIoEb0vy4EJLwuX+sdcjmnH32Xt/3WXvtH+vxmMmctdZee38/+zPP7B/v/d3fXWOMAAAAAADAVF+12wsAAAAAAGB/MmAGAAAAAKDFgBkAAAAAgBYDZgAAAAAAWgyYAQAAAABoMWAGAAAAAKDlQAyYq+r1VfUvdnsdU1XVy6rqrqp6qKpesNvrWTe6oUM3dOiGDt0wlWbo0A0duqFDN3ToZn+oMcZur2FbVfXQhrtfk+SRJI/P7//dMcZbV7+qU1dVH0/yY2OM61Z4zEryxiSvmT/05iQ/OfZDCBPpZqnH/OtJ/kmSb0nyuTHGs1d17FXTzVKP+RNJDiV5VpLPJvn1Mcb/varjr5JulnrM1yX50SRPTfJQkrcn+YkxxmOrWsOq6GZHjn1mkg8lOWeMceGqj7/TNLPUY74hyT/KbA+P+StjjE+sag2ropulH/dbkvxKZl8XP5zk58YYv7rKNayCbpZ6zP+Q5H/Z8NCZST46xvjmVa1hVXSz1GOeleRXk7wsyRlJ/muSvzfG+O+rWsOq6Gapx3xKZt18z/yhXx9jvGEVxz59FQc5VWOMc47drqo/S/KaMcbv7d6KTk1VnT7/RvlZSW5f8eGvTHJFkucnGUluTPKJJL+x4nXsON0s1cNJ3pLkbUlev+Jjr5Rulnv4JK/KbNjzl5L8blXdNca4dsXr2HG6Wap/m+TwGOP+qjovyTszGzi/acXr2HG62RE/keTeJOds98T9SDNL9/Yxxt/cheOulG6WeuynJvmPSf7PzD4/nZnkwP0wK9HNMo0xvmfj/ar6/ST/aZVrWBXdLNVrk/zPSf5Kkv+R5DeT/FqS/33F69hxulmqX85sSP/sJE9PclNVfXKM8S93+sAH5RIZb6iq357ffnZVjar62/NT0T9XVX+vqr61qj5UVfdX1T897vV/qKqOzJ97Q1U96yTHObuqfruq/t/527mlqp4xf9mfVdV3b7OmV1fVp5K8b/4TmtOSfHD+U41U1VVV9fGqerCqPlJVLzvu+H9nvs5jL/+W+ePPrKp3VdVnqupPq+pHt9iuQ0l+aYxxdP6Tr19K8oNT9vug0M3i3Ywx3j/G+K3Mfhix1nQzqZtfHGN8YIzx2Bjjo0muS/Ltkzf9ANDNpG4+Psa4/9ibTPKlJM+dst8HhW4mfX2TqnpOkr+Z5OcnbfQBoplpzTCjm0nd/FiSG8YYbx1jPDLGeHCMcWTilh8Iuul9vKmqZ2d2NvNvLfL8g0Y3k7p5TmYfbz49xvhCkmuTfOOkDT8gdDOpm+9L8otjjP9vjPFnmV214Iem7HfXvjiDuenbklyc5DuTXJ/ZT5q/O7NfLfjjqvqdMcZ/rqorMjsj8/uS3JHkqszO0vxrm7zNQ0m+LslFmZ2yf0mSz09Y0/+a5HlJvjTG+HxVjSTPH2PcOX/5xzP7ZPPnSV6e5Ler6rljjHuq6uVJ3pDZ2ce3ZnY24KNV9VWZne11XZJXZvYT9N+rqo+OMW7YZA3fmOSDG+5/MGv6QeokdLN5N2xNN9t0U1U1P94/n/A+HHS6OUk3VfU3MvvNmq/N7PIqPz7hfTjodHPyjze/Nn+fp6x9HWjm5M18X1Xdl+SeJP90jPHPJrwPB51uNu/mRUk+XFX/LbMfft6c5IfHGJ+a8H4cZLrZ/nupVyV53xjjTye8Dwedbjbv5s1JfrWqnpnk/iQ/kOQ/THgfDjrdnPzjTR13+5smvA9tB+IM5pP4v8YYXxhj/G5mv97/tjHGvfOzdt+X5NgFtv9ukp8fYxyZn8L+c0kuOclPNB5N8heSPHeM8fgY44/GGA9MWNMbxhgPjzE2DXSM8TtjjLvHGF8aY7w9s/hfOH/xazL7KcQtY+bOMcYnk3xrkqeNMX56jPHFMbtu3G8mecVJ1nBOZr9eccz/SHLOfPiDbk7WDVvTzQLryexzzo7/as4+opuTGGP86zHGk5P8T5kNmj894X046HSzifkZIKePMf7NhHWvC81s7h2ZfRP4tCR/J8k/qapXTngfDjrdbO7CzAYQr03y9Un+NLNBBTO62d6rkhyesP51oJvNfSzJp5L89yQPZPY566cnvA8HnW429x+TXFVVX1tVz83s7OWvmfA+tB3kAfPGb0Y/v8n9Y9d4eVZmPxW6v6ruT3JfZhP+CzZ5m7+V5IYk11bV3VX1i1V1xoQ13bXVC6vqVVV124a1fFNmf+Qomf0E5eObvNqzkjzz2OvMX+/1SZ5xksM8lOTJG+4/OclDYxy8P/LXpBs6dLP1sX4ksy+m/7cxxiNbPXfN6GYbY4w7Mrtu2a9PeB8OOt2c+PaflOQXk/wfE9a8TjSziTHGR+bf5D0+xvhvmf1BnO+f8D4cdLrZ3OeT/Jv5AOALSX4qyV+rqq+b8H4cZLrZ+ljfkeQvZnb9bp6gm839syRnZzbwfFKSd8cZzBvpZnM/mtn7f0dmZz2/LcnRCe9D20G+RMai7krys2OBv0o5xng0sy8ifqpm1056T5KPZvarCw/nK38q8Bc3exMne9s1++nJbyZ5cZI/HGM8XlW35YlT2+/K7BT5zdb/p2OMi7db/9ztmf2Bv/fP7z8/u/eHePazdeuG5Vi7bqrqhzL7NaTvHGOs5BPbAbR23Rzn9JO8Xba2Tt1cnNkfMnlfzX4h68wkX1dVf57kRWN2/Tm2t07NbGZsOAaLW7duPnTcOo7d1s4069bNMYeSvHuM8dDE12Nm3bp5fpJ/NMa4b37cX0vy01X11DHGZxd8G6xZN/NefmDDcX8uT8z/dtRBPoN5Ub+R5B9W1TcmSVV9Xc2ufXKCqvrrVfXNVXVaZr+i8GiSx+cvvi3JK6rqjKq6NNPPgHhSZjF+Zn6sv52vvE7Kv0jyD6rqr9bMc+eBvj/JA1X1k1X11VV1WlV9U1V960mO86+S/FhVXVCza/n8ePyKTsdadVNVX1VVZ2d2PaOq2cXvz5y4Vtavmx/I7FeQLpv/Kg8969bNa6rq6fPb35DkHya5aeJaWa9u/iSzsz4umf/3mszOYrkk25xJwldYp2ZSVZdX1bnzt/HCzM74uW7iWlmzbjK71NfLquqSmp3V9o+T/MF44o/Tsph16yZV9dWZXXP18MQ18oR16+aWJK+av59nJPn7Se42XJ5srbqpqr9UVX9h/rzvSXJlkp+ZuNaWtR8wj9m1+n4hs1PgH8jsm5TvOcnTj/06ywNJjiT5z0l+e/6yf5zZTxs+l9lPPP71xHV8JMkvJfnDzL4p+uYk/3XDy38nyc/O3+6DSf6fJOeNMR7P7GLll2R2DbDPZhbmyX5N659ndoHwD8/f138ff3RrsjXs5jsz+zWL92R2vbnPJ/ndKWtlLbv5mcx+peuWqnpo/t9vTFkra9nNt2f2B5Qezuxjznsy+zUwJlinbsYYj40x/vzYf5n96uOX5vcfP/75bG6dmpl7RZI752/jXyX5hTHGNVPWyvp1M8b4T5l9Tvr3Se7N7A/9/Y0pa2X9upm7IrO/f/TeKWvkCWvYzT9I8oXMLnXwmSTfm+RlU9bKWnbzVzOb9z2Y5OeT/MAYYyVXLSiX3gUAAAAAoGPtz2AGAAAAAKDHgBkAAAAAgBYDZgAAAAAAWk5pwFxVL62qj1bVnVV11bIWxcGmGzp0Q4du6NANHbphKs3QoRs6dEOHbpii/Uf+quq0JB9LclmSo0luSfLK+V9G3NSZddY4O09qHY+96wt5OF8cj9Qiz9UNx+iGjp3sRjMH14P53GfHGE9b5Lm64RjdMJWvbejQDR26oUM3dCzazemncIwXJrlzjPGJJKmqa5NcnuSksZ2dJ+Xb6sWncEj2opvHTVOerhuS6IaenexGMwfX7413fnLC03VDEt0wna9t6NANHbqhQzd0LNrNqVwi44Ikd224f3T+2Feoqiur6taquvXRPHIKh+OA0A0duqFj2240wyZ0Q4dumMrXNnTohg7d0KEbJjmVM5g3Oz36hOttjDGuTnJ1kjy5zutdj4ODRDd06IaObbvRDJvQDR26YSpf29ChmxW44e7bvnz7Jc+8ZBdXsjS6oUM3THIqZzAfTXLRhvsXJrn71JbDGtANHbqhQzd06IYO3TCVZujQDR26oUM3THIqA+ZbklxcVc+pqjOTvCLJ9ctZFgeYbujQDR26oUM3dOiGqTRDh27o0A0dumGS9iUyxhiPVdWPJLkhyWlJ3jLGuH1pK+NA0g0duqFDN3Tohg7dMJVm6NANHbqhQzdMdSrXYM4Y4z1J3rOktbAmdEOHbujQDR26oUM3TKUZOnSz8w7IdZe/gm7o0A1TnMolMgAAAAAAWGMGzAAAAAAAtBgwAwAAAADQYsAMAAAAAECLATMAAAAAAC0GzAAAAAAAtBgwAwAAAADQYsAMAAAAAECLATMAAAAAAC0GzAAAAAAAtBgwAwAAAADQYsAMAAAAAECLATMAAAAAAC0GzAAAAAAAtBgwAwAAAADQYsAMAAAAAECLATMAAAAAAC0GzAAAAAAAtBgwAwAAAADQYsAMAAAAAECLATMAAAAAAC0GzAAAAAAAtBgwAwAAAADQYsAMAAAAAECLATMAAAAAAC0GzAAAAAAAtGw7YK6qt1TVvVX1JxseO6+qbqyqO+b/nruzy2S/0Q0duqFDN3Tohg7d0KEbptIMHbqhQzcsyyJnMB9O8tLjHrsqyU1jjIuT3DS/Dxsdjm6Y7nB0w3SHoxumOxzdMN3h6IbpDkc3THM4mmG6w9EN0x2ObliCbQfMY4z/kuS+4x6+PMk189vXJLliyetin9MNHbqhQzd06IYO3dChG6bSDB26oUM3LEv3GszPGGPckyTzf59+sidW1ZVVdWtV3fpoHmkejgNCN3Toho6FutEMx9ENHbqhQzdM5WtiOnRDh26YbMf/yN8Y4+oxxqVjjEvPyFk7fTgOCN3QoRum0gwduqFDN3Tohg7d0KEbOnTDMd0B86er6vwkmf977/KWxAGmGzp0Q4du6NANHbqhQzdMpRk6dEOHbpisO2C+Psmh+e1DSa5bznI44HRDh27o0A0duqFDN3Tohqk0Q4du6NANk207YK6qtyX5wyR/uaqOVtWrk7wxyWVVdUeSy+b34ct0Q4du6NANHbqhQzd06IapNEOHbujQDcty+nZPGGO88iQvevGS18IBohs6dEOHbujQDR26oUM3TKUZOnRDh25Ylh3/I38AAAAAABxMBswAAAAAALQYMAMAAAAA0GLADAAAAABAiwEzAAAAAAAtBswAAAAAALQYMAMAAAAA0GLADAAAAABAiwEzAAAAAAAtBswAAAAAALQYMAMAAAAA0GLADAAAAABAiwEzAAAAAAAtBswAAAAAALQYMAMAAAAA0GLADAAAAABAiwEzAAAAAAAtBswAAAAAALQYMAMAAAAA0GLADAAAAABAy+m7vQAAADgVN9x925dvv+SZl+ziSgAAYP04gxkAAAAAgBYDZgAAAAAAWlwiAwCAfc1lMWB/cDkbADiYnMEMAAAAAEDLtgPmqrqoqt5bVUeq6vaqeu388fOq6saqumP+77k7v1z2C93QoRs6dMNUmqFDN3Tohg7d0KEbOnTDsixyBvNjSX58jPG8JC9K8sNV9Q1Jrkpy0xjj4iQ3ze/DMbqhQzd06IapNEOHbujQDR26oUM3dOiGpdh2wDzGuGeM8YH57QeTHElyQZLLk1wzf9o1Sa7YqUWy/+iGDt3QoRum0gwduqFDN1/pJc+85Mv/cXK6oUM3dOiGZZl0DeaqenaSFyS5Ockzxhj3JLMgkzx92YvjYNANHbqhQzdMpRk6dEOHbujQDR26oUM3nIqFB8xVdU6SdyV53RjjgQmvd2VV3VpVtz6aRzprZB/TDR26oaPTjWbWm481dOiGDt3QoRs6dEOHbjhVCw2Yq+qMzEJ76xjj3fOHP11V589ffn6Sezd73THG1WOMS8cYl56Rs5axZvYJ3dChGzq63WhmfflYQ4du6NANHbqhQzd06IZl2HbAXFWV5M1Jjowx3rThRdcnOTS/fSjJdctfHvuVbujQDR26YSrN0KEbOnRDh27o0A0dumFZTl/gOd+e5G8l+XBV3TZ/7PVJ3pjkHVX16iSfSvLynVki+5Ru6NANHbphKs3QoRs6dEOHbujQDR26YSm2HTCPMf4gSZ3kxS9e7nI4KHRDh27o0A1TaYYO3dChGzp0Q4du6NANy7LwH/kDAAAAAICNDJgBAAAAAGgxYAYAAAAAoMWAGQAAAACAFgNmAAAAAABaDJgBAAAAAGgxYAYAAAAAoMWAGQAAAACAFgNmAAAAAABaDJgBAAAAAGgxYAYAAAAAoMWAGQAAAACAFgNmAAAAAABaDJgBAAAAAGgxYAYAAAAAoMWAGQAAAACAFgNmAAAAAABaDJgBAAAAAGgxYAYAAAAAoMWAGQAAAACAFgNmAAAAAABaDJgBAAAAAGgxYAYAAAAAoMWAGQAAAACAFgNmAAAAAABaDJgBAAAAAGjZdsBcVWdX1fur6oNVdXtV/dT88edU1c1VdUdVvb2qztz55bJf6IYO3dChGzp0w1SaoUM3dOiGDt3QoRuWZZEzmB9J8l1jjOcnuSTJS6vqRUl+IckvjzEuTvK5JK/euWWyD+mGDt3QoRs6dMNUmqFDN3Tohg7d0KEblmLbAfOYeWh+94z5fyPJdyV55/zxa5JcsSMrZF/SDR26oUM3dOiGqTRDh27o0A0duqFDNyzLQtdgrqrTquq2JPcmuTHJx5PcP8Z4bP6Uo0kuOMnrXllVt1bVrY/mkWWsmX1CN3Toho5uN5pZb7phKp+j6NANHbqhQzd06IZlWGjAPMZ4fIxxSZILk7wwyfM2e9pJXvfqMcalY4xLz8hZ/ZWy7+iGDt3Q0e1GM+tNN0zlcxQduqFDN3Tohg7dsAwLDZiPGWPcn+T3k7woyVOq6vT5iy5Mcvdyl8ZBoRs6dEOHbujQDVNphg7d0KEbOnRDh244FdsOmKvqaVX1lPntr07y3UmOJHlvku+fP+1Qkut2apHsP7qhQzd06IYO3TCVZujQDR26oUM3dOiGZTl9+6fk/CTXVNVpmQ2k3zHG+HdV9ZEk11bVzyT54yRv3sF1sv/ohg7d0KEbOnTDVJqhQzd06IYO3dChG5Zi2wHzGONDSV6wyeOfyOzaLHAC3dChGzp0Q4dumEozdOiGDt3QoRs6dMOyTLoGMwAAAAAAHGPADAAAAABAiwEzAAAAAAAtBswAAAAAALQYMAMAAAAA0GLADAAAAABAiwEzAAAAAAAtNcZY3cGqPpPkk0memuSzKzvw/rCf9+RZY4yn7dQbn3fzcPbv/uwk3ZyEbrakm034HLWt/bwvutk9+3lfdLM79vOerOJrG91sbj/viW52z37eE99L7R7dnIRutnTgu1npgPnLB626dYxx6coPvIfZk63Zn83Zl63Zn83Zl63Zn83Zl63Zn83Zl63ZnxPZk+3ZoxPZk+3ZoxPZk63Zn83Zl63Zn82tw764RAYAAAAAAC0GzAAAAAAAtOzWgPnqXTruXmZPtmZ/NmdftmZ/NmdftmZ/NmdftmZ/NmdftmZ/TmRPtmePTmRPtmePTmRPtmZ/NmdftmZ/Nnfg92VXrsEMAAAAAMD+5xIZAAAAAAC0rHTAXFUvraqPVtWdVXXVKo+9l1TVRVX13qo6UlW3V9Vr54+fV1U3VtUd83/P3e217gW60UyHbnTToRvddOhGN1NpZkY30+hmRjfT6GZGN9PoRjMdulnvblZ2iYyqOi3Jx5JcluRokluSvHKM8ZGVLGAPqarzk5w/xvhAVX1tkj9KckWSH0xy3xjjjfP/Gc8dY/zkLi511+lmRjPT6GZGN9PoZkY30+hmRjeL08wTdLM43TxBN4vTzRN0szjdzGhmGt3MrHM3qzyD+YVJ7hxjfGKM8cUk1ya5fIXH3zPGGPeMMT4wv/1gkiNJLshsP66ZP+2azCJcd7qJZhp0E9006Ca6adBNdDORZuZ0M4lu5nQziW7mdDOJbqKZBt1kvbtZ5YD5giR3bbh/dP7YWquqZyd5QZKbkzxjjHFPMosyydN3b2V7hm6Oo5mF6OY4ulmIbo6jm4Xo5ji62ZZmNqGbbelmE7rZlm42oZtt6eY4mlmIbo6zbt2scsBcmzy2mutz7FFVdU6SdyV53Rjjgd1ezx6lmw00szDdbKCbhelmA90sTDcb6GYhmjmObhaim+PoZiG6OY5uFqKbDTSzMN1ssI7drHLAfDTJRRvuX5jk7hUef0+pqjMyi+2tY4x3zx/+9Px6Lceu23Lvbq1vD9HNnGYm0c2cbibRzZxuJtHNnG4WppkNdLMw3Wygm4XpZgPdLEw3c5qZRDdz69rNKgfMtyS5uKqeU1VnJnlFkutXePw9o6oqyZuTHBljvGnDi65Pcmh++1CS61a9tj1IN9FMg26imwbdRDcNuoluJtLMnG4m0c2cbibRzZxuJtFNNNOgm6x3NzXG6s5Yr6rvTfIrSU5L8pYxxs+u7OB7SFV9R5L3Jflwki/NH359ZtdleUeSr0/yqSQvH2PctyuL3EN0o5kO3eimQze66dCNbqbSzIxuptHNjG6m0c2MbqbRjWY6dLPe3ax0wAwAAAAAwMGxyktkAAAAAABwgBgwAwAAAADQYsAMAAAAAECLATMAAAAAAC0GzAAAAAAAtBgwAwAAAADQYsAMAAAAAECLATMAAAAAAC0GzAAAAAAAtBgwAwAAAADQYsAMAAAAAECLATMAAAAAAC2nNGCuqpdW1Uer6s6qumpZi+Jg0w0duqFDN3Tohg7dMJVm6NANHbqhQzdMUWOM3itWnZbkY0kuS3I0yS1JXjnG+MjylsdBoxs6dEOHbujQDR26YSrN0KEbOnRDh26Y6vRTeN0XJrlzjPGJJKmqa5NcnuSksZ1ZZ42z86RTOCR70RfycL44HqkFn64bkuiGnp3sRjMH14P53GfHGE9b8Om6IYlumM7XNnTohg7d0KEbOhbt5lQGzBckuWvD/aNJvu34J1XVlUmuTJKz8zX5tnrxKRySvejmcdOUp+uGJLqhZ9ndaGY9/N545ycnPF03JNEN0/nahg7d0KEbOnRDx6LdnMo1mDebXp9wvY0xxtVjjEvHGJeekbNO4XAcELqhQzd0bNuNZtiEbujQDVP52oYO3dChGzp0wySnMmA+muSiDfcvTHL3qS2HNaAbOnRDh27o0A0dumEqzdChGzp0Q4dumORUBsy3JLm4qp5TVWcmeUWS65ezLA4w3dChGzp0Q4du6NANU2mGDt3QoRs6dMMk7WswjzEeq6ofSXJDktOSvGWMcfvSVsaBpBs6dEOHbujQDR26YSrN0KEbOnRDh26Y6lT+yF/GGO9J8p4lrYU1oRs6dEOHbujQDR26YSrN0KEbOnRDh26Y4lQukQEAAAAAwBozYAYAAAAAoMWAGQAAAACAFgNmAAAAAABaDJgBAAAAAGgxYAYAAAAAoMWAGQAAAACAFgNmAAAAAABaDJgBAAAAAGgxYAYAAAAAoMWAGQAAAACAFgNmAAAAAABaDJgBAAAAAGgxYAYAAAAAoMWAGQAAAACAFgNmAAAAAABaDJgBAAAAAGgxYAYAAAAAoMWAGQAAAACAFgNmAAAAAABaDJgBAAAAAGgxYAYAAAAAoMWAGQAAAACAFgNmAAAAAABaDJgBAAAAAGgxYAYAAAAAoGXbAXNVvaWq7q2qP9nw2HlVdWNV3TH/99ydXSb7jW7o0A0duqFDN3Tohg7dMJVm6NANHbphWRY5g/lwkpce99hVSW4aY1yc5Kb5fdjocHTDdIejG6Y7HN0w3eHohukORzeT3HD3bV/x35o6HN0wzeFohukORzdMdzi6YQm2HTCPMf5LkvuOe/jyJNfMb1+T5Iolr4t9Tjd06IYO3dChGzp0Q4dumEozdOiGDt2wLN1rMD9jjHFPksz/ffrJnlhVV1bVrVV166N5pHk4Dgjd0KEbOhbqRjMcRzd06IYO3TCVr4np0A0dumGy03f6AGOMq5NcnSRPrvPGTh9vr9r4K4EveeYlu7iS/UE3dOiGqTRDh27oWNdufN17ata1G06NbujQDR264ZjuGcyfrqrzk2T+773LWxIHmG7o0A0duqFDN3Tohg7dMJVm6NANHbphsu6A+fokh+a3DyW5bjnL4YDTDR26oUM3dOiGDt3QoRum0gwduqFDN0y27YC5qt6W5A+T/OWqOlpVr07yxiSXVdUdSS6b34cv0w0duqFDN3Tohg7d0KEbptIMHbqhQzcsy7bXYB5jvPIkL3rxktdyoK3b9ed0Q4du6NANHbqhQzd06IapNEOHbujQDcvSvUQGAAAAAABrzoAZAAAAAICWbS+RAex9N9x921fcX7dLsgCr4WMNAACrtPHrT197wt7lDGYAAAAAAFoMmAEAAAAAaDFgBgAAAACgxTWY4QBwLSpgFXysAQBglXz9CfuDM5gBAAAAAGgxYAYAAAAAoMWAGQAAAACAFgNmAAAAAABaDJgBAAAAAGgxYAYAAAAAoMWAGQAAAACAFgNmAAAAAABaDJgBAAAAAGgxYAYAAAAAoMWAGQAAAACAFgNmAAAAAABaDJgBAAAAAGgxYAYAAAAAoMWAGQAAAACAFgNmAAAAAABaDJgBAAAAAGgxYAYAAAAAoMWAGQAAAACAlm0HzFV1UVW9t6qOVNXtVfXa+ePnVdWNVXXH/N9zd3657Be6oUM3dOiGqTRDh27o0A0duqFDN3TohmVZ5Azmx5L8+BjjeUlelOSHq+obklyV5KYxxsVJbprfh2N0Q4du6NANU2mGDt3QoRs6dEOHbujQDUux7YB5jHHPGOMD89sPJjmS5IIR/RGkAAAI6ElEQVQklye5Zv60a5JcsVOLZP/RDR26oUM3TKUZOnRDh27o0A0duqFDNyzLpGswV9Wzk7wgyc1JnjHGuCeZBZnk6Sd5nSur6taquvXRPHJqq2Vf0g0duqFjajeawccaOnRDh27o0A0duqFDN5yKhQfMVXVOkncled0Y44FFX2+McfUY49IxxqVn5KzOGtnHdEOHbujodKOZ9eZjDR26oUM3dOiGDt3QoRtO1UID5qo6I7PQ3jrGePf84U9X1fnzl5+f5N6dWSL7lW7o0A0dumEqzdChGzp0Q4du6NANHbphGbYdMFdVJXlzkiNjjDdteNH1SQ7Nbx9Kct3yl8d+pRs6dEOHbphKM3Tohg7d0KEbOnRDh25YltMXeM63J/lbST5cVbfNH3t9kjcmeUdVvTrJp5K8fGeWyD6lGzp0Q4dumEozdOiGDt3QoRs6dEOHbliKbQfMY4w/SFInefGLl7scDgrd0KEbOnTDVJqhQzd06IYO3dChGzp0w7Is/Ef+AAAAAABgIwNmAAAAAABaFrkGMwAAAADAnnfD3bd9+fZLnnnJLq5kfTiDGQAAAACAFgNmAAAAAABaDJgBAAAAAGhxDWaAXeTaUAAAALA8vrdePWcwAwAAAADQYsAMAAAAAECLS2QA7CK/ugMAAADsZ85gBgAAAACgxYAZAAAAAIAWA2YAAAAAAFoMmAEAAAAAaDFgBgAAAACgxYAZAAAAAIAWA2YAAAAAAFoMmAEAAAAAaDFgBgAAAACgxYAZAAAAAIAWA2YAAAAAAFoMmAEAAAAAaDFgBgAAAACgxYAZAAAAAIAWA2YAAAAAAFoMmAEAAAAAaNl2wFxVZ1fV+6vqg1V1e1X91Pzx51TVzVV1R1W9varO3Pnlsl/ohg7d0KEbOnTDVJqhQzd06IYO3dChG5ZlkTOYH0nyXWOM5ye5JMlLq+pFSX4hyS+PMS5O8rkkr965ZbIP6YYO3dChGzp0w1SaoUM3dOiGDt3QoRuWYtsB85h5aH73jPl/I8l3JXnn/PFrklyxIytkX9INHbqhQzd06IapNEOHbujQDR26oUM3LMtC12CuqtOq6rYk9ya5McnHk9w/xnhs/pSjSS44yeteWVW3VtWtj+aRZayZfUI3dOiGjm43mllvumEqn6Po0A0duqFDN3TohmVYaMA8xnh8jHFJkguTvDDJ8zZ72kle9+oxxqVjjEvPyFn9lbLv6IYO3dDR7UYz6003TOVzFB26oUM3dOiGDt2wDAsNmI8ZY9yf5PeTvCjJU6rq9PmLLkxy93KXxkGhGzp0Q4du6NANU2mGDt3QoRs6dEOHbjgV2w6Yq+ppVfWU+e2vTvLdSY4keW+S758/7VCS63Zqkew/uqFDN3Tohg7dMJVm6NANHbqhQzd06IZlOX37p+T8JNdU1WmZDaTfMcb4d1X1kSTXVtXPJPnjJG/ewXWy/+iGDt3QoRs6dMNUmqFDN3Tohg7d0KEblmLbAfMY40NJXrDJ45/I7NoscALd0KEbOnRDh26YSjN06IYO3dChGzp0w7JMugYzAAAAAAAcY8AMAAAAAECLATMAAAAAAC0GzAAAAAAAtBgwAwAAAADQYsAMAAAAAEBLjTFWd7CqzyT5ZJKnJvnsyg68P+znPXnWGONpO/XG5908nP27PztJNyehmy3pZhM+R21rP++LbnbPft4X3eyO/bwnq/jaRjeb2897opvds5/3xPdSu0c3J6GbLR34blY6YP7yQatuHWNcuvID72H2ZGv2Z3P2ZWv2Z3P2ZWv2Z3P2ZWv2Z3P2ZWv250T2ZHv26ET2ZHv26ET2ZGv2Z3P2ZWv2Z3PrsC8ukQEAAAAAQIsBMwAAAAAALbs1YL56l467l9mTrdmfzdmXrdmfzdmXrdmfzdmXrdmfzdmXrdmfE9mT7dmjE9mT7dmjE9mTrdmfzdmXrdmfzR34fdmVazADAAAAALD/uUQGAAAAAAAtBswAAAAAALSsdMBcVS+tqo9W1Z1VddUqj72XVNVFVfXeqjpSVbdX1Wvnj59XVTdW1R3zf8/d7bXuBbrRTIdudNOhG9106EY3U2lmRjfT6GZGN9PoZkY30+hGMx26We9uVnYN5qo6LcnHklyW5GiSW5K8cozxkZUsYA+pqvOTnD/G+EBVfW2SP0pyRZIfTHLfGOON8/8Zzx1j/OQuLnXX6WZGM9PoZkY30+hmRjfT6GZGN4vTzBN0szjdPEE3i9PNE3SzON3MaGYa3cysczerPIP5hUnuHGN8YozxxSTXJrl8hcffM8YY94wxPjC//WCSI0kuyGw/rpk/7ZrMIlx3uolmGnQT3TToJrpp0E10M5Fm5nQziW7mdDOJbuZ0M4luopkG3WS9u1nlgPmCJHdtuH90/thaq6pnJ3lBkpuTPGOMcU8yizLJ03dvZXuGbo6jmYXo5ji6WYhujqObhejmOLrZlmY2oZtt6WYTutmWbjahm23p5jiaWYhujrNu3axywFybPLaa63PsUVV1TpJ3JXndGOOB3V7PHqWbDTSzMN1soJuF6WYD3SxMNxvoZiGaOY5uFqKb4+hmIbo5jm4WopsNNLMw3Wywjt2scsB8NMlFG+5fmOTuFR5/T6mqMzKL7a1jjHfPH/70/Hotx67bcu9urW8P0c2cZibRzZxuJtHNnG4m0c2cbhammQ10szDdbKCbhelmA90sTDdzmplEN3Pr2s0qB8y3JLm4qp5TVWcmeUWS61d4/D2jqirJm5McGWO8acOLrk9yaH77UJLrVr22PUg30UyDbqKbBt1ENw26iW4m0sycbibRzZxuJtHNnG4m0U0006CbrHc3Ncbqzlivqu9N8itJTkvyljHGz67s4HtIVX1Hkvcl+XCSL80ffn1m12V5R5KvT/KpJC8fY9y3K4vcQ3SjmQ7d6KZDN7rp0I1uptLMjG6m0c2MbqbRzYxuptGNZjp0s97drHTADAAAAADAwbHKS2QAAAAAAHCAGDADAAAAANBiwAwAAAAAQIsBMwAAAAAALQbMAAAAAAC0GDADAAAAANBiwAwAAAAAQMv/D759ScW+PpjlAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1800x360 with 20 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(2, N_1, figsize=(25, 5))\n",
    "\n",
    "for i in range(N_1):\n",
    "    C_1[i].process_event(ev.data[i])\n",
    "\n",
    "    ax[0, i].imshow(layer_1[i].time_surface_on)\n",
    "    ax[1, i].imshow(layer_1[i].time_surface_off)\n",
    "    ax[0, i].set_title('Time surface {}'.format(i))\n",
    "    \n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.15"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}