NematodeWiring.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "9848b0fd-8cd2-4be9-86f2-33913b7c90e6",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import os\n",
    "from tensorflow import keras\n",
    "from ncps import wirings\n",
    "from ncps.tf import LTC\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "df420e14-be84-41a0-a16b-f3bf2440ef3d",
   "metadata": {},
   "outputs": [],
   "source": [
    "m5long_step_4 = pd.read_csv(\"final_complete_m5long_timestep_4.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f54d0925-45a5-4dd7-b360-bb72a2d82a12",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "f1869ccc-ec2a-4623-a71e-2454b363e3de",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.99990315, 0.99988647, 0.99988473, 0.99987063, 0.99986044,\n",
       "       0.99989704, 0.99994875, 0.99994731, 0.99995094, 1.        ,\n",
       "       0.99995313, 0.9993544 , 0.99934512, 0.99934557, 0.99933961,\n",
       "       0.99935349, 0.99935464, 0.99934716, 0.99934867, 0.99936852,\n",
       "       0.99934383, 0.99931553, 0.99929583, 0.9992935 , 0.999299  ,\n",
       "       0.99935502, 0.99933485, 0.99934942, 0.99937833, 0.99937493,\n",
       "       0.99937245, 0.99938166, 0.99941531, 0.99942385, 0.99941607,\n",
       "       0.9994348 , 0.99944537, 0.99940603, 0.99945518, 0.99944234,\n",
       "       0.99941743, 0.99938716, 0.99937425, 0.99935961, 0.99937472,\n",
       "       0.99942641, 0.99941359, 0.99936459, 0.99903505, 0.99906991,\n",
       "       0.99908917, 0.99910751, 0.99916789, 0.99914502, 0.99911468,\n",
       "       0.99911679, 0.99911974, 0.99917544, 0.99919514, 0.99921892,\n",
       "       0.99923393, 0.99918434, 0.99916109, 0.99913959, 0.99913868,\n",
       "       0.99912623, 0.99912509, 0.99912396, 0.99910842, 0.99907543,\n",
       "       0.99911422, 0.99913114, 0.99912827, 0.99909807, 0.99912592,\n",
       "       0.99912796, 0.99912849, 0.99912608, 0.99911626, 0.99910207,\n",
       "       0.99915152, 0.99914978, 0.99914487, 0.99914154, 0.99914751,\n",
       "       0.9991623 , 0.99887095, 0.99889163, 0.99889155, 0.99887337,\n",
       "       0.99884113, 0.99882633, 0.99885593, 0.99887042, 0.99891465,\n",
       "       0.99892673, 0.998937  , 0.99891443, 0.99891042, 0.99888469,\n",
       "       0.99886431])"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a1 = list(m5long_step_4.loc[:100][\"MA 18\"])\n",
    "a1 = np.array(a1)\n",
    "# a1 / (np.max(a1) - np.min(a1))\n",
    "a1/np.max(a1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "0b321748-26dc-410c-93c4-6a8cafc6c008",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7360.1828 7351.5444\n"
     ]
    }
   ],
   "source": [
    "print(np.max(a1), np.min(a1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "c334d2a2-e10c-4b24-8aa3-9274cf635339",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_4\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " ltc_4 (LTC)                 (None, None, 1)           350       \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 350\n",
      "Trainable params: 350\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "Epoch 1/50\n",
      "1/1 [==============================] - 8s 8s/step - loss: 3.0085\n",
      "Epoch 2/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 2.8227\n",
      "Epoch 3/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 2.0238\n",
      "Epoch 4/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 1.6530\n",
      "Epoch 5/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 1.3935\n",
      "Epoch 6/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 1.1847\n",
      "Epoch 7/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 1.0092\n",
      "Epoch 8/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.8584\n",
      "Epoch 9/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.7270\n",
      "Epoch 10/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.6111\n",
      "Epoch 11/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.5080\n",
      "Epoch 12/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.4162\n",
      "Epoch 13/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.3346\n",
      "Epoch 14/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.2633\n",
      "Epoch 15/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.2024\n",
      "Epoch 16/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.1521\n",
      "Epoch 17/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.1121\n",
      "Epoch 18/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0815\n",
      "Epoch 19/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0585\n",
      "Epoch 20/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0416\n",
      "Epoch 21/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0293\n",
      "Epoch 22/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0202\n",
      "Epoch 23/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0135\n",
      "Epoch 24/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0086\n",
      "Epoch 25/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0052\n",
      "Epoch 26/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0028\n",
      "Epoch 27/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0013\n",
      "Epoch 28/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 4.5395e-04\n",
      "Epoch 29/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 1.6857e-04\n",
      "Epoch 30/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 2.8329e-04\n",
      "Epoch 31/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 6.8508e-04\n",
      "Epoch 32/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0013\n",
      "Epoch 33/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0020\n",
      "Epoch 34/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0027\n",
      "Epoch 35/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0034\n",
      "Epoch 36/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0041\n",
      "Epoch 37/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0047\n",
      "Epoch 38/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0051\n",
      "Epoch 39/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0055\n",
      "Epoch 40/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0057\n",
      "Epoch 41/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0057\n",
      "Epoch 42/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0057\n",
      "Epoch 43/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0055\n",
      "Epoch 44/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0053\n",
      "Epoch 45/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0049\n",
      "Epoch 46/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0046\n",
      "Epoch 47/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0041\n",
      "Epoch 48/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0037\n",
      "Epoch 49/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0032\n",
      "Epoch 50/50\n",
      "1/1 [==============================] - 1s 1s/step - loss: 0.0028\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "a1 = np.array(list(m5long_step_4.loc[:1000][\"MA 18\"]))\n",
    "a2 = np.array(list(m5long_step_4.loc[:1000][\"MA 50\"]))\n",
    "a = np.expand_dims(np.stack([a1/np.max(a1), a2/np.max(a2)], axis=1), axis = 0)\n",
    "b = np.array(list(m5long_step_4.loc[:1000][\"Ask\"]))\n",
    "b = b/np.max(b)\n",
    "b=b.reshape(1,len(b), 1)\n",
    "wiring = wirings.AutoNCP(8,1) # 8 neurons in total, 1 output (motor neuron)\n",
    "\n",
    "model = keras.models.Sequential(\n",
    "    [\n",
    "        keras.layers.InputLayer(input_shape=(None, 2)),\n",
    "        # here we could potentially add layers before and after the LTC network\n",
    "        LTC(wiring, return_sequences=True),\n",
    "    ]\n",
    ")\n",
    "model.compile(\n",
    "    optimizer=keras.optimizers.Adam(0.01), loss='mean_squared_error'\n",
    ")\n",
    "\n",
    "model.summary()\n",
    "hist = model.fit(x=a, y=b, batch_size=1, epochs=50,verbose=1)\n",
    "# How does the trained model now fit to the sinusoidal function?\n",
    "prediction = model(a).numpy()\n",
    "plt.figure(figsize=(6, 4))\n",
    "plt.plot(b[0, :, 0], label=\"Target output\")\n",
    "plt.plot(prediction[0, :, 0], label=\"LTC output\",linestyle=\"dashed\")\n",
    "plt.ylim((-1, 1))\n",
    "plt.legend(loc=\"upper right\")\n",
    "plt.title(\"After training\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1e5a3697-e1ef-434d-8aca-3d63ac166a9d",
   "metadata": {},
   "outputs": [],
   "source": [
    "a"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "035ae2e7-872f-4a41-8394-cf34795ed3a7",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "import matplotlib.pyplot as plt\n",
    "a1 = np.array(list(m5long_step_4.loc[:100][\"MA 18\"]))\n",
    "a2 = np.array(list(m5long_step_4.loc[:100][\"MA 50\"]))\n",
    "a = np.expand_dims( np.stack([1 - ( (np.max(a1) - a1)/np.max(a1) ) , 1 - (np.max(a2) - a2)/np.max(a2)], axis=1), axis = 0)\n",
    "b = np.array(list(m5long_step_4.loc[:100][\"Ask\"]))\n",
    "b = 1 - (np.max(b) - b)/np.max(b)\n",
    "b=b.reshape(1,len(b), 1)\n",
    "hist = model.fit(x=a, y=b, batch_size=1, epochs=50,verbose=1)\n",
    "\n",
    "prediction = model(a).numpy()\n",
    "plt.figure(figsize=(6, 4))\n",
    "plt.plot(b[0, :, 0], label=\"Target output\")\n",
    "plt.plot(prediction[0, :, 0], label=\"LTC output\",linestyle=\"dashed\")\n",
    "# plt.ylim((-1, 1))|\n",
    "plt.legend(loc=\"upper right\")\n",
    "plt.title(\"After training\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "7cd80c41-6aa3-4f59-b4a0-f3f667a0e1ea",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0625"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1-((80-5)/80)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "9b37ef3c-a2f7-4f4e-9449-36efd5f1af71",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "data_x.shape:  (1, 48, 2)\n",
      "data_y.shape:  (1, 48, 1)\n"
     ]
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "# import seaborn as sns\n",
    "\n",
    "N = 48 # Length of the time-series\n",
    "# Input feature is a sine and a cosine wave\n",
    "data_x = np.stack(\n",
    "    [np.sin(np.linspace(0, 3 * np.pi, N)), np.cos(np.linspace(0, 3 * np.pi, N))], axis=1\n",
    ")\n",
    "data_x = np.expand_dims(data_x, axis=0).astype(np.float32)  # Add batch dimension\n",
    "# Target output is a sine with double the frequency of the input signal\n",
    "data_y = np.sin(np.linspace(0, 6 * np.pi, N)).reshape([1, N, 1]).astype(np.float32)\n",
    "print(\"data_x.shape: \", str(data_x.shape))\n",
    "print(\"data_y.shape: \", str(data_y.shape))\n",
    "\n",
    "# Let's visualize the training data\n",
    "# sns.set()\n",
    "# plt.figure(figsize=(6, 4))\n",
    "# plt.plot(data_x[0, :, 0], label=\"Input feature 1\")\n",
    "# plt.plot(data_x[0, :, 1], label=\"Input feature 1\")\n",
    "# plt.plot(data_y[0, :, 0], label=\"Target output\")\n",
    "# plt.ylim((-1, 1))\n",
    "# plt.title(\"Training data\")\n",
    "# plt.legend(loc=\"upper right\")\n",
    "# plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "c5c8a68c-cffa-418c-9b34-6e5bc787beaa",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(48,)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "N = 48\n",
    "np.sin(np.linspace(0, 3 * np.pi, N)).shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "59782fd8-bc20-4f92-afbe-c6e11b4417ac",
   "metadata": {},
   "outputs": [],
   "source": [
    "data_x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "2bf2398a-fa83-4aa6-babb-51fb8e0093f4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1, 48, 1)"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "aed70d79-7271-422d-af58-289b48de133c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " ltc_1 (LTC)                 (None, None, 1)           350       \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 350\n",
      "Trainable params: 350\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "wiring = wirings.AutoNCP(8,1) # 8 neurons in total, 1 output (motor neuron)\n",
    "model = keras.models.Sequential(\n",
    "    [\n",
    "        keras.layers.InputLayer(input_shape=(None, 2)),\n",
    "        # here we could potentially add layers before and after the LTC network\n",
    "        LTC(wiring, return_sequences=True),\n",
    "    ]\n",
    ")\n",
    "model.compile(\n",
    "    optimizer=keras.optimizers.Adam(0.01), loss='mean_squared_error'\n",
    ")\n",
    "\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ffa431fc-281c-4a1c-bd1e-c830a2e926b7",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "45da86d9-252e-4242-8373-e5003e695762",
   "metadata": {},
   "outputs": [],
   "source": [
    "# sns.set_style(\"white\")\n",
    "# plt.figure(figsize=(6, 4))\n",
    "# legend_handles = wiring.draw_graph(draw_labels=True, neuron_colors={\"command\": \"tab:cyan\"})\n",
    "# plt.legend(handles=legend_handles, loc=\"upper center\", bbox_to_anchor=(1, 1))\n",
    "# sns.despine(left=True, bottom=True)\n",
    "# plt.tight_layout()\n",
    "# plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2b13810b-e7b0-4d9d-90f7-47014a5e7846",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Let's visualize how LTC initialy performs before the training\n",
    "# sns.set()\n",
    "# prediction = model(data_x).numpy()\n",
    "# plt.figure(figsize=(6, 4))\n",
    "# plt.plot(data_y[0, :, 0], label=\"Target output\")\n",
    "# plt.plot(prediction[0, :, 0], label=\"NCP output\")\n",
    "# plt.ylim((-1, 1))\n",
    "# plt.title(\"Before training\")\n",
    "# plt.legend(loc=\"upper right\")\n",
    "# plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4c356196-cdb4-4adf-944d-1c3fc0d8fd3d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Train the model for 400 epochs (= training steps)\n",
    "hist = model.fit(x=data_x, y=data_y, batch_size=1, epochs=100,verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "37a84fb2-75bd-4aa4-b606-a47941731e9f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Let's visualize the training loss\n",
    "sns.set()\n",
    "plt.figure(figsize=(6, 4))\n",
    "plt.plot(hist.history[\"loss\"], label=\"Training loss\")\n",
    "plt.legend(loc=\"upper right\")\n",
    "plt.xlabel(\"Training steps\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "8268fae8-d9cb-4056-b354-3d2dbe1f141d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# How does the trained model now fit to the sinusoidal function?\n",
    "prediction = model(data_x).numpy()\n",
    "plt.figure(figsize=(6, 4))\n",
    "plt.plot(data_y[0, :, 0], label=\"Target output\")\n",
    "plt.plot(prediction[0, :, 0], label=\"LTC output\",linestyle=\"dashed\")\n",
    "plt.ylim((-1, 1))\n",
    "plt.legend(loc=\"upper right\")\n",
    "plt.title(\"After training\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "65a0d96b-0bfe-444f-b6d9-b4ddb24a83aa",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.10.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}