first_layer.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from copy import copy\n",
    "from six import iteritems\n",
    "from matplotlib import pyplot as plt\n",
    "from pygenn import genn_model, genn_wrapper"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load MNIST data (pre-normalized and coverted to numpy format)\n",
    "training_images = np.load(\"training_images.npy\")\n",
    "training_labels = np.load(\"training_labels.npy\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualize training data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(60000, 784)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x207ffa64c08>"
      ]
     },
     "execution_count": 14,
     "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": [
    "print(training_images.shape)\n",
    "assert np.allclose(np.sum(training_images, axis=1), 1.0)\n",
    "\n",
    "fig, axis = plt.subplots()\n",
    "axis.imshow(np.reshape(training_images[0], (28, 28)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "DT = 0.1\n",
    "INPUT_SCALE = 80.0\n",
    "NUM_PN = 784\n",
    "PRESENT_TIME_MS = 20.0\n",
    "\n",
    "# Standard LIF neurons parameters\n",
    "LIF_PARAMS = {\n",
    "    \"C\": 0.2,\n",
    "    \"TauM\": 20.0,\n",
    "    \"Vrest\": -60.0,\n",
    "    \"Vreset\": -60.0,\n",
    "    \"Vthresh\": -50.0,\n",
    "    \"Ioffset\": 0.0,\n",
    "    \"TauRefrac\": 2.0}\n",
    "\n",
    "# We only want PNs to spike once\n",
    "PN_PARAMS = copy(LIF_PARAMS)\n",
    "PN_PARAMS[\"TauRefrac\"] = 100.0\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Custom models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Current source model, allowing current to be injected into neuron from variable\n",
    "cs_model = genn_model.create_custom_current_source_class(\n",
    "    \"cs_model\",\n",
    "    var_name_types=[(\"magnitude\", \"scalar\")],\n",
    "    injection_code=\"$(injectCurrent, $(magnitude));\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model definition"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create model\n",
    "model = genn_model.GeNNModel(\"float\", \"mnist_mb_first_layer\")\n",
    "model.dT = DT\n",
    "model._model.set_seed(1337)\n",
    "\n",
    "# Create neuron populations\n",
    "lif_init = {\"V\": PN_PARAMS[\"Vreset\"], \"RefracTime\": 0.0}\n",
    "pn = model.add_neuron_population(\"pn\", NUM_PN, \"LIF\", PN_PARAMS, lif_init)\n",
    "\n",
    "# Turn on spike recording\n",
    "pn.spike_recording_enabled = True\n",
    "\n",
    "# Create current sources to deliver input to network\n",
    "pn_input = model.add_current_source(\"pn_input\", cs_model, pn , {}, {\"magnitude\": 0.0})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Build model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Concert present time into timesteps\n",
    "present_timesteps = int(round(PRESENT_TIME_MS / DT))\n",
    "\n",
    "# Build model and load it\n",
    "model.build()\n",
    "model.load(num_recording_timesteps=present_timesteps)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Simulation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "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": [
    "def reset_neuron(pop, var_init):\n",
    "    # Reset variables\n",
    "    for var_name, var_val in iteritems(var_init):\n",
    "        pop.vars[var_name].view[:] = var_val\n",
    "    \n",
    "    # Push the state to GPU\n",
    "    pop.push_state_to_device()\n",
    "    \n",
    "for s in range(4):\n",
    "    # Set training image\n",
    "    pn_input.vars[\"magnitude\"].view[:] = training_images[s] * INPUT_SCALE\n",
    "    pn_input.push_var_to_device(\"magnitude\")\n",
    "\n",
    "    # Simulate timesteps\n",
    "    for i in range(present_timesteps):\n",
    "        model.step_time()\n",
    "\n",
    "    # Reset neuron state for next stimuli\n",
    "    reset_neuron(pn, lif_init)\n",
    "\n",
    "    # Download spikes from GPU\n",
    "    model.pull_recording_buffers_from_device();\n",
    "\n",
    "    # Plot PN spikes\n",
    "    fig, axis = plt.subplots()\n",
    "    pn_spike_times, pn_spike_ids = pn.spike_recording_data\n",
    "    axis.scatter(pn_spike_times, pn_spike_ids, s=1)\n",
    "    axis.set_xlabel(\"Time [ms]\")\n",
    "    axis.set_ylabel(\"PN\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}