9_Appendix.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Introduction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Number of floating point operations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_FFNN_FLOPs(shape):\n",
    "    L = len(shape)\n",
    "    numflop = 0\n",
    "    for l in range(L-1):\n",
    "        numflop+= shape[l+1]*((2*shape[l])-1)\n",
    "    return numflop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1244198, 1710)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "shape0 = (784, 784, 10)\n",
    "shape1 = (10, 10, 10, 10, 10, 10, 10, 10, 10, 10)\n",
    "shape2 = (2, 3, 2, 2)\n",
    "shape3 = (3, 4, 3, 3)    \n",
    "\n",
    "get_FFNN_FLOPs(shape0), get_FFNN_FLOPs(shape1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feed-forward neural networks"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Common activation functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import spyder_utilities as su\n",
    "import matplotlib.pyplot as plt\n",
    "x = np.linspace(-5, 5, 1000)\n",
    "fig, ax = plt.subplots(1, 1, figsize=(6, 3))\n",
    "ax = su.config_axis(ax, x_lim=(-5, 5), y_lim=(-1.5, 1.5), Y_0=1, mult_y=1 , Eng=False)\n",
    "ax.plot(x, 1/(1+np.e**(-x)), linewidth=2, label=\"Sigmoid function\")\n",
    "ax.plot(x, np.tanh(x), linestyle=':', linewidth=2, label=\"Hyperbolic tangent function\")\n",
    "ax.plot(x, np.maximum(0, x), linestyle='--', linewidth=2, label=\"ReLu function\")\n",
    "ax.set_xlabel(\"input to neuron\")\n",
    "ax.set_ylabel(\"output of neuron\")\n",
    "ax.legend()\n",
    "plt.show()\n",
    "# su.save_fig(fig,\n",
    "#             image_nm=\"common_activation_functions\",\n",
    "#             paper_format=[\"pdf\", \"png\"], debug=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "toc-hr-collapsed": false
   },
   "source": [
    "## Neurons"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Hidden layer neuron"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\"\"\"\n",
    "Daniel Maidment\n",
    "\n",
    "Thu Oct 15 2019\n",
    "\"\"\"\n",
    "########################################################################\n",
    "import spyder_utilities as su\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from spnetwork.neuron import Neuron\n",
    "########################################################################\n",
    "\n",
    "seed = 100\n",
    "np.random.seed(seed)\n",
    "size = 10000\n",
    "N = 11\n",
    "\n",
    "fig, ax = plt.subplots(1, 1, figsize=(6, 6))\n",
    "ax = su.config_axis(ax, x_lim=(0, 1), y_lim=(0, 1), Eng=False)\n",
    "P = np.arange(0.01, 1, 0.05)\n",
    "\n",
    "for n in np.arange(1, N, 1):\n",
    "    tNeuron = Neuron(n)\n",
    "    p_x = np.empty(np.shape(P), dtype=float)\n",
    "    p_y = np.empty(np.shape(P), dtype=float)\n",
    "\n",
    "    for i in range(len(P)):\n",
    "        in_ar_1 = np.array(np.random.binomial(1, P[i], size),\n",
    "                           dtype=bool)\n",
    "        p_x[i] = np.mean(in_ar_1)\n",
    "\n",
    "        out_ar = np.zeros(size, bool)\n",
    "\n",
    "        for j in range(size):\n",
    "            out_ar[j] = tNeuron + in_ar_1[j]\n",
    "\n",
    "        p_y[i] = np.mean(out_ar)\n",
    "        tNeuron.reset()\n",
    "\n",
    "    label = f\"N = {n:2d}\"\n",
    "    ax.scatter(p_x, p_y, label=label)\n",
    "\n",
    "s = f\"Samples per point: {size}. Seed:{seed}\"\n",
    "plt.legend()\n",
    "ax.text(0, 1.01, s,\n",
    "        horizontalalignment='left',\n",
    "        verticalalignment='bottom',\n",
    "        transform=ax.transAxes)\n",
    "ax.set_xlabel(\"input frequency, normalised\")\n",
    "ax.set_ylabel(\"output frequency, normalised\")\n",
    "plt.show()\n",
    "su.save_fig(fig=fig, path = \"\\\\images\\\\neuron\",\n",
    "            image_nm=f\"linear_neuron_frequency_model_{N}_{size}\",\n",
    "            paper_format=['pdf', 'png'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "toc-hr-collapsed": false
   },
   "source": [
    "### Hidden layer neuron with feedback"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Frequency model of the hidden layer neuron with feedback for small values of NFF where NFB=NFF-1."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "path specified as: \\images\\neuron\\delayFB2\n"
     ]
    }
   ],
   "source": [
    "\"\"\"\n",
    "Daniel Maidment\n",
    "\n",
    "Thu Oct 24 2019\n",
    "\"\"\"\n",
    "########################################################################\n",
    "\n",
    "import spyder_utilities as su\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from spnetwork.neuron import DelayFBNeuron2\n",
    "\n",
    "########################################################################\n",
    "\n",
    "seed = 1000\n",
    "size = 10000\n",
    "N = 5\n",
    "\n",
    "fig, ax = plt.subplots(1, 1, figsize=(6, 6))\n",
    "ax = su.config_axis(ax, x_lim=(0, 1), y_lim=(0, 1), Eng=False)\n",
    "P = np.arange(0.01, 1, 0.05)\n",
    "\n",
    "for n in np.arange(2, N, 1):\n",
    "    tNeuron = DelayFBNeuron2(n, n-1)\n",
    "\n",
    "    p_x = np.empty(np.shape(P), dtype=float)\n",
    "    p_y = np.empty(np.shape(P), dtype=float)\n",
    "\n",
    "    for i in range(len(P)):\n",
    "        np.random.seed(seed)\n",
    "        in_ar = np.array(np.random.binomial(1, P[i], size), dtype=bool)\n",
    "        p_x[i] = np.mean(in_ar)\n",
    "\n",
    "        out_ar = np.zeros(size, bool)\n",
    "\n",
    "        for j in range(size):\n",
    "            out_ar[j] = tNeuron + in_ar[j]\n",
    "\n",
    "        p_y[i] = np.mean(out_ar)\n",
    "        tNeuron.reset()\n",
    "\n",
    "    label = f\"NFF = {n:2d}, NFB = {n-1:2d}\"\n",
    "    ax.scatter(p_x, p_y, label=label)\n",
    "\n",
    "s = f\"Samples per point: {size}. Seed:{seed}\"\n",
    "ax.legend()\n",
    "ax.text(0, 1.01, s,\n",
    "        horizontalalignment='left',\n",
    "        verticalalignment='bottom',\n",
    "        transform=ax.transAxes)\n",
    "ax.set_xlabel(\"input frequency, normalised\")\n",
    "ax.set_ylabel(\"output frequency, normalised\")\n",
    "plt.show()\n",
    "\n",
    "su.save_fig(fig=fig, path = \"\\\\images\\\\neuron\\\\delayFB2\",\n",
    "            image_nm=f\"feedback_neuron_frequency_model_smallNFF_smallNFB\",\n",
    "            paper_format=['pdf', 'png'], debug=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Frequency model of the hidden layer neuron with feedback where of NFF=4 and NFB is small."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "path specified as: \\images\\neuron\\delayFB2\n"
     ]
    }
   ],
   "source": [
    "\"\"\"\n",
    "Daniel Maidment\n",
    "\n",
    "Thu Nov 11\n",
    "\"\"\"\n",
    "########################################################################\n",
    "\n",
    "import spyder_utilities as su\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from spnetwork.neuron import DelayFBNeuron2\n",
    "\n",
    "########################################################################\n",
    "\n",
    "seed = 1000\n",
    "size = 10000\n",
    "N = 4\n",
    "\n",
    "fig, ax = plt.subplots(1, 1, figsize=(6, 6))\n",
    "ax = su.config_axis(ax, x_lim=(0, 1), y_lim=(0, 1), Eng=False)\n",
    "P = np.arange(0.01, 1, 0.05)\n",
    "\n",
    "for n in np.arange(1, N, 1):\n",
    "    tNeuron = DelayFBNeuron2(N, n)\n",
    "\n",
    "    p_x = np.empty(np.shape(P), dtype=float)\n",
    "    p_y = np.empty(np.shape(P), dtype=float)\n",
    "\n",
    "    for i in range(len(P)):\n",
    "        np.random.seed(seed)\n",
    "        in_ar = np.array(np.random.binomial(1, P[i], size), dtype=bool)\n",
    "        p_x[i] = np.mean(in_ar)\n",
    "\n",
    "        out_ar = np.zeros(size, bool)\n",
    "\n",
    "        for j in range(size):\n",
    "            out_ar[j] = tNeuron + in_ar[j]\n",
    "\n",
    "        p_y[i] = np.mean(out_ar)\n",
    "        tNeuron.reset()\n",
    "\n",
    "    slope = np.mean(np.gradient(p_y, p_x))\n",
    "    label = f\"NFF = {N:2d}, \"\n",
    "    label += f\"NFB = {n:2d}\"\n",
    "    ax.scatter(p_x, p_y, label=label)\n",
    "\n",
    "s = f\"Samples per point: {size}. Seed:{seed}\"\n",
    "ax.legend()\n",
    "ax.text(0, 1.01, s,\n",
    "        horizontalalignment='left',\n",
    "        verticalalignment='bottom',\n",
    "        transform=ax.transAxes)\n",
    "ax.set_xlabel(\"input frequency, normalised\")\n",
    "ax.set_ylabel(\"output frequency, normalised\")\n",
    "plt.show()\n",
    "\n",
    "su.save_fig(fig=fig, path = \"\\\\images\\\\neuron\\\\delayFB2\",\n",
    "            image_nm=f\"frequency_model_NFF4_smallNFB\",\n",
    "            paper_format=['pdf', 'png'], debug=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Time series model of hidden layer neuron with feedback"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x648 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\"\"\"\n",
    "Daniel Maidment\n",
    "\n",
    "Thu Oct 21 2019\n",
    "\"\"\"\n",
    "########################################################################\n",
    "import spyder_utilities as su\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from spnetwork.neuron import DelayFBNeuron2\n",
    "########################################################################\n",
    "np.random.seed(100)\n",
    "tn = neuron.DelayFBNeuron2(N=4, Nfb=2)\n",
    "N = 100\n",
    "p = 0.5\n",
    "nS = 4\n",
    "t = np.arange(0, N, 1)\n",
    "s = np.random.binomial(1, p, (N, nS))\n",
    "tnout = np.zeros(N, dtype=int)\n",
    "for i in range(N):\n",
    "    tn+s[i]\n",
    "    tnout[i] = int(tn.state)\n",
    "\n",
    "fig0, axarr = plt.subplots(nS+1, 1, figsize=(12, 1.5*(nS+nN)),\n",
    "                           sharex=True, sharey=True)\n",
    "for i in range(nS):\n",
    "    axarr[i] = su.binary_ax_config(N=N, ax=axarr[i])\n",
    "    axarr[i].step(t, s[:, i], where='post', linewidth=2,\n",
    "                  label=\"mean={:.3}\".format(np.mean(s[:, i])))\n",
    "    axarr[i].legend(loc=\"upper left\", bbox_to_anchor=(1,1))\n",
    "    axarr[i].set_title(f\"Input {i}\")\n",
    "\n",
    "axarr[-1] = su.binary_ax_config(N=N, ax=axarr[-1])\n",
    "axarr[-1].step(t, tnout, where='post', linewidth=2,\n",
    "               label=\"mean={:.3}\".format(np.mean(tnout[1])))\n",
    "axarr[-1].legend(loc=\"upper left\", bbox_to_anchor=(1,1))\n",
    "axarr[-1].set_title(\"Neuron with feedback Output\")\n",
    "axarr[-1].set_xlabel(\"Number of clock cycles\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Generate Neuron test bench files."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "########################################################################\n",
    "import spyder_utilities as su\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from spnetwork.neuron import DelayFBNeuron2\n",
    "########################################################################\n",
    "np.random.seed(100)\n",
    "tn = DelayFBNeuron2(N=4, Nfb=2)\n",
    "N = 100\n",
    "p = 0.4\n",
    "nS = 4\n",
    "t = np.arange(0, N, 1)\n",
    "s = np.random.binomial(1, p, (N, nS))\n",
    "tnout = np.zeros(N, dtype=int)\n",
    "\n",
    "input_vector_file = open(f\"Vividio\\\\input_vectors{p}.txt\", 'w')\n",
    "for i in range(N):\n",
    "    for j in range(nS):\n",
    "        input_vector_file.write(str(s[i][j]))\n",
    "    input_vector_file.write('\\n')\n",
    "input_vector_file.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "input_vector_file = open(f\"Vividio\\\\input_vectors{p}.txt\", 'r')\n",
    "neuron_output_file = open(f\"Vividio\\\\neuron_outputs{p}.txt\", 'w')\n",
    "\n",
    "tn = DelayFBNeuron2(N=4, Nfb=2)\n",
    "for line in input_vector_file:\n",
    "    synaptic_input = [int(i) for i in list(line.strip())]\n",
    "#     print(synaptic_input)\n",
    "    tn+synaptic_input\n",
    "    neuron_output_file.write(str(int(tn.state))+'\\n')\n",
    "\n",
    "input_vector_file.close()\n",
    "neuron_output_file.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x540 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "python_neuron = []\n",
    "# vhdl_neuron = []\n",
    "synapses = []\n",
    "\n",
    "input_vector_file = open(f\"Vividio\\\\input_vectors{p}.txt\", 'r')\n",
    "neuron_output_file = open(f\"Vividio\\\\neuron_outputs{p}.txt\", 'r')\n",
    "for line in input_vector_file:\n",
    "    synaptic_input = [int(i) for i in list(line.strip())]\n",
    "    neuron_output = int(neuron_output_file.readline().strip())\n",
    "    synapses.append(synaptic_input)\n",
    "    python_neuron.append(neuron_output)\n",
    "#     print(synaptic_input, neuron_output)\n",
    "    \n",
    "input_vector_file.close()\n",
    "neuron_output_file.close()\n",
    "synapses = np.array(synapses)\n",
    "python_neuron = np.array(python_neuron)\n",
    "\n",
    "fig0, axarr = plt.subplots(nS+1, 1, figsize=(12, 1.5*(nS+1)),\n",
    "                           sharex=True, sharey=True)\n",
    "for i in range(nS):\n",
    "    axarr[i] = su.binary_ax_config(N=N, ax=axarr[i])\n",
    "    axarr[i].step(t, synapses[:, i], where='post', linewidth=2,\n",
    "                  label=\"mean={:.3}\".format(np.mean(synapses[:, i])))\n",
    "    axarr[i].legend(loc=\"upper left\", bbox_to_anchor=(1,1))\n",
    "    axarr[i].set_title(f\"Synaptic Input {i}\")\n",
    "\n",
    "axarr[-1] = su.binary_ax_config(N=N, ax=axarr[-1])\n",
    "axarr[-1].step(t, python_neuron, where='post', linewidth=2,\n",
    "               label=\"mean={:.3}\".format(np.mean(python_neuron)))\n",
    "axarr[-1].legend(loc=\"upper left\", bbox_to_anchor=(1,1))\n",
    "axarr[-1].set_title(\"Neuron with feedback Output\")\n",
    "axarr[-1].set_xlabel(\"Number of clock cycles\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "########################################################################\n",
    "\n",
    "import spyder_utilities as su\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "########################################################################\n",
    "N=10000\n",
    "P = np.arange(0.01, 1.0, 0.01)\n",
    "signal_file = open(\"Vividio//neuron_input_multiple_frequencies0.01.txt\",\n",
    "                   'w')\n",
    "for p in P:\n",
    "    for i in range(N):\n",
    "        val = np.random.binomial(1, p)\n",
    "        signal_file.write(str(val)+'\\n')\n",
    "signal_file.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "99\n"
     ]
    }
   ],
   "source": [
    "p_x = P\n",
    "print(len(p_x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "########################################################################\n",
    "\n",
    "import spyder_utilities as su\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from spnetwork.neuron import DelayFBNeuron2\n",
    "\n",
    "########################################################################\n",
    "signal_file = open(\"Vividio//neuron_input_multiple_frequencies0.01.txt\",\n",
    "                   'r')\n",
    "python_neuron_data = open(\"Vividio//python_NeuronOutMultFreq0.01.txt\",\n",
    "                          'w')\n",
    "N = 10000\n",
    "P = np.arange(0.01, 1.0, 0.01)\n",
    "python_neuron = DelayFBNeuron2(4, 2)\n",
    "for i in range(len(P)):\n",
    "    for j in range(N):\n",
    "        neuron_input = bool(int(signal_file.readline().strip()))\n",
    "        python_neuron+neuron_input\n",
    "        neuron_output = int(python_neuron.state)\n",
    "        python_neuron_data.write(str(neuron_output)+'\\n')\n",
    "    python_neuron.reset()\n",
    "        \n",
    "signal_file.close()\n",
    "python_neuron_data.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Extract VHDL Neuron frequency information."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "########################################################################\n",
    "\n",
    "import spyder_utilities as su\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "########################################################################\n",
    "vhdl_neuron = []\n",
    "python_neuron = []\n",
    "vhdl_file = open(\"Vividio//vhdl_NeuronOutMultFreq0.01_part_reset.txt\",\n",
    "            'r')\n",
    "py_file = open(\"Vividio//python_NeuronOutMultFreq0.01.txt\",\n",
    "                          'r')\n",
    "for i in range(len(P)*N):\n",
    "    vhdl_line = vhdl_file.readline().strip()\n",
    "    py_line = py_file.readline().strip()\n",
    "    vno = int(vhdl_line)\n",
    "    pyno = int(py_line)\n",
    "    vhdl_neuron.append(vno)\n",
    "    python_neuron.append(pyno)\n",
    "vhdl_file.close()\n",
    "py_file.close()\n",
    "vhdl_neuron = np.array(vhdl_neuron)\n",
    "python_neuron = np.array(python_neuron)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([0.2878, 0.2839, 0.2901, 0.2826, 0.3018, 0.2948, 0.3194, 0.3136,\n",
       "        0.3327, 0.3304, 0.3285, 0.3141, 0.3421, 0.3304, 0.3396, 0.3651,\n",
       "        0.3681, 0.3669, 0.3652, 0.3762, 0.391 , 0.3776, 0.379 , 0.3941,\n",
       "        0.3904, 0.4038, 0.4067, 0.4051, 0.4238, 0.4131, 0.4297, 0.4427,\n",
       "        0.4328, 0.4327, 0.4384, 0.4484, 0.4626, 0.4569, 0.4651, 0.4647,\n",
       "        0.4619, 0.4759, 0.4698, 0.4856, 0.486 , 0.4906, 0.4938, 0.4914,\n",
       "        0.4934, 0.5016, 0.5051, 0.5064, 0.5091, 0.5117, 0.5149, 0.5023,\n",
       "        0.5217, 0.519 , 0.5203, 0.5256, 0.5271, 0.5372, 0.5354, 0.5444,\n",
       "        0.5353, 0.5403, 0.5436, 0.5422, 0.552 , 0.5498, 0.5468, 0.5498,\n",
       "        0.5558, 0.5531, 0.5601, 0.5575, 0.5555, 0.5588, 0.5598, 0.5563,\n",
       "        0.5614, 0.5592, 0.5582, 0.5556, 0.553 , 0.5546, 0.5511, 0.5514,\n",
       "        0.5482, 0.5482, 0.5464, 0.5458, 0.5367, 0.5353, 0.5325, 0.5279,\n",
       "        0.5199, 0.5116, 0.5041]),\n",
       " array([0.2396, 0.2816, 0.2971, 0.3025, 0.3107, 0.2955, 0.3171, 0.3179,\n",
       "        0.3255, 0.3237, 0.3445, 0.3271, 0.3451, 0.3516, 0.3531, 0.3627,\n",
       "        0.3714, 0.3665, 0.3767, 0.3705, 0.3692, 0.3889, 0.3865, 0.3877,\n",
       "        0.3948, 0.4182, 0.4155, 0.4159, 0.4249, 0.4228, 0.4212, 0.4367,\n",
       "        0.4394, 0.4269, 0.4422, 0.4595, 0.4519, 0.4487, 0.4616, 0.4714,\n",
       "        0.466 , 0.4697, 0.4806, 0.4723, 0.4831, 0.4902, 0.4911, 0.4964,\n",
       "        0.4937, 0.4998, 0.5013, 0.5039, 0.5086, 0.5063, 0.5164, 0.5156,\n",
       "        0.5189, 0.5254, 0.5211, 0.5236, 0.5313, 0.5328, 0.5378, 0.5363,\n",
       "        0.5422, 0.5404, 0.554 , 0.5468, 0.5535, 0.5508, 0.5562, 0.554 ,\n",
       "        0.5615, 0.5666, 0.5591, 0.5667, 0.566 , 0.5653, 0.5623, 0.5705,\n",
       "        0.5756, 0.5669, 0.5724, 0.5692, 0.5764, 0.572 , 0.5765, 0.5759,\n",
       "        0.5788, 0.5837, 0.5792, 0.5793, 0.5758, 0.5736, 0.5789, 0.5751,\n",
       "        0.5757, 0.5741, 0.5733]))"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "N=10000\n",
    "P = np.arange(0.01, 1.0, 0.01)\n",
    "p_x = P\n",
    "vhdl_neuron = vhdl_neuron.reshape(len(P), N)\n",
    "python_neuron = python_neuron.reshape(len(P), N)\n",
    "p_y_vhdl = np.zeros(len(P))\n",
    "p_y_py = np.zeros(len(P))\n",
    "for i in range(len(P)):\n",
    "    p_y_vhdl[i] = np.mean(vhdl_neuron[i])\n",
    "    p_y_py[i] = np.mean(python_neuron[i])\n",
    "p_y_py, p_y_vhdl"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "path specified as: \\images\\neuron\n"
     ]
    }
   ],
   "source": [
    "########################################################################\n",
    "\n",
    "import spyder_utilities as su\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "########################################################################\n",
    "fig2, ax = plt.subplots(1, 1, figsize=(6, 6))\n",
    "ax = su.config_axis(ax, x_lim=(0, 1), y_lim=(0, 1), Eng=False)\n",
    "\n",
    "ax.scatter(p_x, p_y_vhdl, label=\"VHDL neuron output\")\n",
    "ax.scatter(p_x, p_y_py, label=\"Python neuron output\")\n",
    "\n",
    "s = f\"Samples per point: {10000}.\\n\"\n",
    "ax.text(0.0, 1.01, s,\n",
    "        horizontalalignment='left',\n",
    "        verticalalignment='bottom',\n",
    "        transform=ax.transAxes)\n",
    "ax.set_xlabel(\"input frequency, normalised\")\n",
    "ax.set_ylabel(\"output frequency, normalised\")\n",
    "ax.legend()\n",
    "plt.show()\n",
    "su.save_fig(fig=fig2, path = \"\\\\images\\\\neuron\",\n",
    "            image_nm=f\"VHDL_Python_Comparison\",\n",
    "            paper_format=['pdf', 'png'], debug=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Generate python model for comparison"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Synapse"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Changing the synaptic seed frequency [$p_{seed}$] and input frequency for a fixed synaptic length [N]."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "path specified as: \\images\\synapse\n"
     ]
    }
   ],
   "source": [
    "\"\"\"\n",
    "Daniel Maidment\n",
    "\n",
    "Nov 27 2019\n",
    "\"\"\"\n",
    "########################################################################\n",
    "\n",
    "import spyder_utilities as su\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from spnetwork.synapse import Synapse\n",
    "\n",
    "########################################################################\n",
    "\n",
    "size = 10000\n",
    "N = 100\n",
    "\n",
    "fig2, ax = plt.subplots(1, 1, figsize=(6, 6))\n",
    "ax = su.config_axis(ax, x_lim=(0, 1), y_lim=(0, 1), Eng=False)\n",
    "P = np.arange(0.01, 1, 0.05)\n",
    "\n",
    "for p in np.arange(0.1, 1, 0.1):\n",
    "    testSynapse = Synapse(N=N, p=p)\n",
    "    p_x = np.empty(np.shape(P), dtype=float)\n",
    "    p_y = np.empty(np.shape(P), dtype=float)\n",
    "\n",
    "    for i in range(len(P)):\n",
    "        in_ar = np.array(np.random.binomial(1, P[i], size), dtype=bool)\n",
    "        out_ar = np.zeros(size, bool)\n",
    "\n",
    "        for j in range(size):\n",
    "            out_ar[j] = testSynapse*in_ar[j]\n",
    "\n",
    "        p_x[i] = np.mean(in_ar)\n",
    "        p_y[i] = np.mean(out_ar)\n",
    "\n",
    "    slope = np.mean(np.gradient(p_y, p_x))\n",
    "    label = \"[{:4.3}, {:4.3}, {:4.3}]\".format(p, testSynapse.getMean(), slope)\n",
    "    ax.scatter(p_x, p_y, label=label)\n",
    "\n",
    "s = f\"Samples per point: {size}. \"\n",
    "s += f\"Synaptic length: {N}.\\n\"\n",
    "s += r\"[ $p_{seed}$, $p_{synapse}$, $p_{result}$ ]\"\n",
    "plt.legend(bbox_to_anchor=(0.01, 1), loc = \"upper left\")\n",
    "ax.text(0, 1.01, s,\n",
    "        horizontalalignment='left',\n",
    "        verticalalignment='bottom',\n",
    "        transform=ax.transAxes)\n",
    "ax.set_xlabel(\"input frequency, normalised\")\n",
    "ax.set_ylabel(\"output frequency, normalised\")\n",
    "plt.show()\n",
    "su.save_fig(fig=fig2, path = \"\\\\images\\\\synapse\",\n",
    "            image_nm=f\"Synapse_frequency_model_N{N}\",\n",
    "            paper_format=['pdf', 'png'], debug=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Changing the input frequency and synaptic length [N] for a fixed synaptic frequency [$p_{seed}=0.5$]."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "path specified as: \\images\\synapse\n"
     ]
    }
   ],
   "source": [
    "\"\"\"\n",
    "Daniel Maidment\n",
    "\n",
    "Nov 27 2019\n",
    "\"\"\"\n",
    "########################################################################\n",
    "\n",
    "import spyder_utilities as su\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from spnetwork.synapse import Synapse\n",
    "\n",
    "########################################################################\n",
    "\n",
    "size = 10000\n",
    "N = 17\n",
    "p=[0.3, 0.7]\n",
    "seed = 101\n",
    "np.random.seed(seed)\n",
    "\n",
    "fig2, ax = plt.subplots(1, 1, figsize=(6, 6))\n",
    "i=1\n",
    "ax = su.config_axis(ax, x_lim=(0, 1), y_lim=(0, 1), Eng=False)\n",
    "P = np.arange(0.01, 1, 0.05)\n",
    "for n in np.arange(2, N, 2):\n",
    "    testSynapse = Synapse(N=n, p=p[i])\n",
    "    p_x = np.empty(np.shape(P), dtype=float)\n",
    "    p_y = np.empty(np.shape(P), dtype=float)\n",
    "\n",
    "    for k in range(len(P)):\n",
    "        in_ar = np.array(np.random.binomial(1, P[k], size), dtype=bool)\n",
    "        out_ar = np.zeros(size, bool)\n",
    "\n",
    "        for j in range(size):\n",
    "            out_ar[j] = testSynapse*in_ar[j]\n",
    "\n",
    "        p_x[k] = np.mean(in_ar)\n",
    "        p_y[k] = np.mean(out_ar)\n",
    "\n",
    "    slope = np.mean(np.gradient(p_y, p_x))\n",
    "    label = \"[{}, {:.3f}, {:.3f}]\".format(n, testSynapse.getMean(), slope)\n",
    "    ax.scatter(p_x, p_y, label=label)\n",
    "\n",
    "s = f\"Samples per point: {size}\\n\"\n",
    "s += r\"$p_{seed}=$\"+f\"{p[i]}, \"\n",
    "s += f\"seed = {seed}\\n\"\n",
    "s += r\"[ N, $p_{synapse}$, $p_{result}$ ]\"\n",
    "ax.legend(bbox_to_anchor=(0.01, 1), loc = \"upper left\")\n",
    "ax.text(0, 1.01, s,\n",
    "        horizontalalignment='left',\n",
    "        verticalalignment='bottom',\n",
    "        transform=ax.transAxes)\n",
    "ax.set_xlabel(\"input frequency, normalised\")\n",
    "ax.set_ylabel(\"output frequency, normalised\")\n",
    "plt.show()\n",
    "su.save_fig(fig=fig2, path = \"\\\\images\\\\synapse\",\n",
    "            image_nm=f\"Synapse_frequency_model_pseed{p[i]}_seed{seed}\",\n",
    "            paper_format=['pdf', 'png'], debug=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Generating test bench signals"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.1\n",
      "0.2\n",
      "0.30000000000000004\n",
      "0.4\n",
      "0.5\n",
      "0.6\n",
      "0.7000000000000001\n",
      "0.8\n",
      "0.9\n"
     ]
    }
   ],
   "source": [
    "########################################################################\n",
    "\n",
    "import spyder_utilities as su\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from spnetwork.synapse import Synapse\n",
    "\n",
    "########################################################################\n",
    "N=1000\n",
    "signal_file = open(\"Vividio//synapse_input_vector.txt\", 'w')\n",
    "for p in np.arange(0.1, 1.0, 0.1):\n",
    "    for i in range(N):\n",
    "        val = np.random.binomial(1, p)\n",
    "        signal_file.write(str(val)+'\\n')\n",
    "signal_file.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Generate list of synaptic seeds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "########################################################################\n",
    "\n",
    "import spyder_utilities as su\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from spnetwork.synapse import Synapse\n",
    "\n",
    "########################################################################\n",
    "\n",
    "seed = 1011\n",
    "np.random.seed(seed)\n",
    "N_S = 10\n",
    "P = np.arange(0.2, 1, 0.2)\n",
    "file = open(\"Vividio//synapse//synaptic_seeds.txt\", 'w')\n",
    "for p in P:\n",
    "    line = np.random.binomial(1, p, N_S)\n",
    "    for j in line:\n",
    "        file.write(str(j)+', ')\n",
    "    file.write('\\n')\n",
    "file.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Extract VHDL synapse frequency information."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.625\n"
     ]
    }
   ],
   "source": [
    "# print(np.random.binomial(1, 0.5, 16))\n",
    "reg_val = np.array([1, 1, 1, 1, 0, 1, 0, 1, 1, 1, 0, 1, 0, 1, 0, 0],\n",
    "                   dtype=int)\n",
    "print(np.mean(reg_val))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "vhdl_synapse0 = []\n",
    "vhdl_synapse1 = []\n",
    "vhdl_synapse2 = []\n",
    "vhdl_synapse3 = []\n",
    "file0 = open(\"Vividio//synapse//vhdl_synapse_output_0.txt\", 'r')\n",
    "file1 = open(\"Vividio//synapse//vhdl_synapse_output_1.txt\", 'r')\n",
    "file2 = open(\"Vividio//synapse//vhdl_synapse_output_2.txt\", 'r')\n",
    "file3 = open(\"Vividio//synapse//vhdl_synapse_output_3.txt\", 'r')\n",
    "for line0 in file0:\n",
    "    so = int(line0.strip())\n",
    "    vhdl_synapse0.append(so)\n",
    "for line1 in file1:\n",
    "    so = int(line1.strip())\n",
    "    vhdl_synapse1.append(so)\n",
    "for line2 in file2:\n",
    "    so = int(line2.strip())\n",
    "    vhdl_synapse2.append(so)\n",
    "for line3 in file3:\n",
    "    so = int(line3.strip())\n",
    "    vhdl_synapse3.append(so)\n",
    "\n",
    "file0.close()\n",
    "file1.close()\n",
    "file2.close()\n",
    "file3.close()\n",
    "vhdl_synapse0 = np.array(vhdl_synapse0)\n",
    "vhdl_synapse1 = np.array(vhdl_synapse1)\n",
    "vhdl_synapse2 = np.array(vhdl_synapse2)\n",
    "vhdl_synapse3 = np.array(vhdl_synapse3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]),\n",
       " array([0.015, 0.032, 0.078, 0.089, 0.095, 0.117, 0.141, 0.166, 0.181]))"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vhdl_synapse0 = vhdl_synapse0.reshape((9, 1000))\n",
    "vhdl_synapse1 = vhdl_synapse1.reshape((9, 1000))\n",
    "vhdl_synapse2 = vhdl_synapse2.reshape((9, 1000))\n",
    "vhdl_synapse3 = vhdl_synapse3.reshape((9, 1000))\n",
    "p_x = np.arange(0.1, 1, 0.1)\n",
    "p_y0 = np.zeros(9)\n",
    "p_y1 = np.zeros(9)\n",
    "p_y2 = np.zeros(9)\n",
    "p_y3 = np.zeros(9)\n",
    "for i in range(9):\n",
    "    p_y0[i] = np.mean(vhdl_synapse0[i])\n",
    "    p_y1[i] = np.mean(vhdl_synapse1[i])\n",
    "    p_y2[i] = np.mean(vhdl_synapse2[i])\n",
    "    p_y3[i] = np.mean(vhdl_synapse3[i])\n",
    "p_x, p_y0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "path specified as: \\images\\synapse\n"
     ]
    }
   ],
   "source": [
    "########################################################################\n",
    "\n",
    "import spyder_utilities as su\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from spnetwork.synapse import Synapse\n",
    "\n",
    "########################################################################\n",
    "fig2, ax = plt.subplots(1, 1, figsize=(6, 6))\n",
    "ax = su.config_axis(ax, x_lim=(0, 1), y_lim=(0, 1), Eng=False)\n",
    "\n",
    "slope0 = np.mean(np.gradient(p_y0, p_x))\n",
    "label = \"[{:.3f}, {:.3f}]\".format(np.mean([0, 0, 0, 0, 1, 0, 0, 0, 1, 0]), slope0)\n",
    "ax.scatter(p_x, p_y0, label=label)\n",
    "\n",
    "slope1 = np.mean(np.gradient(p_y1, p_x))\n",
    "label = \"[{:.3f}, {:.3f}]\".format(np.mean([0, 0, 0, 1, 0, 1, 0, 0, 1, 0]), slope1)\n",
    "ax.scatter(p_x, p_y1, label=label)\n",
    "\n",
    "slope2 = np.mean(np.gradient(p_y2, p_x))\n",
    "label = \"[{:.3f}, {:.3f}]\".format(np.mean([1, 0, 0, 1, 1, 1, 1, 1, 1, 1]), slope2)\n",
    "ax.scatter(p_x, p_y2, label=label)\n",
    "\n",
    "slope3 = np.mean(np.gradient(p_y3, p_x))\n",
    "label = \"[{:.3f}, {:.3f}]\".format(np.mean([1, 1, 1, 0, 1, 1, 0, 0, 0, 1]), slope3)\n",
    "ax.scatter(p_x, p_y3, label=label)\n",
    "\n",
    "s = f\"Samples per point: {1000}, \"\n",
    "# s += r\"$p_{seed}=$\"+f\"{p[i]}, \"\n",
    "s += f\"seed = {seed}\\n\"\n",
    "s += r\"[$p_{synapse}$, $p_{result}$ ]\"\n",
    "ax.legend(bbox_to_anchor=(0.01, 1), loc = \"upper left\")\n",
    "ax.text(0, 1.01, s,\n",
    "        horizontalalignment='left',\n",
    "        verticalalignment='bottom',\n",
    "        transform=ax.transAxes)\n",
    "ax.set_xlabel(\"input frequency, normalised\")\n",
    "ax.set_ylabel(\"output frequency, normalised\")\n",
    "plt.show()\n",
    "su.save_fig(fig=fig2, path = \"\\\\images\\\\synapse\",\n",
    "            image_nm=f\"VHDL_Synapse_frequency_model\",\n",
    "            paper_format=['pdf', 'png'], debug=True)"
   ]
  },
  {
   "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": "3.7.4"
  },
  "toc-autonumbering": true
 },
 "nbformat": 4,
 "nbformat_minor": 4
}