Deep Fashion Network Re-train.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import PIL\n",
    "from keras.applications import VGG16,imagenet_utils\n",
    "from keras.preprocessing.image import img_to_array\n",
    "from keras.preprocessing.image import load_img\n",
    "from keras.models import Model\n",
    "from keras.layers import Dense\n",
    "import numpy as np\n",
    "import glob\n",
    "import os\n",
    "\n",
    "import pickle\n",
    "import time\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "df = pd.read_table('Data/DeepFashion/list_attr_img.txt',skiprows=1,sep='\\s+',header=None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "bbox = pd.read_table('Data/DeepFashion/list_bbox.txt',sep='\\s+')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "joined = bbox.join(df,lsuffix='image_name',rsuffix='0')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "attr = (df.iloc[:,6:].values>0)\n",
    "a_counts = (np.sum(attr,axis=0))\n",
    "top_a = np.flip(a_counts.argsort(),0)[:206]\n",
    "top_classes = np.where(np.sum(attr,axis=0) > 1000)[0]\n",
    "attr_subset = attr[:,top_a]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "subset = pd.DataFrame([], columns=joined.columns)\n",
    "for a in top_a[::-1]:\n",
    "    needed = 1000 - np.sum(subset.iloc[:,6+a]>0)\n",
    "    if needed > 0:\n",
    "        has_attr = joined[joined.iloc[:,6+a]>0]\n",
    "        sample = np.random.randint(has_attr.shape[0],size=needed)\n",
    "        subset = subset.append(has_attr.iloc[sample])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "pickle.dump(subset,open('../rebalanced_attr_subset.p','wb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "subset = pickle.load(open('../rebalanced_attr_subset.p','rb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "attr_subset = (subset.iloc[:,6+top_a].values>0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(85682, 1006)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "subset.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df = pickle.load(open('../df_subset/attr.p','rb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(25000, 1006)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "inputShape = (224, 224)\n",
    "preprocess = imagenet_utils.preprocess_input\n",
    "\n",
    "def preprocess_DF(bbox):\n",
    "    image = crop_resize_DF(bbox)\n",
    "    image = img_to_array(image)\n",
    "    image = np.expand_dims(image,axis=0)\n",
    "    \n",
    "    return preprocess(image)\n",
    "\n",
    "def crop_resize_DF(bbox):\n",
    "    img = PIL.Image.open('Data/DeepFashion/'+bbox['image_name'])\n",
    "    lx = bbox['x_1']\n",
    "    ly = bbox['y_1']\n",
    "    ux = bbox['x_2']\n",
    "    uy = bbox['y_2']\n",
    "    \n",
    "    img = img.crop((lx,ly,ux,uy))\n",
    "    img = img.resize(inputShape, PIL.Image.ANTIALIAS)\n",
    "    return img"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def generator(df, batch_size):\n",
    "    while True:\n",
    "        subset = np.random.randint(df.shape[0],size=batch_size)\n",
    "        proc_imgs = []\n",
    "        for i in subset:\n",
    "            proc_imgs.append(np.squeeze(preprocess_DF(df.iloc[i])))\n",
    "    \n",
    "        attr = (df.iloc[subset,6+top_a].values>0)\n",
    "        yield np.array(proc_imgs), attr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "joined.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "attr = attr_subset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "pos_weight = np.prod(attr.shape)/np.sum(attr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "221.41995860252149"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pos_weight"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "class_weights = 1/(np.mean(attr,axis=0)*1000+1e-8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "import keras.backend as K\n",
    "import tensorflow as tf\n",
    "\n",
    "def weighted_sigmoid_loss(y_true,y_pred):\n",
    "    return tf.nn.weighted_cross_entropy_with_logits(y_true,y_pred,pos_weight=pos_weight*5)\n",
    "\n",
    "def H_pred(y_true, y_pred, threshold=0.5):\n",
    "    p = tf.where(y_pred>threshold,tf.ones_like(y_pred),tf.zeros_like(y_pred))\n",
    "    return K.mean(K.sum(K.abs(y_true - p),axis=1))\n",
    "\n",
    "def total_labels(y_true, y_pred, threshold=0.5):\n",
    "    p = tf.where(y_pred>threshold,tf.ones_like(y_pred),tf.zeros_like(y_pred))\n",
    "    return K.mean(K.sum(p,axis=1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "t = np.array([[1,1,0,0,1,0],[1,1,0,0,1,0]])\n",
    "p = np.array([[0,1,0.6,0,1,0],[0,1,0,0,1,0]])\n",
    "K.eval(H_pred(t,p))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "t = [[0.7,0.2,0.8,0.3,0.2],[0.7,0.2,0.8,0.3,0.2]]\n",
    "np.flip(np.argsort(t),axis=0)[:3]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def global_precision(y_true, y_pred, k=5):\n",
    "    _,i = tf.nn.top_k(y_pred,k=k)\n",
    "    t = tf.sparse_to_dense(i,(-1,len(top_a)),1.0,0.0)\n",
    "    N_c = tf.reduce_sum(tf.cast(K.equal(y_true,t), tf.float32))\n",
    "    N_p = K.sum(t)\n",
    "    return N_c/N_p"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from keras import optimizers\n",
    "from keras import regularizers\n",
    "from keras.layers.normalization import BatchNormalization\n",
    "from keras.layers import Dropout,Flatten\n",
    "l = 0.0\n",
    "\n",
    "base_model = VGG16(weights='imagenet')\n",
    "base_model = VGG16(weights='imagenet', include_top=False, input_shape=(224, 224, 3))\n",
    "x = base_model.output\n",
    "\n",
    "x = Flatten(name='flatten')(x)\n",
    "x = BatchNormalization()(x)\n",
    "x = Dropout(rate=0.2)(x)\n",
    "\n",
    "x = Dense(4096, activation='relu', name='fc1', kernel_initializer='glorot_normal',\n",
    "         bias_initializer='glorot_uniform', kernel_regularizer=regularizers.l2(l))(x)\n",
    "x = BatchNormalization()(x)\n",
    "x = Dropout(rate=0.2)(x)\n",
    "\n",
    "predictions = Dense(len(top_a), activation='sigmoid',name='predictions', kernel_initializer='glorot_normal',\n",
    "         bias_initializer='glorot_uniform', kernel_regularizer=regularizers.l2(l))(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "new_model = Model(inputs=base_model.input, outputs=predictions)\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": true,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "new_model.compile(optimizer='adam', loss=weighted_sigmoid_loss, \n",
    "                  metrics=[H_pred,total_labels])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/80\n",
      "50/50 [==============================] - 182s - loss: 3.1585 - H_pred: 83.7841 - total_labels: 84.0619   \n",
      "Epoch 2/80\n",
      "50/50 [==============================] - 170s - loss: 3.0322 - H_pred: 79.6831 - total_labels: 80.1228   \n",
      "Epoch 3/80\n",
      "50/50 [==============================] - 170s - loss: 2.9069 - H_pred: 79.5200 - total_labels: 80.0047   \n",
      "Epoch 4/80\n",
      "50/50 [==============================] - 170s - loss: 2.9118 - H_pred: 77.3678 - total_labels: 77.9459   \n",
      "Epoch 5/80\n",
      "50/50 [==============================] - 170s - loss: 2.7942 - H_pred: 75.7281 - total_labels: 76.3575   \n",
      "Epoch 6/80\n",
      "50/50 [==============================] - 170s - loss: 2.7182 - H_pred: 73.7516 - total_labels: 74.4659   \n",
      "Epoch 7/80\n",
      "50/50 [==============================] - 169s - loss: 2.6055 - H_pred: 73.1884 - total_labels: 73.8566   \n",
      "Epoch 8/80\n",
      "50/50 [==============================] - 168s - loss: 2.6558 - H_pred: 71.5216 - total_labels: 72.2234   \n",
      "Epoch 9/80\n",
      "50/50 [==============================] - 168s - loss: 2.7346 - H_pred: 72.7081 - total_labels: 73.4269   \n",
      "Epoch 10/80\n",
      "50/50 [==============================] - 171s - loss: 2.6514 - H_pred: 72.5625 - total_labels: 73.2537   \n",
      "Epoch 11/80\n",
      "50/50 [==============================] - 169s - loss: 2.6327 - H_pred: 73.9559 - total_labels: 74.6884   \n",
      "Epoch 12/80\n",
      "50/50 [==============================] - 169s - loss: 2.6808 - H_pred: 75.5766 - total_labels: 76.3300   \n",
      "Epoch 13/80\n",
      "50/50 [==============================] - 169s - loss: 2.6389 - H_pred: 73.5987 - total_labels: 74.3078   \n",
      "Epoch 14/80\n",
      "50/50 [==============================] - 169s - loss: 2.7199 - H_pred: 71.1769 - total_labels: 71.9406   \n",
      "Epoch 15/80\n",
      "50/50 [==============================] - 168s - loss: 2.6380 - H_pred: 72.0388 - total_labels: 72.7913   \n",
      "Epoch 16/80\n",
      "50/50 [==============================] - 167s - loss: 2.6576 - H_pred: 72.4759 - total_labels: 73.2325   \n",
      "Epoch 17/80\n",
      "50/50 [==============================] - 168s - loss: 2.6341 - H_pred: 74.4400 - total_labels: 75.1972   \n",
      "Epoch 18/80\n",
      "50/50 [==============================] - 168s - loss: 2.6646 - H_pred: 74.7403 - total_labels: 75.5138   \n",
      "Epoch 19/80\n",
      "50/50 [==============================] - 169s - loss: 2.5720 - H_pred: 74.3144 - total_labels: 75.0597   \n",
      "Epoch 20/80\n",
      "50/50 [==============================] - 183s - loss: 2.6191 - H_pred: 73.1994 - total_labels: 73.9672   \n",
      "Epoch 21/80\n",
      "50/50 [==============================] - 166s - loss: 2.6306 - H_pred: 71.8388 - total_labels: 72.5841   \n",
      "Epoch 22/80\n",
      "50/50 [==============================] - 167s - loss: 2.6444 - H_pred: 72.0431 - total_labels: 72.8000   \n",
      "Epoch 23/80\n",
      "50/50 [==============================] - 167s - loss: 2.6216 - H_pred: 70.8263 - total_labels: 71.5831   \n",
      "Epoch 24/80\n",
      " 7/50 [===>..........................] - ETA: 144s - loss: 2.7272 - H_pred: 70.5022 - total_labels: 71.3125"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-22-3d53e0a7092f>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      6\u001b[0m h = new_model.fit_generator(generator(subset,64),steps_per_epoch=50,epochs=80,verbose=1,\n\u001b[0;32m----> 7\u001b[0;31m                             callbacks=[tbCB,checkpointCB],class_weight=class_weights)\n\u001b[0m",
      "\u001b[0;32m/usr/local/lib/python3.4/dist-packages/keras/legacy/interfaces.py\u001b[0m in \u001b[0;36mwrapper\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m     85\u001b[0m                 warnings.warn('Update your `' + object_name +\n\u001b[1;32m     86\u001b[0m                               '` call to the Keras 2 API: ' + signature, stacklevel=2)\n\u001b[0;32m---> 87\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     88\u001b[0m         \u001b[0mwrapper\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_original_function\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     89\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mwrapper\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.4/dist-packages/keras/engine/training.py\u001b[0m in \u001b[0;36mfit_generator\u001b[0;34m(self, generator, steps_per_epoch, epochs, verbose, callbacks, validation_data, validation_steps, class_weight, max_queue_size, workers, use_multiprocessing, initial_epoch)\u001b[0m\n\u001b[1;32m   1838\u001b[0m                     outs = self.train_on_batch(x, y,\n\u001b[1;32m   1839\u001b[0m                                                \u001b[0msample_weight\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0msample_weight\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1840\u001b[0;31m                                                class_weight=class_weight)\n\u001b[0m\u001b[1;32m   1841\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1842\u001b[0m                     \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mouts\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlist\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.4/dist-packages/keras/engine/training.py\u001b[0m in \u001b[0;36mtrain_on_batch\u001b[0;34m(self, x, y, sample_weight, class_weight)\u001b[0m\n\u001b[1;32m   1563\u001b[0m             \u001b[0mins\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mx\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0my\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0msample_weights\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1564\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_make_train_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1565\u001b[0;31m         \u001b[0moutputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mins\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1566\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1567\u001b[0m             \u001b[0;32mreturn\u001b[0m \u001b[0moutputs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.4/dist-packages/keras/backend/tensorflow_backend.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, inputs)\u001b[0m\n\u001b[1;32m   2266\u001b[0m         updated = session.run(self.outputs + [self.updates_op],\n\u001b[1;32m   2267\u001b[0m                               \u001b[0mfeed_dict\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mfeed_dict\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2268\u001b[0;31m                               **self.session_kwargs)\n\u001b[0m\u001b[1;32m   2269\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mupdated\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2270\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.4/dist-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36mrun\u001b[0;34m(self, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m    776\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    777\u001b[0m       result = self._run(None, fetches, feed_dict, options_ptr,\n\u001b[0;32m--> 778\u001b[0;31m                          run_metadata_ptr)\n\u001b[0m\u001b[1;32m    779\u001b[0m       \u001b[0;32mif\u001b[0m \u001b[0mrun_metadata\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    780\u001b[0m         \u001b[0mproto_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf_session\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTF_GetBuffer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrun_metadata_ptr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.4/dist-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_run\u001b[0;34m(self, handle, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m    980\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mfinal_fetches\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0mfinal_targets\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    981\u001b[0m       results = self._do_run(handle, final_targets, final_fetches,\n\u001b[0;32m--> 982\u001b[0;31m                              feed_dict_string, options, run_metadata)\n\u001b[0m\u001b[1;32m    983\u001b[0m     \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    984\u001b[0m       \u001b[0mresults\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",
      "\u001b[0;32m/usr/local/lib/python3.4/dist-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_do_run\u001b[0;34m(self, handle, target_list, fetch_list, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m   1030\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mhandle\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1031\u001b[0m       return self._do_call(_run_fn, self._session, feed_dict, fetch_list,\n\u001b[0;32m-> 1032\u001b[0;31m                            target_list, options, run_metadata)\n\u001b[0m\u001b[1;32m   1033\u001b[0m     \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1034\u001b[0m       return self._do_call(_prun_fn, self._session, handle, feed_dict,\n",
      "\u001b[0;32m/usr/local/lib/python3.4/dist-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_do_call\u001b[0;34m(self, fn, *args)\u001b[0m\n\u001b[1;32m   1037\u001b[0m   \u001b[0;32mdef\u001b[0m \u001b[0m_do_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1038\u001b[0m     \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1039\u001b[0;31m       \u001b[0;32mreturn\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1040\u001b[0m     \u001b[0;32mexcept\u001b[0m \u001b[0merrors\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mOpError\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1041\u001b[0m       \u001b[0mmessage\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcompat\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mas_text\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmessage\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.4/dist-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_run_fn\u001b[0;34m(session, feed_dict, fetch_list, target_list, options, run_metadata)\u001b[0m\n\u001b[1;32m   1019\u001b[0m         return tf_session.TF_Run(session, options,\n\u001b[1;32m   1020\u001b[0m                                  \u001b[0mfeed_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfetch_list\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtarget_list\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1021\u001b[0;31m                                  status, run_metadata)\n\u001b[0m\u001b[1;32m   1022\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1023\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_prun_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msession\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhandle\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeed_dict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfetch_list\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "from keras import callbacks\n",
    "import time\n",
    "tbCB = callbacks.TensorBoard(log_dir=\"logs/attr-retrain{}\".format(time.time()), histogram_freq=0, write_graph=False, write_images=False)\n",
    "checkpointCB = callbacks.ModelCheckpoint(\"checkpoints/retrain{}\".format(time.time()),period=10)\n",
    "\n",
    "h = new_model.fit_generator(generator(subset,64),steps_per_epoch=50,epochs=80,verbose=1,\n",
    "                            callbacks=[tbCB,checkpointCB],class_weight=class_weights)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "h = new_model.fit_generator(generator(joined,128),steps_per_epoch=50,epochs=80,verbose=1,\n",
    "                            callbacks=[tbCB,checkpointCB],class_weight=class_weights)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "p = new_model.predict(proc_tops[:20])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "np.where(p[0,:]>0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "np.where(p[1,:]>0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "new_model.save_weights('DF-attributes-weighted_sigmoid-7-24.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "new_model.layers[:19]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "new_model.load_weights('Data/Trained Models/DF-Retrain 7-19.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "feat_model = Model(inputs=new_model.input,outputs=new_model.get_layer('fc1').output)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "base_model = VGG16(weights='imagenet')\n",
    "\n",
    "feat_model2 = Model(inputs=base_model.input,outputs=base_model.get_layer('fc1').output)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pickle\n",
    "tops = pickle.load(open('Data/feature_matrix/tops_10000_df.p','rb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "proc_tops = []\n",
    "for index, row in tops.iloc[:100].iterrows():\n",
    "    proc_tops.append(np.squeeze(preprocess_DF(row)))\n",
    "proc_tops = np.array(proc_tops)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "start = time.time()\n",
    "fc6_retrain_tops = feat_model.predict(proc_tops)\n",
    "#pickle.dump(fc6_retrain_tops,open('Data/feature_matrix/fc6_retrain_tops.p','wb'))\n",
    "print(f'{time.time()-start:.2f} s')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "p = feat_model.predict(proc_tops[:20])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "p"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "p2 = feat_model2.predict(proc_tops[:20])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "p2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "plt.hist(new_model.get_layer('predictions').get_weights()[0][:,0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "plt.hist(new_model.get_layer('fc1').get_weights()[0][:,0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "l = list(generator(joined,4))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "for x,y in generator(joined,64):\n",
    "    print(K.get_value(H_pred(new_model.predict(x),y)))\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "plt.hist(np.sum(attr[:1000],axis=1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "229000/64"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "64*224*224"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}