Datasets
Models
Downloads: 1
Diff of
/MultiscaleNet.ipynb
[000000]
..
[a8e9b4]
Switch to unified view
| a | b/MultiscaleNet.ipynb | ||
|---|---|---|---|
| 1 | { |
||
| 2 | "cells": [ |
||
| 3 | { |
||
| 4 | "cell_type": "code", |
||
| 5 | "execution_count": 1, |
||
| 6 | "metadata": { |
||
| 7 | "collapsed": true |
||
| 8 | }, |
||
| 9 | "outputs": [], |
||
| 10 | "source": [ |
||
| 11 | "%matplotlib inline\n", |
||
| 12 | "import pandas as pd\n", |
||
| 13 | "import numpy as np\n", |
||
| 14 | "from cdtw import pydtw\n", |
||
| 15 | "import seaborn as sns\n", |
||
| 16 | "from tqdm import tqdm\n", |
||
| 17 | "import os\n", |
||
| 18 | "import json\n", |
||
| 19 | "import h5py" |
||
| 20 | ] |
||
| 21 | }, |
||
| 22 | { |
||
| 23 | "cell_type": "code", |
||
| 24 | "execution_count": 2, |
||
| 25 | "metadata": { |
||
| 26 | "collapsed": false |
||
| 27 | }, |
||
| 28 | "outputs": [ |
||
| 29 | { |
||
| 30 | "name": "stderr", |
||
| 31 | "output_type": "stream", |
||
| 32 | "text": [ |
||
| 33 | "Using TensorFlow backend.\n" |
||
| 34 | ] |
||
| 35 | } |
||
| 36 | ], |
||
| 37 | "source": [ |
||
| 38 | "from keras import backend as K\n", |
||
| 39 | "from keras.regularizers import l2, activity_l2\n", |
||
| 40 | "\n", |
||
| 41 | "from keras import backend as K\n", |
||
| 42 | "from keras.engine.topology import Layer\n", |
||
| 43 | "from keras.optimizers import RMSprop, SGD, Adam\n", |
||
| 44 | "from keras.layers.core import Dense, Dropout, Activation, Flatten, Lambda, Merge\n", |
||
| 45 | "from keras.layers.recurrent import LSTM, GRU\n", |
||
| 46 | "from keras.models import Sequential, Model, load_model\n", |
||
| 47 | "from keras.layers import Input, Bidirectional, merge\n", |
||
| 48 | "from keras.layers.convolutional import Convolution1D, AtrousConvolution1D\n", |
||
| 49 | "from keras.layers.pooling import MaxPooling1D, AveragePooling1D, GlobalMaxPooling1D" |
||
| 50 | ] |
||
| 51 | }, |
||
| 52 | { |
||
| 53 | "cell_type": "code", |
||
| 54 | "execution_count": 3, |
||
| 55 | "metadata": { |
||
| 56 | "collapsed": true |
||
| 57 | }, |
||
| 58 | "outputs": [], |
||
| 59 | "source": [ |
||
| 60 | "def read_train(fname):\n", |
||
| 61 | " subjects = {}\n", |
||
| 62 | " with h5py.File(fname, \"r\") as data_file:\n", |
||
| 63 | " for subject, subject_data in data_file.items():\n", |
||
| 64 | " print(subject)\n", |
||
| 65 | " X = subject_data[\"data\"][:]\n", |
||
| 66 | " y = subject_data[\"labels\"][:]\n", |
||
| 67 | " subjects[subject] = (X, y)\n", |
||
| 68 | " return subjects\n", |
||
| 69 | "\n", |
||
| 70 | "def read_test(fname):\n", |
||
| 71 | " subjects = {}\n", |
||
| 72 | " with h5py.File(fname, \"r\") as data_file:\n", |
||
| 73 | " X = {}\n", |
||
| 74 | " for subject, subject_data in data_file.items():\n", |
||
| 75 | " X[subject] = {}\n", |
||
| 76 | " for chunk_id, chunk in data_file[subject].items():\n", |
||
| 77 | " X[subject][chunk_id] = chunk[:]\n", |
||
| 78 | "\n", |
||
| 79 | " return X\n", |
||
| 80 | "\n", |
||
| 81 | "def batch(ts, y, n=1):\n", |
||
| 82 | " l = len(ts)\n", |
||
| 83 | " for ndx in range(0, l-n, 1):\n", |
||
| 84 | " yield (ts[ndx:min(ndx + n, l)], y[ndx:min(ndx + n, l)])\n", |
||
| 85 | "\n", |
||
| 86 | "def label_batch(batch):\n", |
||
| 87 | " if all([i == 1 for i in batch[1]]):\n", |
||
| 88 | " return 1\n", |
||
| 89 | " elif all([i == 0 for i in batch[1]]):\n", |
||
| 90 | " return 0\n", |
||
| 91 | " elif all([i == 2 for i in batch[1]]):\n", |
||
| 92 | " return 2\n", |
||
| 93 | " return -1\n", |
||
| 94 | " \n", |
||
| 95 | "def get_data():\n", |
||
| 96 | " train = read_train(\"train.h5\")\n", |
||
| 97 | " test = read_test(\"test.h5\")\n", |
||
| 98 | " \n", |
||
| 99 | "\n", |
||
| 100 | " subject_datas = {}\n", |
||
| 101 | " for subject, data in tqdm(train.items()):\n", |
||
| 102 | " subject_ts = data[0].T\n", |
||
| 103 | " subject_y = data[1][0]\n", |
||
| 104 | " batches = [i for i in batch(subject_ts, subject_y, n=1125)]\n", |
||
| 105 | " batches = [(i[0], label_batch(i)) for i in batches]\n", |
||
| 106 | " batches = [i for i in batches if i[1] != -1]\n", |
||
| 107 | " batches = [i for i in batches if len(i[0]) == 1125]\n", |
||
| 108 | " subject_datas[subject] = batches\n", |
||
| 109 | " \n", |
||
| 110 | " X = []\n", |
||
| 111 | " y = []\n", |
||
| 112 | " for subj, subj_data in tqdm(subject_datas.items()):\n", |
||
| 113 | " X.extend([i[0] for i in subj_data])\n", |
||
| 114 | " y.extend([i[1] for i in subj_data])\n", |
||
| 115 | " return X, y, test" |
||
| 116 | ] |
||
| 117 | }, |
||
| 118 | { |
||
| 119 | "cell_type": "code", |
||
| 120 | "execution_count": 4, |
||
| 121 | "metadata": { |
||
| 122 | "collapsed": false |
||
| 123 | }, |
||
| 124 | "outputs": [ |
||
| 125 | { |
||
| 126 | "name": "stdout", |
||
| 127 | "output_type": "stream", |
||
| 128 | "text": [ |
||
| 129 | "subject_0\n", |
||
| 130 | "subject_1\n", |
||
| 131 | "subject_2\n", |
||
| 132 | "subject_3\n" |
||
| 133 | ] |
||
| 134 | }, |
||
| 135 | { |
||
| 136 | "name": "stderr", |
||
| 137 | "output_type": "stream", |
||
| 138 | "text": [ |
||
| 139 | "100%|██████████| 4/4 [01:14<00:00, 19.86s/it]\n", |
||
| 140 | "100%|██████████| 4/4 [00:00<00:00, 145.58it/s]\n" |
||
| 141 | ] |
||
| 142 | } |
||
| 143 | ], |
||
| 144 | "source": [ |
||
| 145 | "X, y, test = get_data()" |
||
| 146 | ] |
||
| 147 | }, |
||
| 148 | { |
||
| 149 | "cell_type": "code", |
||
| 150 | "execution_count": 5, |
||
| 151 | "metadata": { |
||
| 152 | "collapsed": false |
||
| 153 | }, |
||
| 154 | "outputs": [], |
||
| 155 | "source": [ |
||
| 156 | "X = np.array(X)\n", |
||
| 157 | "y = np.array(y)" |
||
| 158 | ] |
||
| 159 | }, |
||
| 160 | { |
||
| 161 | "cell_type": "code", |
||
| 162 | "execution_count": 6, |
||
| 163 | "metadata": { |
||
| 164 | "collapsed": true |
||
| 165 | }, |
||
| 166 | "outputs": [], |
||
| 167 | "source": [ |
||
| 168 | "def shuffle_in_unison_scary(a, b):\n", |
||
| 169 | " rng_state = np.random.get_state()\n", |
||
| 170 | " np.random.shuffle(a)\n", |
||
| 171 | " np.random.set_state(rng_state)\n", |
||
| 172 | " np.random.shuffle(b)" |
||
| 173 | ] |
||
| 174 | }, |
||
| 175 | { |
||
| 176 | "cell_type": "code", |
||
| 177 | "execution_count": 7, |
||
| 178 | "metadata": { |
||
| 179 | "collapsed": true |
||
| 180 | }, |
||
| 181 | "outputs": [], |
||
| 182 | "source": [ |
||
| 183 | "shuffle_in_unison_scary(X, y)" |
||
| 184 | ] |
||
| 185 | }, |
||
| 186 | { |
||
| 187 | "cell_type": "code", |
||
| 188 | "execution_count": 8, |
||
| 189 | "metadata": { |
||
| 190 | "collapsed": false |
||
| 191 | }, |
||
| 192 | "outputs": [], |
||
| 193 | "source": [ |
||
| 194 | "validation_start = len(X) - 30\n", |
||
| 195 | "X_train = X[:validation_start]\n", |
||
| 196 | "y_train = y[:validation_start]\n", |
||
| 197 | "X_val = X[validation_start:]\n", |
||
| 198 | "y_val = y[validation_start:]" |
||
| 199 | ] |
||
| 200 | }, |
||
| 201 | { |
||
| 202 | "cell_type": "code", |
||
| 203 | "execution_count": 9, |
||
| 204 | "metadata": { |
||
| 205 | "collapsed": true |
||
| 206 | }, |
||
| 207 | "outputs": [], |
||
| 208 | "source": [ |
||
| 209 | "from scipy.signal import resample\n", |
||
| 210 | "\n", |
||
| 211 | "\n", |
||
| 212 | "def toarr(label):\n", |
||
| 213 | " arr = np.zeros(3)\n", |
||
| 214 | " arr[label] = 1\n", |
||
| 215 | " return arr\n", |
||
| 216 | "\n", |
||
| 217 | "def data_generator(X, Y, batch_size):\n", |
||
| 218 | " while True:\n", |
||
| 219 | " inds = np.random.choice(len(X), batch_size)\n", |
||
| 220 | " x = X[inds]\n", |
||
| 221 | " y = Y[inds]\n", |
||
| 222 | " y = np.vstack([toarr(i) for i in y])\n", |
||
| 223 | " x_256 = np.array([resample(i, 256) for i in x])\n", |
||
| 224 | " x_500 = np.array([resample(i, 500) for i in x])\n", |
||
| 225 | " x = np.array([i for i in x])\n", |
||
| 226 | " yield ([x_256, x_500, x], y)" |
||
| 227 | ] |
||
| 228 | }, |
||
| 229 | { |
||
| 230 | "cell_type": "code", |
||
| 231 | "execution_count": 10, |
||
| 232 | "metadata": { |
||
| 233 | "collapsed": true |
||
| 234 | }, |
||
| 235 | "outputs": [], |
||
| 236 | "source": [ |
||
| 237 | "def multiscale(chunk):\n", |
||
| 238 | " resampled_256 = resample(chunk, 256)\n", |
||
| 239 | " resampled_500 = resample(chunk, 500)\n", |
||
| 240 | " return [resampled_256, resampled_500, chunk]" |
||
| 241 | ] |
||
| 242 | }, |
||
| 243 | { |
||
| 244 | "cell_type": "code", |
||
| 245 | "execution_count": null, |
||
| 246 | "metadata": { |
||
| 247 | "collapsed": true |
||
| 248 | }, |
||
| 249 | "outputs": [], |
||
| 250 | "source": [ |
||
| 251 | "def get_base_model(input_len, fsize):\n", |
||
| 252 | " '''Base network to be shared (eq. to feature extraction).\n", |
||
| 253 | " '''\n", |
||
| 254 | " with K.tf.device('/gpu:1'):\n", |
||
| 255 | " input_seq = Input(shape=(input_len, 24))\n", |
||
| 256 | " nb_filters = 150\n", |
||
| 257 | " convolved = Convolution1D(nb_filters, fsize, border_mode=\"same\", activation=\"tanh\")(input_seq)\n", |
||
| 258 | " processed = GlobalMaxPooling1D()(convolved)\n", |
||
| 259 | " compressed = Dense(150, activation=\"tanh\")(processed)\n", |
||
| 260 | " compressed = Dropout(0.3)(compressed)\n", |
||
| 261 | " compressed = Dense(150, activation=\"tanh\")(compressed)\n", |
||
| 262 | " model = Model(input=input_seq, output=compressed) \n", |
||
| 263 | " return model" |
||
| 264 | ] |
||
| 265 | }, |
||
| 266 | { |
||
| 267 | "cell_type": "code", |
||
| 268 | "execution_count": null, |
||
| 269 | "metadata": { |
||
| 270 | "collapsed": false |
||
| 271 | }, |
||
| 272 | "outputs": [], |
||
| 273 | "source": [ |
||
| 274 | "with K.tf.device('/gpu:1'):\n", |
||
| 275 | " \n", |
||
| 276 | " input256_seq = Input(shape=(256, 24))\n", |
||
| 277 | " input500_seq = Input(shape=(500, 24))\n", |
||
| 278 | " input1125_seq = Input(shape=(1125, 24))\n", |
||
| 279 | " \n", |
||
| 280 | " base_network256 = get_base_model(256, 4)\n", |
||
| 281 | " base_network500 = get_base_model(500, 7)\n", |
||
| 282 | " base_network1125 = get_base_model(1125, 10)\n", |
||
| 283 | " \n", |
||
| 284 | " embedding_256 = base_network256(input256_seq)\n", |
||
| 285 | " embedding_500 = base_network500(input500_seq)\n", |
||
| 286 | " embedding_1125 = base_network256(input1125_seq)\n", |
||
| 287 | " \n", |
||
| 288 | " merged = merge([embedding_256, embedding_500, embedding_1125], mode=\"concat\")\n", |
||
| 289 | " out = Dense(3, activation='softmax')(merged)\n", |
||
| 290 | " \n", |
||
| 291 | " model = Model(input=[input256_seq, input500_seq, input1125_seq], output=out)\n", |
||
| 292 | " \n", |
||
| 293 | " #opt = SGD(lr=0.001, momentum=0.9, nesterov=True, clipvalue=0.0001)\n", |
||
| 294 | " opt = RMSprop(lr=0.005, clipvalue=10**6)\n", |
||
| 295 | " #opt = Adam(lr=0.001)\n", |
||
| 296 | " model.compile(loss=\"categorical_crossentropy\", optimizer=opt)" |
||
| 297 | ] |
||
| 298 | }, |
||
| 299 | { |
||
| 300 | "cell_type": "code", |
||
| 301 | "execution_count": 13, |
||
| 302 | "metadata": { |
||
| 303 | "collapsed": true |
||
| 304 | }, |
||
| 305 | "outputs": [], |
||
| 306 | "source": [ |
||
| 307 | "with K.tf.device('/gpu:2'):\n", |
||
| 308 | " model = load_model(\"convnet-multiscale-true-022unk\")" |
||
| 309 | ] |
||
| 310 | }, |
||
| 311 | { |
||
| 312 | "cell_type": "code", |
||
| 313 | "execution_count": 21, |
||
| 314 | "metadata": { |
||
| 315 | "collapsed": false |
||
| 316 | }, |
||
| 317 | "outputs": [ |
||
| 318 | { |
||
| 319 | "name": "stdout", |
||
| 320 | "output_type": "stream", |
||
| 321 | "text": [ |
||
| 322 | "Epoch 1/100000\n", |
||
| 323 | "100000/100000 [==============================] - 135s - loss: 0.1939 \n", |
||
| 324 | "Epoch 2/100000\n", |
||
| 325 | " 99500/100000 [============================>.] - ETA: 0s - loss: 0.1922" |
||
| 326 | ] |
||
| 327 | }, |
||
| 328 | { |
||
| 329 | "ename": "KeyboardInterrupt", |
||
| 330 | "evalue": "", |
||
| 331 | "output_type": "error", |
||
| 332 | "traceback": [ |
||
| 333 | "\u001b[0;31m\u001b[0m", |
||
| 334 | "\u001b[0;31mKeyboardInterrupt\u001b[0mTraceback (most recent call last)", |
||
| 335 | "\u001b[0;32m<ipython-input-21-09ad3c297605>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m 6\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0mK\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdevice\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'/gpu:2'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 7\u001b[0m model.fit_generator(data_generator(X_train, y_train, batch_size=50), samples_per_epoch, nb_epoch, \n\u001b[0;32m----> 8\u001b[0;31m callbacks=[earlyStopping], verbose=1)#, nb_val_samples=20000,\n\u001b[0m\u001b[1;32m 9\u001b[0m \u001b[0;31m#validation_data=data_generator(X_val, y_val, batch_size=40))\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", |
||
| 336 | "\u001b[0;32m/usr/local/lib/python2.7/dist-packages/keras/engine/training.pyc\u001b[0m in \u001b[0;36mfit_generator\u001b[0;34m(self, generator, samples_per_epoch, nb_epoch, verbose, callbacks, validation_data, nb_val_samples, class_weight, max_q_size, nb_worker, pickle_safe)\u001b[0m\n\u001b[1;32m 1451\u001b[0m \u001b[0mbatch_logs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0ml\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mo\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1452\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1453\u001b[0;31m \u001b[0mcallbacks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mon_batch_end\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbatch_index\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch_logs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1454\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1455\u001b[0m \u001b[0;31m# construct epoch logs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", |
||
| 337 | "\u001b[0;32m/usr/local/lib/python2.7/dist-packages/keras/callbacks.pyc\u001b[0m in \u001b[0;36mon_batch_end\u001b[0;34m(self, batch, logs)\u001b[0m\n\u001b[1;32m 59\u001b[0m \u001b[0mt_before_callbacks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 60\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mcallback\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcallbacks\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 61\u001b[0;31m \u001b[0mcallback\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mon_batch_end\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbatch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlogs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 62\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_delta_ts_batch_end\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0mt_before_callbacks\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 63\u001b[0m \u001b[0mdelta_t_median\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmedian\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_delta_ts_batch_end\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", |
||
| 338 | "\u001b[0;32m/usr/local/lib/python2.7/dist-packages/keras/callbacks.pyc\u001b[0m in \u001b[0;36mon_batch_end\u001b[0;34m(self, batch, logs)\u001b[0m\n\u001b[1;32m 187\u001b[0m \u001b[0;31m# will be handled by on_epoch_end\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 188\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mverbose\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mseen\u001b[0m \u001b[0;34m<\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mparams\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'nb_sample'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 189\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprogbar\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mupdate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mseen\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlog_values\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 190\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 191\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mon_epoch_end\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mepoch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlogs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m{\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", |
||
| 339 | "\u001b[0;32m/usr/local/lib/python2.7/dist-packages/keras/utils/generic_utils.pyc\u001b[0m in \u001b[0;36mupdate\u001b[0;34m(self, current, values, force)\u001b[0m\n\u001b[1;32m 157\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 158\u001b[0m \u001b[0msys\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstdout\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwrite\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minfo\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 159\u001b[0;31m \u001b[0msys\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstdout\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mflush\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 160\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 161\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mcurrent\u001b[0m \u001b[0;34m>=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtarget\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", |
||
| 340 | "\u001b[0;32m/usr/local/lib/python2.7/dist-packages/ipykernel/iostream.pyc\u001b[0m in \u001b[0;36mflush\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 267\u001b[0m \u001b[0mevt\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mthreading\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mEvent\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 268\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_io_loop\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0madd_callback\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mevt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mset\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 269\u001b[0;31m \u001b[0mevt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwait\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 270\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 271\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_flush\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", |
||
| 341 | "\u001b[0;32m/usr/lib/python2.7/threading.pyc\u001b[0m in \u001b[0;36mwait\u001b[0;34m(self, timeout)\u001b[0m\n\u001b[1;32m 618\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 619\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__flag\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 620\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__cond\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwait\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtimeout\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 621\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__flag\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 622\u001b[0m \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", |
||
| 342 | "\u001b[0;32m/usr/lib/python2.7/threading.pyc\u001b[0m in \u001b[0;36mwait\u001b[0;34m(self, timeout)\u001b[0m\n\u001b[1;32m 337\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;31m# restore state no matter what (e.g., KeyboardInterrupt)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 338\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mtimeout\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 339\u001b[0;31m \u001b[0mwaiter\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0macquire\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 340\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0m__debug__\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 341\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_note\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"%s.wait(): got it\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", |
||
| 343 | "\u001b[0;31mKeyboardInterrupt\u001b[0m: " |
||
| 344 | ] |
||
| 345 | } |
||
| 346 | ], |
||
| 347 | "source": [ |
||
| 348 | "from keras.callbacks import EarlyStopping\n", |
||
| 349 | "nb_epoch = 100000\n", |
||
| 350 | "earlyStopping = EarlyStopping(monitor='val_loss', patience=10, verbose=0, mode='auto')\n", |
||
| 351 | "samples_per_epoch = 100000\n", |
||
| 352 | "\n", |
||
| 353 | "with K.tf.device('/gpu:2'):\n", |
||
| 354 | " model.fit_generator(data_generator(X_train, y_train, batch_size=50), samples_per_epoch, nb_epoch, \n", |
||
| 355 | " callbacks=[earlyStopping], verbose=1)#, nb_val_samples=20000,\n", |
||
| 356 | " #validation_data=data_generator(X_val, y_val, batch_size=40))" |
||
| 357 | ] |
||
| 358 | }, |
||
| 359 | { |
||
| 360 | "cell_type": "code", |
||
| 361 | "execution_count": 26, |
||
| 362 | "metadata": { |
||
| 363 | "collapsed": false |
||
| 364 | }, |
||
| 365 | "outputs": [ |
||
| 366 | { |
||
| 367 | "name": "stdout", |
||
| 368 | "output_type": "stream", |
||
| 369 | "text": [ |
||
| 370 | "Epoch 1/1\n", |
||
| 371 | "30000/30000 [==============================] - 70s - loss: 0.1640 " |
||
| 372 | ] |
||
| 373 | }, |
||
| 374 | { |
||
| 375 | "name": "stderr", |
||
| 376 | "output_type": "stream", |
||
| 377 | "text": [ |
||
| 378 | " 24%|██▍ | 12/49 [00:00<00:00, 119.99it/s]" |
||
| 379 | ] |
||
| 380 | }, |
||
| 381 | { |
||
| 382 | "name": "stdout", |
||
| 383 | "output_type": "stream", |
||
| 384 | "text": [ |
||
| 385 | "\n" |
||
| 386 | ] |
||
| 387 | }, |
||
| 388 | { |
||
| 389 | "name": "stderr", |
||
| 390 | "output_type": "stream", |
||
| 391 | "text": [ |
||
| 392 | "100%|██████████| 49/49 [00:00<00:00, 135.88it/s]\n", |
||
| 393 | "100%|██████████| 89/89 [00:00<00:00, 215.36it/s]\n", |
||
| 394 | "100%|██████████| 49/49 [00:00<00:00, 248.00it/s]\n", |
||
| 395 | "100%|██████████| 89/89 [00:00<00:00, 219.17it/s]\n" |
||
| 396 | ] |
||
| 397 | }, |
||
| 398 | { |
||
| 399 | "name": "stdout", |
||
| 400 | "output_type": "stream", |
||
| 401 | "text": [ |
||
| 402 | "Epoch 1/1\n", |
||
| 403 | "30000/30000 [==============================] - 67s - loss: 0.1592 " |
||
| 404 | ] |
||
| 405 | }, |
||
| 406 | { |
||
| 407 | "name": "stderr", |
||
| 408 | "output_type": "stream", |
||
| 409 | "text": [ |
||
| 410 | "\r", |
||
| 411 | " 0%| | 0/49 [00:00<?, ?it/s]" |
||
| 412 | ] |
||
| 413 | }, |
||
| 414 | { |
||
| 415 | "name": "stdout", |
||
| 416 | "output_type": "stream", |
||
| 417 | "text": [ |
||
| 418 | "\n" |
||
| 419 | ] |
||
| 420 | }, |
||
| 421 | { |
||
| 422 | "name": "stderr", |
||
| 423 | "output_type": "stream", |
||
| 424 | "text": [ |
||
| 425 | "100%|██████████| 49/49 [00:00<00:00, 126.00it/s]\n", |
||
| 426 | "100%|██████████| 89/89 [00:00<00:00, 190.92it/s]\n", |
||
| 427 | "100%|██████████| 49/49 [00:00<00:00, 161.84it/s]\n", |
||
| 428 | "100%|██████████| 89/89 [00:00<00:00, 190.56it/s]\n" |
||
| 429 | ] |
||
| 430 | }, |
||
| 431 | { |
||
| 432 | "name": "stdout", |
||
| 433 | "output_type": "stream", |
||
| 434 | "text": [ |
||
| 435 | "Epoch 1/1\n", |
||
| 436 | "30000/30000 [==============================] - 60s - loss: 0.1557 " |
||
| 437 | ] |
||
| 438 | }, |
||
| 439 | { |
||
| 440 | "name": "stderr", |
||
| 441 | "output_type": "stream", |
||
| 442 | "text": [ |
||
| 443 | "\r", |
||
| 444 | " 0%| | 0/49 [00:00<?, ?it/s]" |
||
| 445 | ] |
||
| 446 | }, |
||
| 447 | { |
||
| 448 | "name": "stdout", |
||
| 449 | "output_type": "stream", |
||
| 450 | "text": [ |
||
| 451 | "\n" |
||
| 452 | ] |
||
| 453 | }, |
||
| 454 | { |
||
| 455 | "name": "stderr", |
||
| 456 | "output_type": "stream", |
||
| 457 | "text": [ |
||
| 458 | "100%|██████████| 49/49 [00:00<00:00, 175.59it/s]\n", |
||
| 459 | "100%|██████████| 89/89 [00:00<00:00, 188.29it/s]\n", |
||
| 460 | "100%|██████████| 49/49 [00:00<00:00, 170.44it/s]\n", |
||
| 461 | "100%|██████████| 89/89 [00:00<00:00, 202.76it/s]\n" |
||
| 462 | ] |
||
| 463 | }, |
||
| 464 | { |
||
| 465 | "name": "stdout", |
||
| 466 | "output_type": "stream", |
||
| 467 | "text": [ |
||
| 468 | "Epoch 1/1\n", |
||
| 469 | "30000/30000 [==============================] - 73s - loss: 0.1545 " |
||
| 470 | ] |
||
| 471 | }, |
||
| 472 | { |
||
| 473 | "name": "stderr", |
||
| 474 | "output_type": "stream", |
||
| 475 | "text": [ |
||
| 476 | " 29%|██▊ | 14/49 [00:00<00:00, 139.19it/s]" |
||
| 477 | ] |
||
| 478 | }, |
||
| 479 | { |
||
| 480 | "name": "stdout", |
||
| 481 | "output_type": "stream", |
||
| 482 | "text": [ |
||
| 483 | "\n" |
||
| 484 | ] |
||
| 485 | }, |
||
| 486 | { |
||
| 487 | "name": "stderr", |
||
| 488 | "output_type": "stream", |
||
| 489 | "text": [ |
||
| 490 | "100%|██████████| 49/49 [00:00<00:00, 190.23it/s]\n", |
||
| 491 | "100%|██████████| 89/89 [00:00<00:00, 235.47it/s]\n", |
||
| 492 | "100%|██████████| 49/49 [00:00<00:00, 221.90it/s]\n", |
||
| 493 | "100%|██████████| 89/89 [00:00<00:00, 171.74it/s]\n" |
||
| 494 | ] |
||
| 495 | }, |
||
| 496 | { |
||
| 497 | "name": "stdout", |
||
| 498 | "output_type": "stream", |
||
| 499 | "text": [ |
||
| 500 | "Epoch 1/1\n", |
||
| 501 | "30000/30000 [==============================] - 72s - loss: 0.1556 " |
||
| 502 | ] |
||
| 503 | }, |
||
| 504 | { |
||
| 505 | "name": "stderr", |
||
| 506 | "output_type": "stream", |
||
| 507 | "text": [ |
||
| 508 | "\r", |
||
| 509 | " 0%| | 0/49 [00:00<?, ?it/s]" |
||
| 510 | ] |
||
| 511 | }, |
||
| 512 | { |
||
| 513 | "name": "stdout", |
||
| 514 | "output_type": "stream", |
||
| 515 | "text": [ |
||
| 516 | "\n" |
||
| 517 | ] |
||
| 518 | }, |
||
| 519 | { |
||
| 520 | "name": "stderr", |
||
| 521 | "output_type": "stream", |
||
| 522 | "text": [ |
||
| 523 | "100%|██████████| 49/49 [00:00<00:00, 119.83it/s]\n", |
||
| 524 | "100%|██████████| 89/89 [00:00<00:00, 195.06it/s]\n", |
||
| 525 | "100%|██████████| 49/49 [00:00<00:00, 187.22it/s]\n", |
||
| 526 | "100%|██████████| 89/89 [00:00<00:00, 201.47it/s]\n" |
||
| 527 | ] |
||
| 528 | }, |
||
| 529 | { |
||
| 530 | "name": "stdout", |
||
| 531 | "output_type": "stream", |
||
| 532 | "text": [ |
||
| 533 | "Epoch 1/1\n", |
||
| 534 | "30000/30000 [==============================] - 68s - loss: 0.1563 " |
||
| 535 | ] |
||
| 536 | }, |
||
| 537 | { |
||
| 538 | "name": "stderr", |
||
| 539 | "output_type": "stream", |
||
| 540 | "text": [ |
||
| 541 | " 27%|██▋ | 13/49 [00:00<00:00, 119.17it/s]" |
||
| 542 | ] |
||
| 543 | }, |
||
| 544 | { |
||
| 545 | "name": "stdout", |
||
| 546 | "output_type": "stream", |
||
| 547 | "text": [ |
||
| 548 | "\n" |
||
| 549 | ] |
||
| 550 | }, |
||
| 551 | { |
||
| 552 | "name": "stderr", |
||
| 553 | "output_type": "stream", |
||
| 554 | "text": [ |
||
| 555 | "100%|██████████| 49/49 [00:00<00:00, 142.53it/s]\n", |
||
| 556 | "100%|██████████| 89/89 [00:00<00:00, 203.55it/s]\n", |
||
| 557 | "100%|██████████| 49/49 [00:00<00:00, 192.02it/s]\n", |
||
| 558 | "100%|██████████| 89/89 [00:00<00:00, 193.69it/s]\n" |
||
| 559 | ] |
||
| 560 | }, |
||
| 561 | { |
||
| 562 | "name": "stdout", |
||
| 563 | "output_type": "stream", |
||
| 564 | "text": [ |
||
| 565 | "Epoch 1/1\n", |
||
| 566 | "30000/30000 [==============================] - 72s - loss: 0.1588 " |
||
| 567 | ] |
||
| 568 | }, |
||
| 569 | { |
||
| 570 | "name": "stderr", |
||
| 571 | "output_type": "stream", |
||
| 572 | "text": [ |
||
| 573 | " 31%|███ | 15/49 [00:00<00:00, 145.08it/s]" |
||
| 574 | ] |
||
| 575 | }, |
||
| 576 | { |
||
| 577 | "name": "stdout", |
||
| 578 | "output_type": "stream", |
||
| 579 | "text": [ |
||
| 580 | "\n" |
||
| 581 | ] |
||
| 582 | }, |
||
| 583 | { |
||
| 584 | "name": "stderr", |
||
| 585 | "output_type": "stream", |
||
| 586 | "text": [ |
||
| 587 | "100%|██████████| 49/49 [00:00<00:00, 181.02it/s]\n", |
||
| 588 | "100%|██████████| 89/89 [00:00<00:00, 199.11it/s]\n", |
||
| 589 | "100%|██████████| 49/49 [00:00<00:00, 229.50it/s]\n", |
||
| 590 | "100%|██████████| 89/89 [00:00<00:00, 204.20it/s]\n" |
||
| 591 | ] |
||
| 592 | }, |
||
| 593 | { |
||
| 594 | "name": "stdout", |
||
| 595 | "output_type": "stream", |
||
| 596 | "text": [ |
||
| 597 | "Epoch 1/1\n", |
||
| 598 | "30000/30000 [==============================] - 67s - loss: 0.1598 " |
||
| 599 | ] |
||
| 600 | }, |
||
| 601 | { |
||
| 602 | "name": "stderr", |
||
| 603 | "output_type": "stream", |
||
| 604 | "text": [ |
||
| 605 | " 37%|███▋ | 18/49 [00:00<00:00, 174.26it/s]" |
||
| 606 | ] |
||
| 607 | }, |
||
| 608 | { |
||
| 609 | "name": "stdout", |
||
| 610 | "output_type": "stream", |
||
| 611 | "text": [ |
||
| 612 | "\n" |
||
| 613 | ] |
||
| 614 | }, |
||
| 615 | { |
||
| 616 | "name": "stderr", |
||
| 617 | "output_type": "stream", |
||
| 618 | "text": [ |
||
| 619 | "100%|██████████| 49/49 [00:00<00:00, 130.09it/s]\n", |
||
| 620 | "100%|██████████| 89/89 [00:00<00:00, 208.28it/s]\n", |
||
| 621 | "100%|██████████| 49/49 [00:00<00:00, 189.43it/s]\n", |
||
| 622 | "100%|██████████| 89/89 [00:00<00:00, 181.52it/s]\n" |
||
| 623 | ] |
||
| 624 | }, |
||
| 625 | { |
||
| 626 | "name": "stdout", |
||
| 627 | "output_type": "stream", |
||
| 628 | "text": [ |
||
| 629 | "Epoch 1/1\n", |
||
| 630 | "30000/30000 [==============================] - 53s - loss: 0.1634 " |
||
| 631 | ] |
||
| 632 | }, |
||
| 633 | { |
||
| 634 | "name": "stderr", |
||
| 635 | "output_type": "stream", |
||
| 636 | "text": [ |
||
| 637 | " 41%|████ | 20/49 [00:00<00:00, 191.91it/s]" |
||
| 638 | ] |
||
| 639 | }, |
||
| 640 | { |
||
| 641 | "name": "stdout", |
||
| 642 | "output_type": "stream", |
||
| 643 | "text": [ |
||
| 644 | "\n" |
||
| 645 | ] |
||
| 646 | }, |
||
| 647 | { |
||
| 648 | "name": "stderr", |
||
| 649 | "output_type": "stream", |
||
| 650 | "text": [ |
||
| 651 | "100%|██████████| 49/49 [00:00<00:00, 202.34it/s]\n", |
||
| 652 | "100%|██████████| 89/89 [00:00<00:00, 216.69it/s]\n", |
||
| 653 | "100%|██████████| 49/49 [00:00<00:00, 212.14it/s]\n", |
||
| 654 | "100%|██████████| 89/89 [00:00<00:00, 212.55it/s]\n" |
||
| 655 | ] |
||
| 656 | }, |
||
| 657 | { |
||
| 658 | "name": "stdout", |
||
| 659 | "output_type": "stream", |
||
| 660 | "text": [ |
||
| 661 | "Epoch 1/1\n", |
||
| 662 | "30000/30000 [==============================] - 70s - loss: 0.1491 " |
||
| 663 | ] |
||
| 664 | }, |
||
| 665 | { |
||
| 666 | "name": "stderr", |
||
| 667 | "output_type": "stream", |
||
| 668 | "text": [ |
||
| 669 | " 29%|██▊ | 14/49 [00:00<00:00, 135.21it/s]" |
||
| 670 | ] |
||
| 671 | }, |
||
| 672 | { |
||
| 673 | "name": "stdout", |
||
| 674 | "output_type": "stream", |
||
| 675 | "text": [ |
||
| 676 | "\n" |
||
| 677 | ] |
||
| 678 | }, |
||
| 679 | { |
||
| 680 | "name": "stderr", |
||
| 681 | "output_type": "stream", |
||
| 682 | "text": [ |
||
| 683 | "100%|██████████| 49/49 [00:00<00:00, 188.83it/s]\n", |
||
| 684 | "100%|██████████| 89/89 [00:00<00:00, 169.04it/s]\n", |
||
| 685 | "100%|██████████| 49/49 [00:00<00:00, 178.81it/s]\n", |
||
| 686 | "100%|██████████| 89/89 [00:00<00:00, 182.79it/s]\n" |
||
| 687 | ] |
||
| 688 | } |
||
| 689 | ], |
||
| 690 | "source": [ |
||
| 691 | "# BLEND NNS\n", |
||
| 692 | "for blend_id in range(25, 35):\n", |
||
| 693 | " with K.tf.device('/gpu:2'):\n", |
||
| 694 | " model.fit_generator(data_generator(X_train, y_train, batch_size=100), samples_per_epoch=30000, nb_epoch=1, \n", |
||
| 695 | " callbacks=[earlyStopping], verbose=1)\n", |
||
| 696 | " \n", |
||
| 697 | " df = []\n", |
||
| 698 | " for subj in test:\n", |
||
| 699 | " for chunk in tqdm(test[subj]):\n", |
||
| 700 | " data = {}\n", |
||
| 701 | " data[\"subject_id\"] = int(subj.split(\"_\")[-1])\n", |
||
| 702 | " data[\"chunk_id\"] = int(chunk.split(\"_\")[-1])\n", |
||
| 703 | " arr = test[subj][chunk].T\n", |
||
| 704 | " preds = model.predict([np.array([i]) for i in multiscale(arr)])[0]\n", |
||
| 705 | " data[\"class_0_score\"] = preds[0]\n", |
||
| 706 | " data[\"class_1_score\"] = preds[1]\n", |
||
| 707 | " data[\"class_2_score\"] = preds[2]\n", |
||
| 708 | " for i in range(0, 1125):\n", |
||
| 709 | " data[\"tick\"] = i\n", |
||
| 710 | " df.append(data.copy())\n", |
||
| 711 | " df = pd.DataFrame(df)\n", |
||
| 712 | " df = df[[\"subject_id\", \"chunk_id\", \"tick\", \"class_0_score\",\n", |
||
| 713 | " \"class_1_score\",\"class_2_score\"]]\n", |
||
| 714 | " \n", |
||
| 715 | " df.to_csv('submit_blended_' + str(blend_id) + '.csv', index=False)" |
||
| 716 | ] |
||
| 717 | }, |
||
| 718 | { |
||
| 719 | "cell_type": "code", |
||
| 720 | "execution_count": 15, |
||
| 721 | "metadata": { |
||
| 722 | "collapsed": true |
||
| 723 | }, |
||
| 724 | "outputs": [], |
||
| 725 | "source": [ |
||
| 726 | "model.save(\"convnet-multiscale-deep-021unk\")" |
||
| 727 | ] |
||
| 728 | }, |
||
| 729 | { |
||
| 730 | "cell_type": "code", |
||
| 731 | "execution_count": 19, |
||
| 732 | "metadata": { |
||
| 733 | "collapsed": false |
||
| 734 | }, |
||
| 735 | "outputs": [ |
||
| 736 | { |
||
| 737 | "name": "stderr", |
||
| 738 | "output_type": "stream", |
||
| 739 | "text": [ |
||
| 740 | "100%|██████████| 49/49 [00:00<00:00, 165.60it/s]\n", |
||
| 741 | "100%|██████████| 89/89 [00:00<00:00, 185.89it/s]\n", |
||
| 742 | "100%|██████████| 49/49 [00:00<00:00, 189.62it/s]\n", |
||
| 743 | "100%|██████████| 89/89 [00:00<00:00, 175.60it/s]\n" |
||
| 744 | ] |
||
| 745 | } |
||
| 746 | ], |
||
| 747 | "source": [ |
||
| 748 | "df = []\n", |
||
| 749 | "for subj in test:\n", |
||
| 750 | " for chunk in tqdm(test[subj]):\n", |
||
| 751 | " data = {}\n", |
||
| 752 | " data[\"subject_id\"] = int(subj.split(\"_\")[-1])\n", |
||
| 753 | " data[\"chunk_id\"] = int(chunk.split(\"_\")[-1])\n", |
||
| 754 | " arr = test[subj][chunk].T\n", |
||
| 755 | " preds = model.predict([np.array([i]) for i in multiscale(arr)])[0]\n", |
||
| 756 | " data[\"class_0_score\"] = preds[0]\n", |
||
| 757 | " data[\"class_1_score\"] = preds[1]\n", |
||
| 758 | " data[\"class_2_score\"] = preds[2]\n", |
||
| 759 | " for i in range(0, 1125):\n", |
||
| 760 | " data[\"tick\"] = i\n", |
||
| 761 | " df.append(data.copy())\n", |
||
| 762 | "df = pd.DataFrame(df)\n", |
||
| 763 | "df = df[[\"subject_id\", \"chunk_id\", \"tick\", \"class_0_score\",\n", |
||
| 764 | " \"class_1_score\",\"class_2_score\"]]" |
||
| 765 | ] |
||
| 766 | }, |
||
| 767 | { |
||
| 768 | "cell_type": "code", |
||
| 769 | "execution_count": 20, |
||
| 770 | "metadata": { |
||
| 771 | "collapsed": true |
||
| 772 | }, |
||
| 773 | "outputs": [], |
||
| 774 | "source": [ |
||
| 775 | "df.to_csv('submit_true_multiscale_016_large_batch.csv', index=False)" |
||
| 776 | ] |
||
| 777 | }, |
||
| 778 | { |
||
| 779 | "cell_type": "code", |
||
| 780 | "execution_count": null, |
||
| 781 | "metadata": { |
||
| 782 | "collapsed": true |
||
| 783 | }, |
||
| 784 | "outputs": [], |
||
| 785 | "source": [] |
||
| 786 | } |
||
| 787 | ], |
||
| 788 | "metadata": { |
||
| 789 | "kernelspec": { |
||
| 790 | "display_name": "Python 2", |
||
| 791 | "language": "python", |
||
| 792 | "name": "python2" |
||
| 793 | }, |
||
| 794 | "language_info": { |
||
| 795 | "codemirror_mode": { |
||
| 796 | "name": "ipython", |
||
| 797 | "version": 2 |
||
| 798 | }, |
||
| 799 | "file_extension": ".py", |
||
| 800 | "mimetype": "text/x-python", |
||
| 801 | "name": "python", |
||
| 802 | "nbconvert_exporter": "python", |
||
| 803 | "pygments_lexer": "ipython2", |
||
| 804 | "version": "2.7.6" |
||
| 805 | } |
||
| 806 | }, |
||
| 807 | "nbformat": 4, |
||
| 808 | "nbformat_minor": 1 |
||
| 809 | } |