nb_model_task3.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "gradient": {
     "editing": false
    }
   },
   "source": [
    "# Task 3 Models\n",
    "\n",
    "1. Train the PCA models\n",
    "\n",
    "        This results in pickled PCA models.\n",
    "        \n",
    "2. Train autoencoder models (as replacement for PCA)\n",
    "\n",
    "        This results in saved autoencoder models.\n",
    "        \n",
    "3. Train a k-means cluster that sits on top of PCA & autoencoder\n",
    "\n",
    "        This results in pickled k-means models."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "gradient": {
     "editing": false
    }
   },
   "source": [
    "## LOAD"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "task_name = 'task3_jenga_openface'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "model_path = './models/{0}/'.format(task_name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import warnings\n",
    "import pickle\n",
    "\n",
    "from tqdm import tqdm\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "save_path = './data/tmp_analysis/{0}'.format(task_name)\n",
    "save_train_path = save_path + '_train.csv'\n",
    "save_test_path = save_path + '_test.csv'\n",
    "save_valid_path = save_path + '_valid.csv'\n",
    "\n",
    "\n",
    "Train = pd.read_csv(save_train_path)\n",
    "Test = pd.read_csv(save_test_path)\n",
    "Valid = pd.read_csv(save_valid_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "X = np.asarray(Train.to_numpy()[:,3:]).astype('float32')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "del Train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "Xtest = np.asarray(Test.to_numpy()[:,3:]).astype('float32')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "del Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "Xvalid = np.asarray(Valid.to_numpy()[:,3:]).astype('float32')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "del Valid"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## PCA Fit\n",
    "I found 32 components to be roughly the inflection curve.\n",
    "I am also creating an 18  component variant just in case we want a direct comparison with action units"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "from sklearn.decomposition import PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "gradient": {}
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[4.62986871e+06 2.72925010e+06 1.54130969e+05 7.60114766e+04\n",
      " 4.60983120e+04 2.44411292e+04 2.20384672e+04 7.54350491e+03\n",
      " 3.89412041e+03 2.03379258e+03 1.64928420e+03 1.51437092e+03\n",
      " 1.28886960e+03 9.60897509e+02 7.84298305e+02 6.18583225e+02\n",
      " 4.42433013e+02 3.89113599e+02 3.61640861e+02 3.32080732e+02\n",
      " 2.67733152e+02 2.32968815e+02 2.16726499e+02 1.77224026e+02\n",
      " 1.62874182e+02 1.37203929e+02 1.19198117e+02 1.09041353e+02\n",
      " 1.05151065e+02 8.80905426e+01 8.07717768e+01 6.83803279e+01]\n"
     ]
    }
   ],
   "source": [
    "pca_n_32 = PCA(n_components=32)\n",
    "face_pca_n_32 = pca_n_32.fit(X)\n",
    "\n",
    "print(face_pca_n_32.explained_variance_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "pickle.dump(face_pca_n_32,open(model_path+'pca_n_32.p','wb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "gradient": {}
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[4.62987002e+06 2.72925075e+06 1.54131010e+05 7.60114766e+04\n",
      " 4.60983221e+04 2.44411580e+04 2.20384689e+04 7.54351677e+03\n",
      " 3.89411660e+03 2.03379301e+03 1.64928250e+03 1.51436976e+03\n",
      " 1.28886960e+03 9.60897721e+02 7.84297987e+02 6.18582854e+02\n",
      " 4.42432166e+02 3.89111401e+02]\n"
     ]
    }
   ],
   "source": [
    "pca_n_18 = PCA(n_components=18)\n",
    "face_pca_n_18 = pca_n_18.fit(X)\n",
    "\n",
    "print(face_pca_n_18.explained_variance_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "pickle.dump(face_pca_n_18,open(model_path+'pca_n_18.p','wb'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Autoencoder\n",
    "Right now just doing a single layer \n",
    "Perhaps we can try a two layer encoder / two layer decoder next?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "32 Wide encoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "import keras\n",
    "from keras import layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "gradient": {}
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(151068, 136)\n"
     ]
    }
   ],
   "source": [
    "print(X.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "input_dim = X.shape[1]\n",
    "encoding_dim = 32\n",
    "\n",
    "input_layer = keras.Input(shape=(input_dim,))\n",
    "encoded_layer = layers.Dense(encoding_dim, activation='relu')(input_layer)\n",
    "decoded_layer = layers.Dense(input_dim, activation='relu')(encoded_layer)\n",
    "\n",
    "autoencoder_32 = keras.Model(input_layer,decoded_layer)\n",
    "\n",
    "encoder_32 = keras.Model(input_layer, encoded_layer)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "autoencoder_32.compile(optimizer='adam',\n",
    "                    loss='binary_crossentropy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "gradient": {}
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_1 (InputLayer)         [(None, 136)]             0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 32)                4384      \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 136)               4488      \n",
      "=================================================================\n",
      "Total params: 8,872\n",
      "Trainable params: 8,872\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "autoencoder_32.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "gradient": {}
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/2\n",
      "2361/2361 [==============================] - 7s 3ms/step - loss: -4276.7771 - val_loss: -5852.8838\n",
      "Epoch 2/2\n",
      "2361/2361 [==============================] - 6s 2ms/step - loss: -5842.1122 - val_loss: -5849.9346\n"
     ]
    }
   ],
   "source": [
    "history = autoencoder_32.fit(X, X, epochs=2, batch_size=64, verbose=1, validation_data=(Xtest, Xtest))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "gradient": {}
   },
   "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"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "\n",
    "plt.plot([0]+history.history['loss'], label='train')\n",
    "plt.plot([0]+history.history['val_loss'], label='test')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "predictions = autoencoder_32.predict(Xvalid)\n",
    "# TODO I should probably do something with the validation set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "autoencoder_32.save(model_path + 'autoencoder_32.h5', save_format='h5')\n",
    "encoder_32.save(model_path + 'encoder_32.h5', save_format='h5')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "18 wide encoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "import keras\n",
    "from keras import layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "input_dim = X.shape[1]\n",
    "encoding_dim = 18\n",
    "\n",
    "input_layer = keras.Input(shape=(input_dim,))\n",
    "encoded_layer = layers.Dense(encoding_dim, activation='relu')(input_layer)\n",
    "decoded_layer = layers.Dense(input_dim, activation='relu')(encoded_layer)\n",
    "\n",
    "autoencoder_18 = keras.Model(input_layer,decoded_layer)\n",
    "\n",
    "encoder_18 = keras.Model(input_layer, encoded_layer)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "autoencoder_18.compile(optimizer='adam',\n",
    "                    loss='binary_crossentropy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "gradient": {}
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model_2\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_2 (InputLayer)         [(None, 136)]             0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 18)                2466      \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 136)               2584      \n",
      "=================================================================\n",
      "Total params: 5,050\n",
      "Trainable params: 5,050\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "autoencoder_18.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "gradient": {}
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/2\n",
      "2361/2361 [==============================] - 6s 2ms/step - loss: -4883.8581 - val_loss: -6609.8623\n",
      "Epoch 2/2\n",
      "2361/2361 [==============================] - 6s 2ms/step - loss: -6821.5070 - val_loss: -6981.7183\n"
     ]
    }
   ],
   "source": [
    "history = autoencoder_18.fit(X, X, epochs=2, batch_size=64, verbose=1, validation_data=(Xtest, Xtest))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "gradient": {}
   },
   "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"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "\n",
    "plt.plot([0]+history.history['loss'], label='train')\n",
    "plt.plot([0]+history.history['val_loss'], label='test')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "predictions = autoencoder_18.predict(Xvalid)\n",
    "# TODO I should probably do something with the validation set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "autoencoder_18.save(model_path + 'autoencoder_18.h5', save_format='h5')\n",
    "encoder_18.save(model_path + 'encoder_18.h5', save_format='h5')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## K-means\n",
    "Not sure what this is going to do \n",
    "(my guess is it is going to cluster similar participants)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "from sklearn.cluster import KMeans"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "def visualize_labels(y):\n",
    "    idx = np.arange(len(y))\n",
    "\n",
    "    fig, (ax1,ax2) = plt.subplots(nrows=2, sharex=True)\n",
    "\n",
    "    ax1.imshow(y[np.newaxis,:], cmap=\"plasma\", aspect=\"auto\")\n",
    "    ax1.set_yticks([])\n",
    "\n",
    "    ax2.scatter(idx,y)\n",
    "\n",
    "    plt.tight_layout()\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "X_pca = face_pca_n_32.transform(X)\n",
    "kmeans_5 = KMeans(n_clusters=5, random_state=0).fit(X_pca)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "gradient": {}
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "y = kmeans_5.predict(face_pca_n_32.transform(Xtest))\n",
    "visualize_labels(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "gradient": {}
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "y = kmeans_5.predict(face_pca_n_32.transform(Xvalid))\n",
    "visualize_labels(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "pickle.dump(kmeans_5,open(model_path+'kmeans_5_from_pca.p','wb'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "gradient": {}
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:No training configuration found in the save file, so the model was *not* compiled. Compile it manually.\n"
     ]
    }
   ],
   "source": [
    "encoder_32 = keras.models.load_model(model_path + 'encoder_32.h5')\n",
    "\n",
    "X_auto = encoder_32.predict(X)\n",
    "kmeans_5 = KMeans(n_clusters=5, random_state=0).fit(X_auto)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "gradient": {}
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "y = kmeans_5.predict(encoder_32.predict(Xtest))\n",
    "visualize_labels(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "gradient": {}
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "y = kmeans_5.predict(encoder_32.predict(Xvalid))\n",
    "visualize_labels(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "gradient": {}
   },
   "outputs": [],
   "source": [
    "pickle.dump(kmeans_5,open(model_path+'kmeans_5_from_auto.p','wb'))"
   ]
  }
 ],
 "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": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}