classifier_confidence_rectangle_ndim_gaussian.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib notebook\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import math\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "\n",
    "from tensorflow import set_random_seed\n",
    "# for reproducibility \n",
    "np.random.seed(1)\n",
    "set_random_seed(1)\n",
    "\n",
    "from keras.layers import Dense, Input, concatenate\n",
    "from keras.layers import Conv2D, MaxPooling2D, Flatten, Lambda, Dropout, Activation, Lambda\n",
    "from keras.layers import Reshape, Conv2DTranspose\n",
    "from keras.models import Model, Sequential\n",
    "from keras.optimizers import Adam\n",
    "from keras.datasets import mnist\n",
    "from keras.datasets import fashion_mnist\n",
    "from keras.losses import mse, binary_crossentropy\n",
    "from keras.utils import plot_model\n",
    "from keras import backend as K\n",
    "\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "from keras.utils import np_utils\n",
    "\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Multi-input classifier - Rectangle outlier samples\n",
    "\n",
    "class Classifier:\n",
    "\n",
    "    # 3 Class classifier - class 2 represents outliers and class 0,1 represent inliers.\n",
    "    # Class 0 sampled from a circle centered at (x, y) = (-2, 0) with radius 0.5\n",
    "    # Class 1 sampled from a circle centered at (x, y) = (+2, 0) with radius 0.5\n",
    "    # Class 2 sampled from rectagle with origin (x, y) = (-3, -1) with h=2, w=6 excluding regions covered by inliers.\n",
    "\n",
    "    # Generate n points uniformly within a circle of radius r, centered at (x, y)\n",
    "    # return as a numpy array n x 2 of dtype = float32\n",
    "    def __init__(self):\n",
    "        self.number_of_classes = 2\n",
    "        self.input_dim = 2\n",
    "        self.epochs = 200\n",
    "        self.batch_size = 32\n",
    "        # There are 2 inlier circles\n",
    "        # Inlier circle centers on the 1st axis; for all other dimensions, the centers are at 0.\n",
    "        self.x1 = -10\n",
    "        self.x2 = +10\n",
    "        self.r = 2.0\n",
    "        # Number of inlier samples for each class\n",
    "        self.n = 1000\n",
    "        # width of the outlier outer square side\n",
    "        self.s = 100\n",
    "\n",
    "        self.data, self.labels, self.data_out, self.labels_out = self.get_data()\n",
    "        print(self.data.shape, self.labels.shape)\n",
    "        self.X_train, self.X_test, self.y_train, self.y_test = train_test_split(self.data, self.labels, test_size=0.5, shuffle=True, stratify=self.labels,\n",
    "                                                            random_state=1)\n",
    "        self.X_train_outliers, self.X_test_outliers, self.y_train_outliers, self.y_test_outliers = train_test_split(self.data_out, self.labels_out, test_size=0.5, shuffle=True,\n",
    "                                                            random_state=1)\n",
    "        self.multi_input_classifier, self.model = self.get_model()\n",
    "\n",
    "        self.mu = np.mean(np.concatenate((self.X_train, self.X_train_outliers), axis=0), axis=0)\n",
    "        self.sigma = np.std(np.concatenate((self.X_train, self.X_train_outliers), axis=0), axis=0)\n",
    "\n",
    "        self.data = self.preprocess_data(self.data)\n",
    "        self.data_out = self.preprocess_data(self.data_out)\n",
    "        self.X_train = self.preprocess_data(self.X_train)\n",
    "        self.X_test = self.preprocess_data(self.X_test)\n",
    "        self.X_train_outlier = self.preprocess_data(self.X_train_outliers)\n",
    "        self.X_test_outliers = self.preprocess_data(self.X_test_outliers)\n",
    "\n",
    "\n",
    "    # return True if point p (a list) lies inside a hypersphere centered at the origin with radius r\n",
    "    def is_in_hypersphere(self, r, p):\n",
    "        p = np.array(p)\n",
    "        return np.sum(p*p) <= r**2\n",
    "\n",
    "\n",
    "    # return True if point p (a list) lies inside a hypersphere centered at the (x, 0, 0, ...) with radius r\n",
    "    def is_in_hypersphere_centered_at(self, x, r, p):\n",
    "        p = np.array(p)\n",
    "        p[0] -= x\n",
    "        return np.sum(p*p) <= r**2\n",
    "\n",
    "\n",
    "    # uniformly sample a point from a square centered at the origin with side = s\n",
    "    def sample_point_from_square(self, s):\n",
    "        p = []\n",
    "        # first point\n",
    "        for i in range(0, self.input_dim):\n",
    "            p.append(np.random.uniform(-1*s/2, s/2, 1))\n",
    "        return p\n",
    "\n",
    "\n",
    "    def preprocess_data(self, x):\n",
    "        #return (x-self.mu)/self.sigma\n",
    "        return x\n",
    "\n",
    "    # sample points from 2D gaussian centered at origin\n",
    "    def sample_point_from_gaussian(self, sigma):\n",
    "        mean = np.zeros(self.input_dim)\n",
    "        cov = np.zeros((self.input_dim, self.input_dim), np.float32)\n",
    "        np.fill_diagonal(cov, sigma*sigma)\n",
    "        p = list(np.random.multivariate_normal(mean, cov, 1).T.squeeze())\n",
    "        return p\n",
    "\n",
    "        \n",
    "    # sample from a hypersphre with rejection sampling\n",
    "    def get_gaussian_data(self, x, r, n):\n",
    "        data = np.zeros((n, self.input_dim))\n",
    "        count = 0\n",
    "        while count<n:\n",
    "            data[count, :] = self.sample_point_from_gaussian(r)\n",
    "            # update the point with offset for the 1st dimension\n",
    "            data[count, 0] += x\n",
    "            count += 1\n",
    "\n",
    "        data = data.astype(np.float32)\n",
    "        return data\n",
    "        # plt.scatter(x, y, s = 4)\n",
    "\n",
    "    # generate n outlier points uniformly sampled within square centered at the origin\n",
    "    # with side=s, excuding the inlier circles\n",
    "    def get_outlier_data_sub(self, s, x1, x2, r, n):\n",
    "        count = 0\n",
    "        data = np.zeros((n, self.input_dim))\n",
    "        while count < n:\n",
    "            p = self.sample_point_from_square(s)\n",
    "            if (not self.is_in_hypersphere_centered_at(x1, r, p)) and (not self.is_in_hypersphere_centered_at(x2, r, p)):\n",
    "                data[count, :] = p\n",
    "                count += 1\n",
    "        return data\n",
    "\n",
    "    def get_outlier_data(self):\n",
    "        data = self.get_outlier_data_sub(self.s, self.x1, self.x2, self.r*3, 2*self.n)\n",
    "        return data.astype(np.float32)\n",
    "\n",
    "        \n",
    "    def get_data(self):\n",
    "        # class 0 data\n",
    "        data = self.get_gaussian_data(self.x1, self.r, self.n)\n",
    "        labels = np.ones((self.n, 1))*0\n",
    "        print(labels.shape)\n",
    "        # class 1 data\n",
    "        data = np.concatenate((data, self.get_gaussian_data(self.x2, self.r, self.n)), axis=0)\n",
    "        labels = np.concatenate((labels, np.ones((self.n, 1))*1), axis=0)\n",
    "        print(labels.shape)\n",
    "        # outlier data\n",
    "        data_out = self.get_outlier_data()\n",
    "        labels_out = np.ones((2*self.n,1))*2\n",
    "        print(labels.shape)\n",
    "        return data.astype(np.float32), labels.astype(np.float32), data_out.astype(np.float32), labels_out.astype(np.float32)\n",
    "\n",
    "    \n",
    "    def base_model(self):\n",
    "        model = Sequential()\n",
    "        model.add(Dense(500, input_shape=(self.input_dim,)))\n",
    "        model.add(Activation('relu'))\n",
    "        model.add(Dense(500))\n",
    "        model.add(Activation('relu'))\n",
    "        model.add(Dense(self.number_of_classes))\n",
    "        # model.add(Convolution2D(10,3,3, border_mode='same'))\n",
    "        # model.add(GlobalAveragePooling2D())\n",
    "        model.add(Activation('softmax'))\n",
    "        #model.compile(loss='binary_crossentropy', optimizer=Adam(0.001), metrics=['accuracy'])\n",
    "        return model\n",
    "\n",
    "    def get_model(self):\n",
    "        # instantiate classifier model\n",
    "        classifier = self.base_model()\n",
    "        classifier.compile(loss='categorical_crossentropy', optimizer=Adam(0.001))\n",
    "\n",
    "        # just the classifier\n",
    "        input_data = Input(shape=(self.input_dim, ))\n",
    "        y_classifier = classifier(input_data)\n",
    "\n",
    "\n",
    "        # 2 input classifier\n",
    "        input_a = Input(shape=(self.input_dim, ))\n",
    "        processed_a = classifier(input_a)\n",
    "        input_b = Input(shape=(self.input_dim, ))\n",
    "        processed_b = classifier(input_b)\n",
    "\n",
    "        out_a = Lambda(lambda x: x, name = 'out_a')(processed_a)\n",
    "        out_b = Lambda(lambda x: x, name = 'out_b')(processed_b)\n",
    "\n",
    "        multi_input_classifier = Model(inputs=[input_a, input_b], outputs=[out_a, out_b])\n",
    "        multi_input_classifier.compile(loss={'out_a':'categorical_crossentropy', 'out_b':'categorical_crossentropy'}, \n",
    "                                       loss_weights={'out_a': 1.0, 'out_b': 1.0}, optimizer=Adam(0.001), \n",
    "                                       metrics={'out_a':['acc'], 'out_b':['categorical_crossentropy']})\n",
    "\n",
    "\n",
    "        mnist_model = Model(input_data, y_classifier)\n",
    "        mnist_model.compile(loss='categorical_crossentropy', optimizer=Adam(0.001), metrics=['acc'])\n",
    "\n",
    "        return multi_input_classifier, mnist_model\n",
    "\n",
    "\n",
    "    def train(self):\n",
    "        # fit the model\n",
    "        Y_train = np_utils.to_categorical(self.y_train, self.number_of_classes)\n",
    "        Y_train_outliers = np.ones((self.y_train_outliers.shape[0], self.number_of_classes))\n",
    "        Y_test = np_utils.to_categorical(self.y_test, self.number_of_classes)\n",
    "        Y_test_outliers = np.ones((self.y_test_outliers.shape[0], self.number_of_classes))\n",
    "\n",
    "        self.multi_input_classifier.fit([self.X_train, self.X_train_outliers], [Y_train, Y_train_outliers], \n",
    "                                        epochs=self.epochs, validation_data=([self.X_test, self.X_test_outliers], [Y_test, Y_test_outliers]), batch_size=self.batch_size)     \n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1000, 1)\n",
      "(2000, 1)\n",
      "(2000, 1)\n",
      "((2000, 2), (2000, 1))\n",
      "Train on 1000 samples, validate on 1000 samples\n",
      "Epoch 1/200\n",
      "1000/1000 [==============================] - 0s 415us/step - loss: 2.6484 - out_a_loss: 0.5141 - out_b_loss: 2.1343 - out_a_acc: 0.8660 - out_b_categorical_crossentropy: 2.1343 - val_loss: 1.9879 - val_out_a_loss: 0.4369 - val_out_b_loss: 1.5509 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.5509\n",
      "Epoch 2/200\n",
      "1000/1000 [==============================] - 0s 113us/step - loss: 1.9467 - out_a_loss: 0.4437 - out_b_loss: 1.5030 - out_a_acc: 0.9990 - out_b_categorical_crossentropy: 1.5030 - val_loss: 1.8762 - val_out_a_loss: 0.4162 - val_out_b_loss: 1.4599 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4599\n",
      "Epoch 3/200\n",
      "1000/1000 [==============================] - 0s 115us/step - loss: 1.8693 - out_a_loss: 0.3696 - out_b_loss: 1.4997 - out_a_acc: 0.9990 - out_b_categorical_crossentropy: 1.4997 - val_loss: 1.8499 - val_out_a_loss: 0.3356 - val_out_b_loss: 1.5143 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.5143\n",
      "Epoch 4/200\n",
      "1000/1000 [==============================] - 0s 120us/step - loss: 1.9314 - out_a_loss: 0.2950 - out_b_loss: 1.6364 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.6364 - val_loss: 1.7821 - val_out_a_loss: 0.2588 - val_out_b_loss: 1.5233 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.5233\n",
      "Epoch 5/200\n",
      "1000/1000 [==============================] - 0s 113us/step - loss: 1.7035 - out_a_loss: 0.2267 - out_b_loss: 1.4768 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4768 - val_loss: 1.6314 - val_out_a_loss: 0.1674 - val_out_b_loss: 1.4641 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4641\n",
      "Epoch 6/200\n",
      "1000/1000 [==============================] - 0s 117us/step - loss: 1.6111 - out_a_loss: 0.1526 - out_b_loss: 1.4586 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4586 - val_loss: 1.5743 - val_out_a_loss: 0.1192 - val_out_b_loss: 1.4551 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4551\n",
      "Epoch 7/200\n",
      "1000/1000 [==============================] - 0s 115us/step - loss: 1.6132 - out_a_loss: 0.1072 - out_b_loss: 1.5060 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.5060 - val_loss: 1.6004 - val_out_a_loss: 0.1009 - val_out_b_loss: 1.4996 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4996\n",
      "Epoch 8/200\n",
      "1000/1000 [==============================] - 0s 129us/step - loss: 1.6054 - out_a_loss: 0.0917 - out_b_loss: 1.5136 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.5136 - val_loss: 1.5490 - val_out_a_loss: 0.0857 - val_out_b_loss: 1.4634 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4634\n",
      "Epoch 9/200\n",
      "1000/1000 [==============================] - 0s 114us/step - loss: 1.5600 - out_a_loss: 0.0748 - out_b_loss: 1.4852 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4852 - val_loss: 1.5175 - val_out_a_loss: 0.0709 - val_out_b_loss: 1.4466 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4466\n",
      "Epoch 10/200\n",
      "1000/1000 [==============================] - 0s 117us/step - loss: 1.5202 - out_a_loss: 0.0629 - out_b_loss: 1.4573 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4573 - val_loss: 1.5078 - val_out_a_loss: 0.0554 - val_out_b_loss: 1.4524 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4524\n",
      "Epoch 11/200\n",
      "1000/1000 [==============================] - 0s 112us/step - loss: 1.5085 - out_a_loss: 0.0524 - out_b_loss: 1.4561 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4561 - val_loss: 1.5168 - val_out_a_loss: 0.0431 - val_out_b_loss: 1.4737 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4737\n",
      "Epoch 12/200\n",
      "1000/1000 [==============================] - 0s 114us/step - loss: 1.4978 - out_a_loss: 0.0457 - out_b_loss: 1.4521 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4521 - val_loss: 1.5285 - val_out_a_loss: 0.0384 - val_out_b_loss: 1.4902 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4902\n",
      "Epoch 13/200\n",
      "1000/1000 [==============================] - 0s 119us/step - loss: 1.5109 - out_a_loss: 0.0399 - out_b_loss: 1.4710 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4710 - val_loss: 1.4896 - val_out_a_loss: 0.0388 - val_out_b_loss: 1.4508 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4508\n",
      "Epoch 14/200\n",
      "1000/1000 [==============================] - 0s 120us/step - loss: 1.4914 - out_a_loss: 0.0366 - out_b_loss: 1.4548 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4548 - val_loss: 1.4981 - val_out_a_loss: 0.0361 - val_out_b_loss: 1.4619 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4619\n",
      "Epoch 15/200\n",
      "1000/1000 [==============================] - 0s 109us/step - loss: 1.4747 - out_a_loss: 0.0336 - out_b_loss: 1.4411 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4411 - val_loss: 1.4668 - val_out_a_loss: 0.0270 - val_out_b_loss: 1.4398 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4398\n",
      "Epoch 16/200\n",
      "1000/1000 [==============================] - 0s 118us/step - loss: 1.4680 - out_a_loss: 0.0322 - out_b_loss: 1.4358 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4358 - val_loss: 1.4598 - val_out_a_loss: 0.0257 - val_out_b_loss: 1.4342 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4342\n",
      "Epoch 17/200\n",
      "1000/1000 [==============================] - 0s 120us/step - loss: 1.4522 - out_a_loss: 0.0297 - out_b_loss: 1.4225 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4225 - val_loss: 1.4494 - val_out_a_loss: 0.0242 - val_out_b_loss: 1.4251 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4251\n",
      "Epoch 18/200\n",
      "1000/1000 [==============================] - 0s 112us/step - loss: 1.4489 - out_a_loss: 0.0243 - out_b_loss: 1.4246 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4246 - val_loss: 1.4601 - val_out_a_loss: 0.0218 - val_out_b_loss: 1.4383 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4383\n",
      "Epoch 19/200\n",
      "1000/1000 [==============================] - 0s 134us/step - loss: 1.4444 - out_a_loss: 0.0251 - out_b_loss: 1.4192 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4192 - val_loss: 1.4556 - val_out_a_loss: 0.0152 - val_out_b_loss: 1.4404 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4404\n",
      "Epoch 20/200\n",
      "1000/1000 [==============================] - 0s 106us/step - loss: 1.4423 - out_a_loss: 0.0209 - out_b_loss: 1.4214 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4214 - val_loss: 1.4442 - val_out_a_loss: 0.0202 - val_out_b_loss: 1.4239 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4239\n",
      "Epoch 21/200\n",
      "1000/1000 [==============================] - 0s 133us/step - loss: 1.4376 - out_a_loss: 0.0208 - out_b_loss: 1.4168 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4168 - val_loss: 1.4510 - val_out_a_loss: 0.0140 - val_out_b_loss: 1.4370 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4370\n",
      "Epoch 22/200\n",
      "1000/1000 [==============================] - 0s 114us/step - loss: 1.4391 - out_a_loss: 0.0185 - out_b_loss: 1.4206 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4206 - val_loss: 1.4422 - val_out_a_loss: 0.0143 - val_out_b_loss: 1.4280 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4280\n",
      "Epoch 23/200\n",
      "1000/1000 [==============================] - 0s 116us/step - loss: 1.4367 - out_a_loss: 0.0189 - out_b_loss: 1.4178 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4178 - val_loss: 1.4377 - val_out_a_loss: 0.0219 - val_out_b_loss: 1.4158 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4158\n",
      "Epoch 24/200\n",
      "1000/1000 [==============================] - 0s 111us/step - loss: 1.4312 - out_a_loss: 0.0171 - out_b_loss: 1.4141 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4141 - val_loss: 1.4334 - val_out_a_loss: 0.0198 - val_out_b_loss: 1.4136 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4136\n",
      "Epoch 25/200\n",
      "1000/1000 [==============================] - 0s 119us/step - loss: 1.4279 - out_a_loss: 0.0175 - out_b_loss: 1.4104 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4104 - val_loss: 1.4429 - val_out_a_loss: 0.0089 - val_out_b_loss: 1.4339 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4339\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 26/200\n",
      "1000/1000 [==============================] - 0s 116us/step - loss: 1.4257 - out_a_loss: 0.0150 - out_b_loss: 1.4107 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4107 - val_loss: 1.4303 - val_out_a_loss: 0.0125 - val_out_b_loss: 1.4178 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4178\n",
      "Epoch 27/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.4282 - out_a_loss: 0.0155 - out_b_loss: 1.4126 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4126 - val_loss: 1.4320 - val_out_a_loss: 0.0207 - val_out_b_loss: 1.4113 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4113\n",
      "Epoch 28/200\n",
      "1000/1000 [==============================] - 0s 111us/step - loss: 1.4242 - out_a_loss: 0.0182 - out_b_loss: 1.4059 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4059 - val_loss: 1.4224 - val_out_a_loss: 0.0128 - val_out_b_loss: 1.4096 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4096\n",
      "Epoch 29/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.4221 - out_a_loss: 0.0147 - out_b_loss: 1.4074 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4074 - val_loss: 1.4278 - val_out_a_loss: 0.0140 - val_out_b_loss: 1.4138 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4138\n",
      "Epoch 30/200\n",
      "1000/1000 [==============================] - 0s 107us/step - loss: 1.4199 - out_a_loss: 0.0146 - out_b_loss: 1.4053 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4053 - val_loss: 1.4232 - val_out_a_loss: 0.0127 - val_out_b_loss: 1.4105 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4105\n",
      "Epoch 31/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.4192 - out_a_loss: 0.0131 - out_b_loss: 1.4061 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4061 - val_loss: 1.4181 - val_out_a_loss: 0.0134 - val_out_b_loss: 1.4047 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4047\n",
      "Epoch 32/200\n",
      "1000/1000 [==============================] - 0s 102us/step - loss: 1.4245 - out_a_loss: 0.0156 - out_b_loss: 1.4089 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4089 - val_loss: 1.4327 - val_out_a_loss: 0.0202 - val_out_b_loss: 1.4125 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4125\n",
      "Epoch 33/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.4276 - out_a_loss: 0.0145 - out_b_loss: 1.4131 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4131 - val_loss: 1.4338 - val_out_a_loss: 0.0133 - val_out_b_loss: 1.4205 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4205\n",
      "Epoch 34/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.4195 - out_a_loss: 0.0114 - out_b_loss: 1.4081 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4081 - val_loss: 1.4216 - val_out_a_loss: 0.0103 - val_out_b_loss: 1.4114 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4114\n",
      "Epoch 35/200\n",
      "1000/1000 [==============================] - 0s 108us/step - loss: 1.4145 - out_a_loss: 0.0122 - out_b_loss: 1.4024 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4024 - val_loss: 1.4163 - val_out_a_loss: 0.0084 - val_out_b_loss: 1.4079 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4079\n",
      "Epoch 36/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.4111 - out_a_loss: 0.0122 - out_b_loss: 1.3989 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3989 - val_loss: 1.4169 - val_out_a_loss: 0.0111 - val_out_b_loss: 1.4059 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4059\n",
      "Epoch 37/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.4160 - out_a_loss: 0.0136 - out_b_loss: 1.4025 - out_a_acc: 0.9990 - out_b_categorical_crossentropy: 1.4025 - val_loss: 1.4165 - val_out_a_loss: 0.0114 - val_out_b_loss: 1.4050 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4050\n",
      "Epoch 38/200\n",
      "1000/1000 [==============================] - 0s 102us/step - loss: 1.4129 - out_a_loss: 0.0117 - out_b_loss: 1.4012 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4012 - val_loss: 1.4181 - val_out_a_loss: 0.0068 - val_out_b_loss: 1.4113 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4113\n",
      "Epoch 39/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.4131 - out_a_loss: 0.0113 - out_b_loss: 1.4018 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4018 - val_loss: 1.4190 - val_out_a_loss: 0.0173 - val_out_b_loss: 1.4017 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4017\n",
      "Epoch 40/200\n",
      "1000/1000 [==============================] - 0s 106us/step - loss: 1.4116 - out_a_loss: 0.0116 - out_b_loss: 1.4000 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4000 - val_loss: 1.4161 - val_out_a_loss: 0.0167 - val_out_b_loss: 1.3994 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3994\n",
      "Epoch 41/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.4122 - out_a_loss: 0.0125 - out_b_loss: 1.3996 - out_a_acc: 0.9990 - out_b_categorical_crossentropy: 1.3996 - val_loss: 1.4116 - val_out_a_loss: 0.0088 - val_out_b_loss: 1.4027 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4027\n",
      "Epoch 42/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.4105 - out_a_loss: 0.0117 - out_b_loss: 1.3988 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3988 - val_loss: 1.4309 - val_out_a_loss: 0.0069 - val_out_b_loss: 1.4241 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4241\n",
      "Epoch 43/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.4098 - out_a_loss: 0.0103 - out_b_loss: 1.3995 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3995 - val_loss: 1.4152 - val_out_a_loss: 0.0055 - val_out_b_loss: 1.4097 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4097\n",
      "Epoch 44/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.4083 - out_a_loss: 0.0105 - out_b_loss: 1.3978 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3978 - val_loss: 1.4170 - val_out_a_loss: 0.0083 - val_out_b_loss: 1.4087 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4087\n",
      "Epoch 45/200\n",
      "1000/1000 [==============================] - 0s 109us/step - loss: 1.4155 - out_a_loss: 0.0129 - out_b_loss: 1.4026 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4026 - val_loss: 1.4147 - val_out_a_loss: 0.0072 - val_out_b_loss: 1.4074 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4074\n",
      "Epoch 46/200\n",
      "1000/1000 [==============================] - 0s 112us/step - loss: 1.4201 - out_a_loss: 0.0132 - out_b_loss: 1.4069 - out_a_acc: 0.9990 - out_b_categorical_crossentropy: 1.4069 - val_loss: 1.4087 - val_out_a_loss: 0.0102 - val_out_b_loss: 1.3985 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3985\n",
      "Epoch 47/200\n",
      "1000/1000 [==============================] - 0s 107us/step - loss: 1.4094 - out_a_loss: 0.0111 - out_b_loss: 1.3983 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3983 - val_loss: 1.4078 - val_out_a_loss: 0.0112 - val_out_b_loss: 1.3966 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3966\n",
      "Epoch 48/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.4088 - out_a_loss: 0.0112 - out_b_loss: 1.3976 - out_a_acc: 0.9990 - out_b_categorical_crossentropy: 1.3976 - val_loss: 1.4119 - val_out_a_loss: 0.0154 - val_out_b_loss: 1.3965 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3965\n",
      "Epoch 49/200\n",
      "1000/1000 [==============================] - 0s 107us/step - loss: 1.4061 - out_a_loss: 0.0104 - out_b_loss: 1.3957 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3957 - val_loss: 1.4121 - val_out_a_loss: 0.0051 - val_out_b_loss: 1.4069 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4069\n",
      "Epoch 50/200\n",
      "1000/1000 [==============================] - 0s 111us/step - loss: 1.4046 - out_a_loss: 0.0089 - out_b_loss: 1.3957 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3957 - val_loss: 1.4218 - val_out_a_loss: 0.0069 - val_out_b_loss: 1.4149 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4149\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 51/200\n",
      "1000/1000 [==============================] - 0s 114us/step - loss: 1.4106 - out_a_loss: 0.0131 - out_b_loss: 1.3976 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3976 - val_loss: 1.4338 - val_out_a_loss: 0.0074 - val_out_b_loss: 1.4264 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4264\n",
      "Epoch 52/200\n",
      "1000/1000 [==============================] - 0s 102us/step - loss: 1.4104 - out_a_loss: 0.0105 - out_b_loss: 1.3999 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3999 - val_loss: 1.4122 - val_out_a_loss: 0.0070 - val_out_b_loss: 1.4052 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4052\n",
      "Epoch 53/200\n",
      "1000/1000 [==============================] - 0s 102us/step - loss: 1.4073 - out_a_loss: 0.0090 - out_b_loss: 1.3982 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3982 - val_loss: 1.4071 - val_out_a_loss: 0.0090 - val_out_b_loss: 1.3981 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3981\n",
      "Epoch 54/200\n",
      "1000/1000 [==============================] - 0s 106us/step - loss: 1.4058 - out_a_loss: 0.0120 - out_b_loss: 1.3938 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3938 - val_loss: 1.4099 - val_out_a_loss: 0.0074 - val_out_b_loss: 1.4024 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4024\n",
      "Epoch 55/200\n",
      "1000/1000 [==============================] - 0s 102us/step - loss: 1.4039 - out_a_loss: 0.0079 - out_b_loss: 1.3960 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3960 - val_loss: 1.4071 - val_out_a_loss: 0.0093 - val_out_b_loss: 1.3978 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3978\n",
      "Epoch 56/200\n",
      "1000/1000 [==============================] - 0s 102us/step - loss: 1.4042 - out_a_loss: 0.0081 - out_b_loss: 1.3962 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3962 - val_loss: 1.4077 - val_out_a_loss: 0.0078 - val_out_b_loss: 1.3999 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3999\n",
      "Epoch 57/200\n",
      "1000/1000 [==============================] - 0s 101us/step - loss: 1.4033 - out_a_loss: 0.0097 - out_b_loss: 1.3937 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3937 - val_loss: 1.4127 - val_out_a_loss: 0.0072 - val_out_b_loss: 1.4055 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.4055\n",
      "Epoch 58/200\n",
      "1000/1000 [==============================] - 0s 106us/step - loss: 1.4083 - out_a_loss: 0.0088 - out_b_loss: 1.3995 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3995 - val_loss: 1.4060 - val_out_a_loss: 0.0069 - val_out_b_loss: 1.3991 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3991\n",
      "Epoch 59/200\n",
      "1000/1000 [==============================] - 0s 114us/step - loss: 1.4065 - out_a_loss: 0.0114 - out_b_loss: 1.3951 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3951 - val_loss: 1.4101 - val_out_a_loss: 0.0063 - val_out_b_loss: 1.4038 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4038\n",
      "Epoch 60/200\n",
      "1000/1000 [==============================] - 0s 112us/step - loss: 1.4039 - out_a_loss: 0.0082 - out_b_loss: 1.3956 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3956 - val_loss: 1.4050 - val_out_a_loss: 0.0063 - val_out_b_loss: 1.3987 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3987\n",
      "Epoch 61/200\n",
      "1000/1000 [==============================] - 0s 102us/step - loss: 1.4007 - out_a_loss: 0.0073 - out_b_loss: 1.3934 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3934 - val_loss: 1.4032 - val_out_a_loss: 0.0060 - val_out_b_loss: 1.3973 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3973\n",
      "Epoch 62/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.4010 - out_a_loss: 0.0083 - out_b_loss: 1.3927 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3927 - val_loss: 1.4083 - val_out_a_loss: 0.0049 - val_out_b_loss: 1.4034 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4034\n",
      "Epoch 63/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.4053 - out_a_loss: 0.0091 - out_b_loss: 1.3962 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3962 - val_loss: 1.4106 - val_out_a_loss: 0.0046 - val_out_b_loss: 1.4060 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4060\n",
      "Epoch 64/200\n",
      "1000/1000 [==============================] - 0s 110us/step - loss: 1.4080 - out_a_loss: 0.0091 - out_b_loss: 1.3989 - out_a_acc: 0.9990 - out_b_categorical_crossentropy: 1.3989 - val_loss: 1.4305 - val_out_a_loss: 0.0022 - val_out_b_loss: 1.4283 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4283\n",
      "Epoch 65/200\n",
      "1000/1000 [==============================] - 0s 119us/step - loss: 1.4174 - out_a_loss: 0.0115 - out_b_loss: 1.4059 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.4059 - val_loss: 1.4062 - val_out_a_loss: 0.0083 - val_out_b_loss: 1.3979 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3979\n",
      "Epoch 66/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.4087 - out_a_loss: 0.0103 - out_b_loss: 1.3985 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3985 - val_loss: 1.4057 - val_out_a_loss: 0.0103 - val_out_b_loss: 1.3954 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3954\n",
      "Epoch 67/200\n",
      "1000/1000 [==============================] - 0s 113us/step - loss: 1.4022 - out_a_loss: 0.0081 - out_b_loss: 1.3941 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3941 - val_loss: 1.4030 - val_out_a_loss: 0.0098 - val_out_b_loss: 1.3932 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3932\n",
      "Epoch 68/200\n",
      "1000/1000 [==============================] - 0s 125us/step - loss: 1.4050 - out_a_loss: 0.0095 - out_b_loss: 1.3955 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3955 - val_loss: 1.4071 - val_out_a_loss: 0.0097 - val_out_b_loss: 1.3974 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3974\n",
      "Epoch 69/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.4033 - out_a_loss: 0.0076 - out_b_loss: 1.3956 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3956 - val_loss: 1.4034 - val_out_a_loss: 0.0061 - val_out_b_loss: 1.3973 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3973\n",
      "Epoch 70/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.4003 - out_a_loss: 0.0077 - out_b_loss: 1.3926 - out_a_acc: 0.9990 - out_b_categorical_crossentropy: 1.3926 - val_loss: 1.4119 - val_out_a_loss: 0.0063 - val_out_b_loss: 1.4056 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4056\n",
      "Epoch 71/200\n",
      "1000/1000 [==============================] - 0s 109us/step - loss: 1.4031 - out_a_loss: 0.0086 - out_b_loss: 1.3946 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3946 - val_loss: 1.4022 - val_out_a_loss: 0.0104 - val_out_b_loss: 1.3917 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3917\n",
      "Epoch 72/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.4010 - out_a_loss: 0.0078 - out_b_loss: 1.3932 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3932 - val_loss: 1.4014 - val_out_a_loss: 0.0104 - val_out_b_loss: 1.3910 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3910\n",
      "Epoch 73/200\n",
      "1000/1000 [==============================] - 0s 115us/step - loss: 1.4006 - out_a_loss: 0.0075 - out_b_loss: 1.3931 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3931 - val_loss: 1.4073 - val_out_a_loss: 0.0078 - val_out_b_loss: 1.3995 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3995\n",
      "Epoch 74/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.4095 - out_a_loss: 0.0103 - out_b_loss: 1.3991 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3991 - val_loss: 1.4071 - val_out_a_loss: 0.0097 - val_out_b_loss: 1.3973 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3973\n",
      "Epoch 75/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.4073 - out_a_loss: 0.0114 - out_b_loss: 1.3959 - out_a_acc: 0.9990 - out_b_categorical_crossentropy: 1.3959 - val_loss: 1.4032 - val_out_a_loss: 0.0039 - val_out_b_loss: 1.3993 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3993\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 76/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.4014 - out_a_loss: 0.0074 - out_b_loss: 1.3939 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3939 - val_loss: 1.4026 - val_out_a_loss: 0.0071 - val_out_b_loss: 1.3955 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3955\n",
      "Epoch 77/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.3989 - out_a_loss: 0.0064 - out_b_loss: 1.3925 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3925 - val_loss: 1.4060 - val_out_a_loss: 0.0050 - val_out_b_loss: 1.4010 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4010\n",
      "Epoch 78/200\n",
      "1000/1000 [==============================] - 0s 109us/step - loss: 1.4058 - out_a_loss: 0.0104 - out_b_loss: 1.3954 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3954 - val_loss: 1.4039 - val_out_a_loss: 0.0130 - val_out_b_loss: 1.3909 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3909\n",
      "Epoch 79/200\n",
      "1000/1000 [==============================] - 0s 102us/step - loss: 1.4026 - out_a_loss: 0.0086 - out_b_loss: 1.3940 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3940 - val_loss: 1.4174 - val_out_a_loss: 0.0244 - val_out_b_loss: 1.3931 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3931\n",
      "Epoch 80/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.3972 - out_a_loss: 0.0068 - out_b_loss: 1.3904 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3904 - val_loss: 1.4019 - val_out_a_loss: 0.0048 - val_out_b_loss: 1.3970 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3970\n",
      "Epoch 81/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.4016 - out_a_loss: 0.0085 - out_b_loss: 1.3932 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3932 - val_loss: 1.4028 - val_out_a_loss: 0.0041 - val_out_b_loss: 1.3987 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3987\n",
      "Epoch 82/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.3998 - out_a_loss: 0.0077 - out_b_loss: 1.3921 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3921 - val_loss: 1.4006 - val_out_a_loss: 0.0069 - val_out_b_loss: 1.3936 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3936\n",
      "Epoch 83/200\n",
      "1000/1000 [==============================] - 0s 111us/step - loss: 1.4013 - out_a_loss: 0.0075 - out_b_loss: 1.3939 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3939 - val_loss: 1.4088 - val_out_a_loss: 0.0026 - val_out_b_loss: 1.4062 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4062\n",
      "Epoch 84/200\n",
      "1000/1000 [==============================] - 0s 102us/step - loss: 1.4001 - out_a_loss: 0.0077 - out_b_loss: 1.3925 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3925 - val_loss: 1.4082 - val_out_a_loss: 0.0172 - val_out_b_loss: 1.3911 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3911\n",
      "Epoch 85/200\n",
      "1000/1000 [==============================] - 0s 107us/step - loss: 1.4004 - out_a_loss: 0.0085 - out_b_loss: 1.3919 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3919 - val_loss: 1.4006 - val_out_a_loss: 0.0075 - val_out_b_loss: 1.3931 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3931\n",
      "Epoch 86/200\n",
      "1000/1000 [==============================] - 0s 102us/step - loss: 1.3979 - out_a_loss: 0.0059 - out_b_loss: 1.3920 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3920 - val_loss: 1.4013 - val_out_a_loss: 0.0063 - val_out_b_loss: 1.3949 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3949\n",
      "Epoch 87/200\n",
      "1000/1000 [==============================] - 0s 102us/step - loss: 1.3981 - out_a_loss: 0.0072 - out_b_loss: 1.3910 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3910 - val_loss: 1.4086 - val_out_a_loss: 0.0024 - val_out_b_loss: 1.4062 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4062\n",
      "Epoch 88/200\n",
      "1000/1000 [==============================] - 0s 111us/step - loss: 1.3989 - out_a_loss: 0.0071 - out_b_loss: 1.3919 - out_a_acc: 0.9990 - out_b_categorical_crossentropy: 1.3919 - val_loss: 1.4024 - val_out_a_loss: 0.0063 - val_out_b_loss: 1.3961 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3961\n",
      "Epoch 89/200\n",
      "1000/1000 [==============================] - 0s 101us/step - loss: 1.3978 - out_a_loss: 0.0059 - out_b_loss: 1.3919 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3919 - val_loss: 1.4009 - val_out_a_loss: 0.0065 - val_out_b_loss: 1.3945 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3945\n",
      "Epoch 90/200\n",
      "1000/1000 [==============================] - 0s 113us/step - loss: 1.3962 - out_a_loss: 0.0051 - out_b_loss: 1.3910 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3910 - val_loss: 1.4128 - val_out_a_loss: 0.0216 - val_out_b_loss: 1.3912 - val_out_a_acc: 0.9970 - val_out_b_categorical_crossentropy: 1.3912\n",
      "Epoch 91/200\n",
      "1000/1000 [==============================] - 0s 130us/step - loss: 1.4008 - out_a_loss: 0.0082 - out_b_loss: 1.3926 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3926 - val_loss: 1.4007 - val_out_a_loss: 0.0091 - val_out_b_loss: 1.3917 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3917\n",
      "Epoch 92/200\n",
      "1000/1000 [==============================] - 0s 111us/step - loss: 1.4020 - out_a_loss: 0.0084 - out_b_loss: 1.3935 - out_a_acc: 0.9990 - out_b_categorical_crossentropy: 1.3935 - val_loss: 1.4034 - val_out_a_loss: 0.0079 - val_out_b_loss: 1.3955 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3955\n",
      "Epoch 93/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.4011 - out_a_loss: 0.0081 - out_b_loss: 1.3930 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3930 - val_loss: 1.4048 - val_out_a_loss: 0.0079 - val_out_b_loss: 1.3969 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3969\n",
      "Epoch 94/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.4022 - out_a_loss: 0.0069 - out_b_loss: 1.3954 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3954 - val_loss: 1.4048 - val_out_a_loss: 0.0117 - val_out_b_loss: 1.3931 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3931\n",
      "Epoch 95/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.3970 - out_a_loss: 0.0053 - out_b_loss: 1.3918 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3918 - val_loss: 1.4026 - val_out_a_loss: 0.0049 - val_out_b_loss: 1.3976 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3976\n",
      "Epoch 96/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.3987 - out_a_loss: 0.0063 - out_b_loss: 1.3924 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3924 - val_loss: 1.4041 - val_out_a_loss: 0.0062 - val_out_b_loss: 1.3979 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3979\n",
      "Epoch 97/200\n",
      "1000/1000 [==============================] - 0s 110us/step - loss: 1.4020 - out_a_loss: 0.0074 - out_b_loss: 1.3946 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3946 - val_loss: 1.4031 - val_out_a_loss: 0.0066 - val_out_b_loss: 1.3965 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3965\n",
      "Epoch 98/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.4012 - out_a_loss: 0.0088 - out_b_loss: 1.3924 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3924 - val_loss: 1.4062 - val_out_a_loss: 0.0091 - val_out_b_loss: 1.3970 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3970\n",
      "Epoch 99/200\n",
      "1000/1000 [==============================] - 0s 109us/step - loss: 1.3988 - out_a_loss: 0.0056 - out_b_loss: 1.3932 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3932 - val_loss: 1.4021 - val_out_a_loss: 0.0070 - val_out_b_loss: 1.3951 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3951\n",
      "Epoch 100/200\n",
      "1000/1000 [==============================] - 0s 106us/step - loss: 1.3996 - out_a_loss: 0.0066 - out_b_loss: 1.3930 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3930 - val_loss: 1.4048 - val_out_a_loss: 0.0144 - val_out_b_loss: 1.3904 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3904\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 101/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.3990 - out_a_loss: 0.0071 - out_b_loss: 1.3919 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3919 - val_loss: 1.4031 - val_out_a_loss: 0.0079 - val_out_b_loss: 1.3952 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3952\n",
      "Epoch 102/200\n",
      "1000/1000 [==============================] - 0s 107us/step - loss: 1.4016 - out_a_loss: 0.0071 - out_b_loss: 1.3945 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3945 - val_loss: 1.4089 - val_out_a_loss: 0.0125 - val_out_b_loss: 1.3963 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3963\n",
      "Epoch 103/200\n",
      "1000/1000 [==============================] - 0s 102us/step - loss: 1.4057 - out_a_loss: 0.0093 - out_b_loss: 1.3964 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3964 - val_loss: 1.4037 - val_out_a_loss: 0.0094 - val_out_b_loss: 1.3943 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3943\n",
      "Epoch 104/200\n",
      "1000/1000 [==============================] - 0s 107us/step - loss: 1.3991 - out_a_loss: 0.0065 - out_b_loss: 1.3926 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3926 - val_loss: 1.4099 - val_out_a_loss: 0.0045 - val_out_b_loss: 1.4055 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4055\n",
      "Epoch 105/200\n",
      "1000/1000 [==============================] - 0s 106us/step - loss: 1.4021 - out_a_loss: 0.0080 - out_b_loss: 1.3941 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3941 - val_loss: 1.4028 - val_out_a_loss: 0.0045 - val_out_b_loss: 1.3983 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3983\n",
      "Epoch 106/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.3971 - out_a_loss: 0.0047 - out_b_loss: 1.3923 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3923 - val_loss: 1.4060 - val_out_a_loss: 0.0036 - val_out_b_loss: 1.4025 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4025\n",
      "Epoch 107/200\n",
      "1000/1000 [==============================] - 0s 111us/step - loss: 1.4021 - out_a_loss: 0.0096 - out_b_loss: 1.3925 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3925 - val_loss: 1.4154 - val_out_a_loss: 0.0023 - val_out_b_loss: 1.4130 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4130\n",
      "Epoch 108/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.3991 - out_a_loss: 0.0064 - out_b_loss: 1.3927 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3927 - val_loss: 1.4071 - val_out_a_loss: 0.0054 - val_out_b_loss: 1.4018 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4018\n",
      "Epoch 109/200\n",
      "1000/1000 [==============================] - 0s 107us/step - loss: 1.4152 - out_a_loss: 0.0181 - out_b_loss: 1.3971 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3971 - val_loss: 1.4161 - val_out_a_loss: 0.0091 - val_out_b_loss: 1.4071 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4071\n",
      "Epoch 110/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.4023 - out_a_loss: 0.0074 - out_b_loss: 1.3949 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3949 - val_loss: 1.4016 - val_out_a_loss: 0.0065 - val_out_b_loss: 1.3951 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3951\n",
      "Epoch 111/200\n",
      "1000/1000 [==============================] - 0s 121us/step - loss: 1.3982 - out_a_loss: 0.0065 - out_b_loss: 1.3917 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3917 - val_loss: 1.4066 - val_out_a_loss: 0.0049 - val_out_b_loss: 1.4017 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4017\n",
      "Epoch 112/200\n",
      "1000/1000 [==============================] - 0s 108us/step - loss: 1.4257 - out_a_loss: 0.0253 - out_b_loss: 1.4004 - out_a_acc: 0.9980 - out_b_categorical_crossentropy: 1.4004 - val_loss: 1.4119 - val_out_a_loss: 0.0058 - val_out_b_loss: 1.4061 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4061\n",
      "Epoch 113/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.3993 - out_a_loss: 0.0053 - out_b_loss: 1.3940 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3940 - val_loss: 1.4096 - val_out_a_loss: 0.0099 - val_out_b_loss: 1.3997 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3997\n",
      "Epoch 114/200\n",
      "1000/1000 [==============================] - 0s 106us/step - loss: 1.3966 - out_a_loss: 0.0063 - out_b_loss: 1.3903 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3903 - val_loss: 1.4088 - val_out_a_loss: 0.0086 - val_out_b_loss: 1.4002 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.4002\n",
      "Epoch 115/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.4011 - out_a_loss: 0.0085 - out_b_loss: 1.3926 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3926 - val_loss: 1.4047 - val_out_a_loss: 0.0064 - val_out_b_loss: 1.3983 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3983\n",
      "Epoch 116/200\n",
      "1000/1000 [==============================] - 0s 110us/step - loss: 1.3978 - out_a_loss: 0.0069 - out_b_loss: 1.3909 - out_a_acc: 0.9990 - out_b_categorical_crossentropy: 1.3909 - val_loss: 1.4118 - val_out_a_loss: 0.0033 - val_out_b_loss: 1.4085 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.4085\n",
      "Epoch 117/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.4025 - out_a_loss: 0.0073 - out_b_loss: 1.3952 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3952 - val_loss: 1.3997 - val_out_a_loss: 0.0056 - val_out_b_loss: 1.3941 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3941\n",
      "Epoch 118/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.3943 - out_a_loss: 0.0053 - out_b_loss: 1.3890 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3890 - val_loss: 1.4190 - val_out_a_loss: 0.0020 - val_out_b_loss: 1.4171 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4171\n",
      "Epoch 119/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.4005 - out_a_loss: 0.0092 - out_b_loss: 1.3914 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3914 - val_loss: 1.4102 - val_out_a_loss: 0.0028 - val_out_b_loss: 1.4074 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4074\n",
      "Epoch 120/200\n",
      "1000/1000 [==============================] - 0s 112us/step - loss: 1.3978 - out_a_loss: 0.0043 - out_b_loss: 1.3935 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3935 - val_loss: 1.4005 - val_out_a_loss: 0.0092 - val_out_b_loss: 1.3913 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3913\n",
      "Epoch 121/200\n",
      "1000/1000 [==============================] - 0s 114us/step - loss: 1.3949 - out_a_loss: 0.0052 - out_b_loss: 1.3898 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3898 - val_loss: 1.4041 - val_out_a_loss: 0.0045 - val_out_b_loss: 1.3996 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3996\n",
      "Epoch 122/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.3953 - out_a_loss: 0.0058 - out_b_loss: 1.3895 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3895 - val_loss: 1.4010 - val_out_a_loss: 0.0064 - val_out_b_loss: 1.3946 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3946\n",
      "Epoch 123/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.3950 - out_a_loss: 0.0063 - out_b_loss: 1.3886 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3886 - val_loss: 1.4025 - val_out_a_loss: 0.0052 - val_out_b_loss: 1.3973 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3973\n",
      "Epoch 124/200\n",
      "1000/1000 [==============================] - 0s 106us/step - loss: 1.4009 - out_a_loss: 0.0089 - out_b_loss: 1.3920 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3920 - val_loss: 1.4059 - val_out_a_loss: 0.0036 - val_out_b_loss: 1.4023 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4023\n",
      "Epoch 125/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.4039 - out_a_loss: 0.0082 - out_b_loss: 1.3957 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3957 - val_loss: 1.4039 - val_out_a_loss: 0.0045 - val_out_b_loss: 1.3994 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3994\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 126/200\n",
      "1000/1000 [==============================] - 0s 111us/step - loss: 1.3962 - out_a_loss: 0.0052 - out_b_loss: 1.3910 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3910 - val_loss: 1.4012 - val_out_a_loss: 0.0086 - val_out_b_loss: 1.3926 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3926\n",
      "Epoch 127/200\n",
      "1000/1000 [==============================] - 0s 113us/step - loss: 1.4020 - out_a_loss: 0.0082 - out_b_loss: 1.3938 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3938 - val_loss: 1.4015 - val_out_a_loss: 0.0060 - val_out_b_loss: 1.3954 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3954\n",
      "Epoch 128/200\n",
      "1000/1000 [==============================] - 0s 137us/step - loss: 1.4016 - out_a_loss: 0.0075 - out_b_loss: 1.3941 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3941 - val_loss: 1.4175 - val_out_a_loss: 0.0251 - val_out_b_loss: 1.3923 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3923\n",
      "Epoch 129/200\n",
      "1000/1000 [==============================] - 0s 126us/step - loss: 1.4008 - out_a_loss: 0.0071 - out_b_loss: 1.3937 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3937 - val_loss: 1.4156 - val_out_a_loss: 0.0240 - val_out_b_loss: 1.3916 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3916\n",
      "Epoch 130/200\n",
      "1000/1000 [==============================] - 0s 109us/step - loss: 1.4001 - out_a_loss: 0.0099 - out_b_loss: 1.3903 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3903 - val_loss: 1.4032 - val_out_a_loss: 0.0029 - val_out_b_loss: 1.4003 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4003\n",
      "Epoch 131/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.3987 - out_a_loss: 0.0064 - out_b_loss: 1.3923 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3923 - val_loss: 1.4084 - val_out_a_loss: 0.0095 - val_out_b_loss: 1.3989 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3989\n",
      "Epoch 132/200\n",
      "1000/1000 [==============================] - 0s 102us/step - loss: 1.3973 - out_a_loss: 0.0057 - out_b_loss: 1.3916 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3916 - val_loss: 1.4051 - val_out_a_loss: 0.0132 - val_out_b_loss: 1.3919 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3919\n",
      "Epoch 133/200\n",
      "1000/1000 [==============================] - 0s 113us/step - loss: 1.3968 - out_a_loss: 0.0067 - out_b_loss: 1.3901 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3901 - val_loss: 1.4017 - val_out_a_loss: 0.0045 - val_out_b_loss: 1.3972 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3972\n",
      "Epoch 134/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.3944 - out_a_loss: 0.0057 - out_b_loss: 1.3887 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3887 - val_loss: 1.4017 - val_out_a_loss: 0.0076 - val_out_b_loss: 1.3941 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3941\n",
      "Epoch 135/200\n",
      "1000/1000 [==============================] - 0s 109us/step - loss: 1.3979 - out_a_loss: 0.0073 - out_b_loss: 1.3906 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3906 - val_loss: 1.4026 - val_out_a_loss: 0.0090 - val_out_b_loss: 1.3936 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3936\n",
      "Epoch 136/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.4063 - out_a_loss: 0.0130 - out_b_loss: 1.3932 - out_a_acc: 0.9980 - out_b_categorical_crossentropy: 1.3932 - val_loss: 1.4116 - val_out_a_loss: 0.0037 - val_out_b_loss: 1.4079 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4079\n",
      "Epoch 137/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.3971 - out_a_loss: 0.0046 - out_b_loss: 1.3926 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3926 - val_loss: 1.4013 - val_out_a_loss: 0.0072 - val_out_b_loss: 1.3941 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3941\n",
      "Epoch 138/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.4028 - out_a_loss: 0.0112 - out_b_loss: 1.3916 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3916 - val_loss: 1.4042 - val_out_a_loss: 0.0086 - val_out_b_loss: 1.3956 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3956\n",
      "Epoch 139/200\n",
      "1000/1000 [==============================] - 0s 134us/step - loss: 1.3964 - out_a_loss: 0.0060 - out_b_loss: 1.3904 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3904 - val_loss: 1.4104 - val_out_a_loss: 0.0068 - val_out_b_loss: 1.4037 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.4037\n",
      "Epoch 140/200\n",
      "1000/1000 [==============================] - 0s 123us/step - loss: 1.3962 - out_a_loss: 0.0052 - out_b_loss: 1.3910 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3910 - val_loss: 1.4078 - val_out_a_loss: 0.0159 - val_out_b_loss: 1.3919 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3919\n",
      "Epoch 141/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.4001 - out_a_loss: 0.0062 - out_b_loss: 1.3939 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3939 - val_loss: 1.4022 - val_out_a_loss: 0.0105 - val_out_b_loss: 1.3916 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3916\n",
      "Epoch 142/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.3974 - out_a_loss: 0.0076 - out_b_loss: 1.3897 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3897 - val_loss: 1.4022 - val_out_a_loss: 0.0068 - val_out_b_loss: 1.3954 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3954\n",
      "Epoch 143/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.3947 - out_a_loss: 0.0049 - out_b_loss: 1.3898 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3898 - val_loss: 1.4107 - val_out_a_loss: 0.0063 - val_out_b_loss: 1.4044 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.4044\n",
      "Epoch 144/200\n",
      "1000/1000 [==============================] - 0s 113us/step - loss: 1.3985 - out_a_loss: 0.0077 - out_b_loss: 1.3908 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3908 - val_loss: 1.4102 - val_out_a_loss: 0.0024 - val_out_b_loss: 1.4078 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4078\n",
      "Epoch 145/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.3977 - out_a_loss: 0.0061 - out_b_loss: 1.3917 - out_a_acc: 0.9990 - out_b_categorical_crossentropy: 1.3917 - val_loss: 1.4039 - val_out_a_loss: 0.0063 - val_out_b_loss: 1.3976 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3976\n",
      "Epoch 146/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.3980 - out_a_loss: 0.0049 - out_b_loss: 1.3931 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3931 - val_loss: 1.4062 - val_out_a_loss: 0.0141 - val_out_b_loss: 1.3921 - val_out_a_acc: 0.9970 - val_out_b_categorical_crossentropy: 1.3921\n",
      "Epoch 147/200\n",
      "1000/1000 [==============================] - 0s 112us/step - loss: 1.3994 - out_a_loss: 0.0086 - out_b_loss: 1.3908 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3908 - val_loss: 1.4077 - val_out_a_loss: 0.0060 - val_out_b_loss: 1.4017 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4017\n",
      "Epoch 148/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.4016 - out_a_loss: 0.0054 - out_b_loss: 1.3963 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3963 - val_loss: 1.4105 - val_out_a_loss: 0.0049 - val_out_b_loss: 1.4055 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4055\n",
      "Epoch 149/200\n",
      "1000/1000 [==============================] - 0s 111us/step - loss: 1.4043 - out_a_loss: 0.0053 - out_b_loss: 1.3991 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3991 - val_loss: 1.4065 - val_out_a_loss: 0.0049 - val_out_b_loss: 1.4016 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4016\n",
      "Epoch 150/200\n",
      "1000/1000 [==============================] - 0s 106us/step - loss: 1.3968 - out_a_loss: 0.0056 - out_b_loss: 1.3912 - out_a_acc: 0.9990 - out_b_categorical_crossentropy: 1.3912 - val_loss: 1.4075 - val_out_a_loss: 0.0148 - val_out_b_loss: 1.3927 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3927\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 151/200\n",
      "1000/1000 [==============================] - 0s 102us/step - loss: 1.4015 - out_a_loss: 0.0079 - out_b_loss: 1.3936 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3936 - val_loss: 1.4049 - val_out_a_loss: 0.0098 - val_out_b_loss: 1.3951 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3951\n",
      "Epoch 152/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.3982 - out_a_loss: 0.0077 - out_b_loss: 1.3905 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3905 - val_loss: 1.4080 - val_out_a_loss: 0.0068 - val_out_b_loss: 1.4012 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.4012\n",
      "Epoch 153/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.4075 - out_a_loss: 0.0106 - out_b_loss: 1.3968 - out_a_acc: 0.9980 - out_b_categorical_crossentropy: 1.3968 - val_loss: 1.4089 - val_out_a_loss: 0.0117 - val_out_b_loss: 1.3972 - val_out_a_acc: 0.9970 - val_out_b_categorical_crossentropy: 1.3972\n",
      "Epoch 154/200\n",
      "1000/1000 [==============================] - 0s 108us/step - loss: 1.4205 - out_a_loss: 0.0241 - out_b_loss: 1.3965 - out_a_acc: 0.9980 - out_b_categorical_crossentropy: 1.3965 - val_loss: 1.4067 - val_out_a_loss: 0.0049 - val_out_b_loss: 1.4018 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4018\n",
      "Epoch 155/200\n",
      "1000/1000 [==============================] - 0s 102us/step - loss: 1.3957 - out_a_loss: 0.0059 - out_b_loss: 1.3897 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3897 - val_loss: 1.4042 - val_out_a_loss: 0.0064 - val_out_b_loss: 1.3978 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3978\n",
      "Epoch 156/200\n",
      "1000/1000 [==============================] - 0s 113us/step - loss: 1.3991 - out_a_loss: 0.0089 - out_b_loss: 1.3902 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3902 - val_loss: 1.4025 - val_out_a_loss: 0.0067 - val_out_b_loss: 1.3957 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3957\n",
      "Epoch 157/200\n",
      "1000/1000 [==============================] - 0s 107us/step - loss: 1.3965 - out_a_loss: 0.0052 - out_b_loss: 1.3913 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3913 - val_loss: 1.4101 - val_out_a_loss: 0.0063 - val_out_b_loss: 1.4037 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4037\n",
      "Epoch 158/200\n",
      "1000/1000 [==============================] - 0s 109us/step - loss: 1.3958 - out_a_loss: 0.0045 - out_b_loss: 1.3913 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3913 - val_loss: 1.4128 - val_out_a_loss: 0.0059 - val_out_b_loss: 1.4068 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4068\n",
      "Epoch 159/200\n",
      "1000/1000 [==============================] - 0s 106us/step - loss: 1.3935 - out_a_loss: 0.0054 - out_b_loss: 1.3881 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3881 - val_loss: 1.4025 - val_out_a_loss: 0.0040 - val_out_b_loss: 1.3985 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3985\n",
      "Epoch 160/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.3943 - out_a_loss: 0.0054 - out_b_loss: 1.3889 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3889 - val_loss: 1.4055 - val_out_a_loss: 0.0032 - val_out_b_loss: 1.4023 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4023\n",
      "Epoch 161/200\n",
      "1000/1000 [==============================] - 0s 106us/step - loss: 1.3996 - out_a_loss: 0.0086 - out_b_loss: 1.3910 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3910 - val_loss: 1.4074 - val_out_a_loss: 0.0034 - val_out_b_loss: 1.4040 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4040\n",
      "Epoch 162/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.3956 - out_a_loss: 0.0044 - out_b_loss: 1.3912 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3912 - val_loss: 1.4022 - val_out_a_loss: 0.0070 - val_out_b_loss: 1.3951 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3951\n",
      "Epoch 163/200\n",
      "1000/1000 [==============================] - 0s 110us/step - loss: 1.3930 - out_a_loss: 0.0051 - out_b_loss: 1.3879 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3879 - val_loss: 1.4087 - val_out_a_loss: 0.0032 - val_out_b_loss: 1.4055 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.4055\n",
      "Epoch 164/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.3939 - out_a_loss: 0.0049 - out_b_loss: 1.3890 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3890 - val_loss: 1.4019 - val_out_a_loss: 0.0042 - val_out_b_loss: 1.3977 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3977\n",
      "Epoch 165/200\n",
      "1000/1000 [==============================] - 0s 108us/step - loss: 1.3933 - out_a_loss: 0.0045 - out_b_loss: 1.3888 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3888 - val_loss: 1.4065 - val_out_a_loss: 0.0033 - val_out_b_loss: 1.4031 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4031\n",
      "Epoch 166/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.3939 - out_a_loss: 0.0038 - out_b_loss: 1.3900 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3900 - val_loss: 1.4039 - val_out_a_loss: 0.0057 - val_out_b_loss: 1.3983 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3983\n",
      "Epoch 167/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.3947 - out_a_loss: 0.0045 - out_b_loss: 1.3901 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3901 - val_loss: 1.4192 - val_out_a_loss: 0.0042 - val_out_b_loss: 1.4150 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4150\n",
      "Epoch 168/200\n",
      "1000/1000 [==============================] - 0s 110us/step - loss: 1.3965 - out_a_loss: 0.0067 - out_b_loss: 1.3897 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3897 - val_loss: 1.4044 - val_out_a_loss: 0.0126 - val_out_b_loss: 1.3917 - val_out_a_acc: 0.9970 - val_out_b_categorical_crossentropy: 1.3917\n",
      "Epoch 169/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.3968 - out_a_loss: 0.0067 - out_b_loss: 1.3900 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3900 - val_loss: 1.4014 - val_out_a_loss: 0.0044 - val_out_b_loss: 1.3970 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3970\n",
      "Epoch 170/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.3953 - out_a_loss: 0.0045 - out_b_loss: 1.3908 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3908 - val_loss: 1.4066 - val_out_a_loss: 0.0025 - val_out_b_loss: 1.4040 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4040\n",
      "Epoch 171/200\n",
      "1000/1000 [==============================] - 0s 108us/step - loss: 1.3948 - out_a_loss: 0.0039 - out_b_loss: 1.3910 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3910 - val_loss: 1.4048 - val_out_a_loss: 0.0051 - val_out_b_loss: 1.3997 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3997\n",
      "Epoch 172/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.3934 - out_a_loss: 0.0045 - out_b_loss: 1.3889 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3889 - val_loss: 1.3990 - val_out_a_loss: 0.0055 - val_out_b_loss: 1.3934 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3934\n",
      "Epoch 173/200\n",
      "1000/1000 [==============================] - 0s 112us/step - loss: 1.3941 - out_a_loss: 0.0050 - out_b_loss: 1.3891 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3891 - val_loss: 1.4015 - val_out_a_loss: 0.0050 - val_out_b_loss: 1.3965 - val_out_a_acc: 0.9970 - val_out_b_categorical_crossentropy: 1.3965\n",
      "Epoch 174/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.3926 - out_a_loss: 0.0036 - out_b_loss: 1.3890 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3890 - val_loss: 1.4060 - val_out_a_loss: 0.0102 - val_out_b_loss: 1.3958 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3958\n",
      "Epoch 175/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.3995 - out_a_loss: 0.0056 - out_b_loss: 1.3939 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3939 - val_loss: 1.4042 - val_out_a_loss: 0.0105 - val_out_b_loss: 1.3937 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3937\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 176/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.4010 - out_a_loss: 0.0100 - out_b_loss: 1.3910 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3910 - val_loss: 1.4063 - val_out_a_loss: 0.0049 - val_out_b_loss: 1.4014 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4014\n",
      "Epoch 177/200\n",
      "1000/1000 [==============================] - 0s 103us/step - loss: 1.3954 - out_a_loss: 0.0051 - out_b_loss: 1.3903 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3903 - val_loss: 1.4017 - val_out_a_loss: 0.0085 - val_out_b_loss: 1.3932 - val_out_a_acc: 0.9970 - val_out_b_categorical_crossentropy: 1.3932\n",
      "Epoch 178/200\n",
      "1000/1000 [==============================] - 0s 112us/step - loss: 1.3950 - out_a_loss: 0.0055 - out_b_loss: 1.3896 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3896 - val_loss: 1.4007 - val_out_a_loss: 0.0047 - val_out_b_loss: 1.3960 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3960\n",
      "Epoch 179/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.3960 - out_a_loss: 0.0056 - out_b_loss: 1.3903 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3903 - val_loss: 1.4012 - val_out_a_loss: 0.0055 - val_out_b_loss: 1.3957 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3957\n",
      "Epoch 180/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.3939 - out_a_loss: 0.0037 - out_b_loss: 1.3903 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3903 - val_loss: 1.4014 - val_out_a_loss: 0.0061 - val_out_b_loss: 1.3953 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3953\n",
      "Epoch 181/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.3952 - out_a_loss: 0.0045 - out_b_loss: 1.3907 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3907 - val_loss: 1.4053 - val_out_a_loss: 0.0035 - val_out_b_loss: 1.4018 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4018\n",
      "Epoch 182/200\n",
      "1000/1000 [==============================] - 0s 107us/step - loss: 1.3948 - out_a_loss: 0.0049 - out_b_loss: 1.3899 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3899 - val_loss: 1.4014 - val_out_a_loss: 0.0043 - val_out_b_loss: 1.3971 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.3971\n",
      "Epoch 183/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.3946 - out_a_loss: 0.0059 - out_b_loss: 1.3887 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3887 - val_loss: 1.4030 - val_out_a_loss: 0.0037 - val_out_b_loss: 1.3992 - val_out_a_acc: 0.9980 - val_out_b_categorical_crossentropy: 1.3992\n",
      "Epoch 184/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.3992 - out_a_loss: 0.0073 - out_b_loss: 1.3918 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3918 - val_loss: 1.4120 - val_out_a_loss: 0.0146 - val_out_b_loss: 1.3975 - val_out_a_acc: 0.9980 - val_out_b_categorical_crossentropy: 1.3975\n",
      "Epoch 185/200\n",
      "1000/1000 [==============================] - 0s 106us/step - loss: 1.3985 - out_a_loss: 0.0069 - out_b_loss: 1.3916 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3916 - val_loss: 1.4119 - val_out_a_loss: 0.0031 - val_out_b_loss: 1.4088 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4088\n",
      "Epoch 186/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.3977 - out_a_loss: 0.0059 - out_b_loss: 1.3918 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3918 - val_loss: 1.4050 - val_out_a_loss: 0.0061 - val_out_b_loss: 1.3988 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3988\n",
      "Epoch 187/200\n",
      "1000/1000 [==============================] - 0s 107us/step - loss: 1.3979 - out_a_loss: 0.0063 - out_b_loss: 1.3916 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3916 - val_loss: 1.4218 - val_out_a_loss: 0.0269 - val_out_b_loss: 1.3949 - val_out_a_acc: 0.9890 - val_out_b_categorical_crossentropy: 1.3949\n",
      "Epoch 188/200\n",
      "1000/1000 [==============================] - 0s 102us/step - loss: 1.3967 - out_a_loss: 0.0058 - out_b_loss: 1.3909 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3909 - val_loss: 1.4088 - val_out_a_loss: 0.0036 - val_out_b_loss: 1.4053 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4053\n",
      "Epoch 189/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.3966 - out_a_loss: 0.0058 - out_b_loss: 1.3908 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3908 - val_loss: 1.4014 - val_out_a_loss: 0.0075 - val_out_b_loss: 1.3940 - val_out_a_acc: 0.9980 - val_out_b_categorical_crossentropy: 1.3940\n",
      "Epoch 190/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.3949 - out_a_loss: 0.0045 - out_b_loss: 1.3904 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3904 - val_loss: 1.4021 - val_out_a_loss: 0.0080 - val_out_b_loss: 1.3941 - val_out_a_acc: 0.9950 - val_out_b_categorical_crossentropy: 1.3941\n",
      "Epoch 191/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.3934 - out_a_loss: 0.0052 - out_b_loss: 1.3882 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3882 - val_loss: 1.4059 - val_out_a_loss: 0.0046 - val_out_b_loss: 1.4013 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4013\n",
      "Epoch 192/200\n",
      "1000/1000 [==============================] - 0s 111us/step - loss: 1.3951 - out_a_loss: 0.0050 - out_b_loss: 1.3901 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3901 - val_loss: 1.4016 - val_out_a_loss: 0.0061 - val_out_b_loss: 1.3955 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3955\n",
      "Epoch 193/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.3942 - out_a_loss: 0.0054 - out_b_loss: 1.3888 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3888 - val_loss: 1.4025 - val_out_a_loss: 0.0077 - val_out_b_loss: 1.3948 - val_out_a_acc: 0.9980 - val_out_b_categorical_crossentropy: 1.3948\n",
      "Epoch 194/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.3999 - out_a_loss: 0.0076 - out_b_loss: 1.3923 - out_a_acc: 0.9990 - out_b_categorical_crossentropy: 1.3923 - val_loss: 1.4104 - val_out_a_loss: 0.0110 - val_out_b_loss: 1.3994 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3994\n",
      "Epoch 195/200\n",
      "1000/1000 [==============================] - 0s 105us/step - loss: 1.4000 - out_a_loss: 0.0076 - out_b_loss: 1.3924 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3924 - val_loss: 1.4041 - val_out_a_loss: 0.0029 - val_out_b_loss: 1.4013 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4013\n",
      "Epoch 196/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.3974 - out_a_loss: 0.0056 - out_b_loss: 1.3918 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3918 - val_loss: 1.4123 - val_out_a_loss: 0.0035 - val_out_b_loss: 1.4088 - val_out_a_acc: 0.9990 - val_out_b_categorical_crossentropy: 1.4088\n",
      "Epoch 197/200\n",
      "1000/1000 [==============================] - 0s 112us/step - loss: 1.3972 - out_a_loss: 0.0047 - out_b_loss: 1.3926 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3926 - val_loss: 1.4055 - val_out_a_loss: 0.0097 - val_out_b_loss: 1.3958 - val_out_a_acc: 0.9980 - val_out_b_categorical_crossentropy: 1.3958\n",
      "Epoch 198/200\n",
      "1000/1000 [==============================] - 0s 104us/step - loss: 1.3961 - out_a_loss: 0.0061 - out_b_loss: 1.3900 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3900 - val_loss: 1.4030 - val_out_a_loss: 0.0040 - val_out_b_loss: 1.3990 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.3990\n",
      "Epoch 199/200\n",
      "1000/1000 [==============================] - 0s 107us/step - loss: 1.3923 - out_a_loss: 0.0039 - out_b_loss: 1.3884 - out_a_acc: 1.0000 - out_b_categorical_crossentropy: 1.3884 - val_loss: 1.4164 - val_out_a_loss: 0.0027 - val_out_b_loss: 1.4137 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4137\n",
      "Epoch 200/200\n",
      "1000/1000 [==============================] - 0s 106us/step - loss: 1.3961 - out_a_loss: 0.0059 - out_b_loss: 1.3902 - out_a_acc: 0.9990 - out_b_categorical_crossentropy: 1.3902 - val_loss: 1.4272 - val_out_a_loss: 0.0020 - val_out_b_loss: 1.4252 - val_out_a_acc: 1.0000 - val_out_b_categorical_crossentropy: 1.4252\n"
     ]
    }
   ],
   "source": [
    "classifier = Classifier()\n",
    "\n",
    "classifier.train()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\n",
       "\n",
       "mpl.get_websocket_type = function() {\n",
       "    if (typeof(WebSocket) !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof(MozWebSocket) !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert('Your browser does not have WebSocket support.' +\n",
       "              'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "              'Firefox 4 and 5 are also supported but you ' +\n",
       "              'have to enable WebSockets in about:config.');\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = (this.ws.binaryType != undefined);\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById(\"mpl-warnings\");\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent = (\n",
       "                \"This browser does not support binary websocket messages. \" +\n",
       "                    \"Performance may be slow.\");\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = $('<div/>');\n",
       "    this._root_extra_style(this.root)\n",
       "    this.root.attr('style', 'display: inline-block');\n",
       "\n",
       "    $(parent_element).append(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen =  function () {\n",
       "            fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n",
       "            fig.send_message(\"send_image_mode\", {});\n",
       "            if (mpl.ratio != 1) {\n",
       "                fig.send_message(\"set_dpi_ratio\", {'dpi_ratio': mpl.ratio});\n",
       "            }\n",
       "            fig.send_message(\"refresh\", {});\n",
       "        }\n",
       "\n",
       "    this.imageObj.onload = function() {\n",
       "            if (fig.image_mode == 'full') {\n",
       "                // Full images could contain transparency (where diff images\n",
       "                // almost always do), so we need to clear the canvas so that\n",
       "                // there is no ghosting.\n",
       "                fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "            }\n",
       "            fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "        };\n",
       "\n",
       "    this.imageObj.onunload = function() {\n",
       "        fig.ws.close();\n",
       "    }\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_header = function() {\n",
       "    var titlebar = $(\n",
       "        '<div class=\"ui-dialog-titlebar ui-widget-header ui-corner-all ' +\n",
       "        'ui-helper-clearfix\"/>');\n",
       "    var titletext = $(\n",
       "        '<div class=\"ui-dialog-title\" style=\"width: 100%; ' +\n",
       "        'text-align: center; padding: 3px;\"/>');\n",
       "    titlebar.append(titletext)\n",
       "    this.root.append(titlebar);\n",
       "    this.header = titletext[0];\n",
       "}\n",
       "\n",
       "\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = $('<div/>');\n",
       "\n",
       "    canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n",
       "\n",
       "    function canvas_keyboard_event(event) {\n",
       "        return fig.key_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    canvas_div.keydown('key_press', canvas_keyboard_event);\n",
       "    canvas_div.keyup('key_release', canvas_keyboard_event);\n",
       "    this.canvas_div = canvas_div\n",
       "    this._canvas_extra_style(canvas_div)\n",
       "    this.root.append(canvas_div);\n",
       "\n",
       "    var canvas = $('<canvas/>');\n",
       "    canvas.addClass('mpl-canvas');\n",
       "    canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n",
       "\n",
       "    this.canvas = canvas[0];\n",
       "    this.context = canvas[0].getContext(\"2d\");\n",
       "\n",
       "    var backingStore = this.context.backingStorePixelRatio ||\n",
       "\tthis.context.webkitBackingStorePixelRatio ||\n",
       "\tthis.context.mozBackingStorePixelRatio ||\n",
       "\tthis.context.msBackingStorePixelRatio ||\n",
       "\tthis.context.oBackingStorePixelRatio ||\n",
       "\tthis.context.backingStorePixelRatio || 1;\n",
       "\n",
       "    mpl.ratio = (window.devicePixelRatio || 1) / backingStore;\n",
       "\n",
       "    var rubberband = $('<canvas/>');\n",
       "    rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n",
       "\n",
       "    var pass_mouse_events = true;\n",
       "\n",
       "    canvas_div.resizable({\n",
       "        start: function(event, ui) {\n",
       "            pass_mouse_events = false;\n",
       "        },\n",
       "        resize: function(event, ui) {\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "        stop: function(event, ui) {\n",
       "            pass_mouse_events = true;\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "    });\n",
       "\n",
       "    function mouse_event_fn(event) {\n",
       "        if (pass_mouse_events)\n",
       "            return fig.mouse_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    rubberband.mousedown('button_press', mouse_event_fn);\n",
       "    rubberband.mouseup('button_release', mouse_event_fn);\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband.mousemove('motion_notify', mouse_event_fn);\n",
       "\n",
       "    rubberband.mouseenter('figure_enter', mouse_event_fn);\n",
       "    rubberband.mouseleave('figure_leave', mouse_event_fn);\n",
       "\n",
       "    canvas_div.on(\"wheel\", function (event) {\n",
       "        event = event.originalEvent;\n",
       "        event['data'] = 'scroll'\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        mouse_event_fn(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.append(canvas);\n",
       "    canvas_div.append(rubberband);\n",
       "\n",
       "    this.rubberband = rubberband;\n",
       "    this.rubberband_canvas = rubberband[0];\n",
       "    this.rubberband_context = rubberband[0].getContext(\"2d\");\n",
       "    this.rubberband_context.strokeStyle = \"#000000\";\n",
       "\n",
       "    this._resize_canvas = function(width, height) {\n",
       "        // Keep the size of the canvas, canvas container, and rubber band\n",
       "        // canvas in synch.\n",
       "        canvas_div.css('width', width)\n",
       "        canvas_div.css('height', height)\n",
       "\n",
       "        canvas.attr('width', width * mpl.ratio);\n",
       "        canvas.attr('height', height * mpl.ratio);\n",
       "        canvas.attr('style', 'width: ' + width + 'px; height: ' + height + 'px;');\n",
       "\n",
       "        rubberband.attr('width', width);\n",
       "        rubberband.attr('height', height);\n",
       "    }\n",
       "\n",
       "    // Set the figure to an initial 600x600px, this will subsequently be updated\n",
       "    // upon first draw.\n",
       "    this._resize_canvas(600, 600);\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus () {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            // put a spacer in here.\n",
       "            continue;\n",
       "        }\n",
       "        var button = $('<button/>');\n",
       "        button.addClass('ui-button ui-widget ui-state-default ui-corner-all ' +\n",
       "                        'ui-button-icon-only');\n",
       "        button.attr('role', 'button');\n",
       "        button.attr('aria-disabled', 'false');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "\n",
       "        var icon_img = $('<span/>');\n",
       "        icon_img.addClass('ui-button-icon-primary ui-icon');\n",
       "        icon_img.addClass(image);\n",
       "        icon_img.addClass('ui-corner-all');\n",
       "\n",
       "        var tooltip_span = $('<span/>');\n",
       "        tooltip_span.addClass('ui-button-text');\n",
       "        tooltip_span.html(tooltip);\n",
       "\n",
       "        button.append(icon_img);\n",
       "        button.append(tooltip_span);\n",
       "\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    var fmt_picker_span = $('<span/>');\n",
       "\n",
       "    var fmt_picker = $('<select/>');\n",
       "    fmt_picker.addClass('mpl-toolbar-option ui-widget ui-widget-content');\n",
       "    fmt_picker_span.append(fmt_picker);\n",
       "    nav_element.append(fmt_picker_span);\n",
       "    this.format_dropdown = fmt_picker[0];\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = $(\n",
       "            '<option/>', {selected: fmt === mpl.default_extension}).html(fmt);\n",
       "        fmt_picker.append(option)\n",
       "    }\n",
       "\n",
       "    // Add hover states to the ui-buttons\n",
       "    $( \".ui-button\" ).hover(\n",
       "        function() { $(this).addClass(\"ui-state-hover\");},\n",
       "        function() { $(this).removeClass(\"ui-state-hover\");}\n",
       "    );\n",
       "\n",
       "    var status_bar = $('<span class=\"mpl-message\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.request_resize = function(x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', {'width': x_pixels, 'height': y_pixels});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_message = function(type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function() {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({type: \"draw\", figure_id: this.id}));\n",
       "    }\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function(fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] != fig.canvas.width || size[1] != fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1]);\n",
       "        fig.send_message(\"refresh\", {});\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function(fig, msg) {\n",
       "    var x0 = msg['x0'] / mpl.ratio;\n",
       "    var y0 = (fig.canvas.height - msg['y0']) / mpl.ratio;\n",
       "    var x1 = msg['x1'] / mpl.ratio;\n",
       "    var y1 = (fig.canvas.height - msg['y1']) / mpl.ratio;\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0, 0, fig.canvas.width, fig.canvas.height);\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function(fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function(fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch(cursor)\n",
       "    {\n",
       "    case 0:\n",
       "        cursor = 'pointer';\n",
       "        break;\n",
       "    case 1:\n",
       "        cursor = 'default';\n",
       "        break;\n",
       "    case 2:\n",
       "        cursor = 'crosshair';\n",
       "        break;\n",
       "    case 3:\n",
       "        cursor = 'move';\n",
       "        break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_message = function(fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function(fig, msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function(fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message(\"ack\", {});\n",
       "}\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function(fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            /* FIXME: We get \"Resource interpreted as Image but\n",
       "             * transferred with MIME type text/plain:\" errors on\n",
       "             * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "             * to be part of the websocket stream */\n",
       "            evt.data.type = \"image/png\";\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src);\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                evt.data);\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "        else if (typeof evt.data === 'string' && evt.data.slice(0, 21) == \"data:image/png;base64\") {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig[\"handle_\" + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\"No handler for the '\" + msg_type + \"' message type: \", msg);\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\"Exception inside the 'handler_\" + msg_type + \"' callback:\", e, e.stack, msg);\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "}\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function(e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e)\n",
       "        e = window.event;\n",
       "    if (e.target)\n",
       "        targ = e.target;\n",
       "    else if (e.srcElement)\n",
       "        targ = e.srcElement;\n",
       "    if (targ.nodeType == 3) // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "\n",
       "    // jQuery normalizes the pageX and pageY\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    // offset() returns the position of the element relative to the document\n",
       "    var x = e.pageX - $(targ).offset().left;\n",
       "    var y = e.pageY - $(targ).offset().top;\n",
       "\n",
       "    return {\"x\": x, \"y\": y};\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys (original) {\n",
       "  return Object.keys(original).reduce(function (obj, key) {\n",
       "    if (typeof original[key] !== 'object')\n",
       "        obj[key] = original[key]\n",
       "    return obj;\n",
       "  }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function(event, name) {\n",
       "    var canvas_pos = mpl.findpos(event)\n",
       "\n",
       "    if (name === 'button_press')\n",
       "    {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x * mpl.ratio;\n",
       "    var y = canvas_pos.y * mpl.ratio;\n",
       "\n",
       "    this.send_message(name, {x: x, y: y, button: event.button,\n",
       "                             step: event.step,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.key_event = function(event, name) {\n",
       "\n",
       "    // Prevent repeat events\n",
       "    if (name == 'key_press')\n",
       "    {\n",
       "        if (event.which === this._key)\n",
       "            return;\n",
       "        else\n",
       "            this._key = event.which;\n",
       "    }\n",
       "    if (name == 'key_release')\n",
       "        this._key = null;\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.which != 17)\n",
       "        value += \"ctrl+\";\n",
       "    if (event.altKey && event.which != 18)\n",
       "        value += \"alt+\";\n",
       "    if (event.shiftKey && event.which != 16)\n",
       "        value += \"shift+\";\n",
       "\n",
       "    value += 'k';\n",
       "    value += event.which.toString();\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, {key: value,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function(name) {\n",
       "    if (name == 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message(\"toolbar_button\", {name: name});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function(tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to  previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Pan axes with left mouse, zoom with right\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"jpeg\", \"pdf\", \"png\", \"ps\", \"raw\", \"svg\", \"tif\"];\n",
       "\n",
       "mpl.default_extension = \"png\";var comm_websocket_adapter = function(comm) {\n",
       "    // Create a \"websocket\"-like object which calls the given IPython comm\n",
       "    // object with the appropriate methods. Currently this is a non binary\n",
       "    // socket, so there is still some room for performance tuning.\n",
       "    var ws = {};\n",
       "\n",
       "    ws.close = function() {\n",
       "        comm.close()\n",
       "    };\n",
       "    ws.send = function(m) {\n",
       "        //console.log('sending', m);\n",
       "        comm.send(m);\n",
       "    };\n",
       "    // Register the callback with on_msg.\n",
       "    comm.on_msg(function(msg) {\n",
       "        //console.log('receiving', msg['content']['data'], msg);\n",
       "        // Pass the mpl event to the overridden (by mpl) onmessage function.\n",
       "        ws.onmessage(msg['content']['data'])\n",
       "    });\n",
       "    return ws;\n",
       "}\n",
       "\n",
       "mpl.mpl_figure_comm = function(comm, msg) {\n",
       "    // This is the function which gets called when the mpl process\n",
       "    // starts-up an IPython Comm through the \"matplotlib\" channel.\n",
       "\n",
       "    var id = msg.content.data.id;\n",
       "    // Get hold of the div created by the display call when the Comm\n",
       "    // socket was opened in Python.\n",
       "    var element = $(\"#\" + id);\n",
       "    var ws_proxy = comm_websocket_adapter(comm)\n",
       "\n",
       "    function ondownload(figure, format) {\n",
       "        window.open(figure.imageObj.src);\n",
       "    }\n",
       "\n",
       "    var fig = new mpl.figure(id, ws_proxy,\n",
       "                           ondownload,\n",
       "                           element.get(0));\n",
       "\n",
       "    // Call onopen now - mpl needs it, as it is assuming we've passed it a real\n",
       "    // web socket which is closed, not our websocket->open comm proxy.\n",
       "    ws_proxy.onopen();\n",
       "\n",
       "    fig.parent_element = element.get(0);\n",
       "    fig.cell_info = mpl.find_output_cell(\"<div id='\" + id + \"'></div>\");\n",
       "    if (!fig.cell_info) {\n",
       "        console.error(\"Failed to find cell for figure\", id, fig);\n",
       "        return;\n",
       "    }\n",
       "\n",
       "    var output_index = fig.cell_info[2]\n",
       "    var cell = fig.cell_info[0];\n",
       "\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_close = function(fig, msg) {\n",
       "    var width = fig.canvas.width/mpl.ratio\n",
       "    fig.root.unbind('remove')\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable()\n",
       "    $(fig.parent_element).html('<img src=\"' + dataURL + '\" width=\"' + width + '\">');\n",
       "    fig.close_ws(fig, msg);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.close_ws = function(fig, msg){\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function(remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var width = this.canvas.width/mpl.ratio\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\" width=\"' + width + '\">';\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message(\"ack\", {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () { fig.push_to_output() }, 1000);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items){\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) { continue; };\n",
       "\n",
       "        var button = $('<button class=\"btn btn-default\" href=\"#\" title=\"' + name + '\"><i class=\"fa ' + image + ' fa-lg\"></i></button>');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = $('<span class=\"mpl-message\" style=\"text-align:right; float: right;\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = $('<div class=\"btn-group inline pull-right\"></div>');\n",
       "    var button = $('<button class=\"btn btn-mini btn-primary\" href=\"#\" title=\"Stop Interaction\"><i class=\"fa fa-power-off icon-remove icon-large\"></i></button>');\n",
       "    button.click(function (evt) { fig.handle_close(fig, {}); } );\n",
       "    button.mouseover('Stop Interaction', toolbar_mouse_event);\n",
       "    buttongrp.append(button);\n",
       "    var titlebar = this.root.find($('.ui-dialog-titlebar'));\n",
       "    titlebar.prepend(buttongrp);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(el){\n",
       "    var fig = this\n",
       "    el.on(\"remove\", function(){\n",
       "\tfig.close_ws(fig, {});\n",
       "    });\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(el){\n",
       "    // this is important to make the div 'focusable\n",
       "    el.attr('tabindex', 0)\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    }\n",
       "    else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager)\n",
       "        manager = IPython.keyboard_manager;\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which == 13) {\n",
       "        this.canvas_div.blur();\n",
       "        event.shiftKey = false;\n",
       "        // Send a \"J\" for go to next cell\n",
       "        event.which = 74;\n",
       "        event.keyCode = 74;\n",
       "        manager.command_mode();\n",
       "        manager.handle_keydown(event);\n",
       "    }\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.find_output_cell = function(html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i=0; i<ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code'){\n",
       "            for (var j=0; j<cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] == html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "}\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel != null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\" width=\"638.888905813665\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x7f99439b47d0>"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Plot input data\n",
    "\n",
    "data = np.concatenate((classifier.data, classifier.data_out), axis=0)\n",
    "labels = np.concatenate((classifier.labels, classifier.labels_out), axis=0)\n",
    "fig = plt.figure()\n",
    "plt.scatter(data[:, 0], data[:, 1], s=4, c=labels.squeeze())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\n",
       "\n",
       "mpl.get_websocket_type = function() {\n",
       "    if (typeof(WebSocket) !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof(MozWebSocket) !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert('Your browser does not have WebSocket support.' +\n",
       "              'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "              'Firefox 4 and 5 are also supported but you ' +\n",
       "              'have to enable WebSockets in about:config.');\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = (this.ws.binaryType != undefined);\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById(\"mpl-warnings\");\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent = (\n",
       "                \"This browser does not support binary websocket messages. \" +\n",
       "                    \"Performance may be slow.\");\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = $('<div/>');\n",
       "    this._root_extra_style(this.root)\n",
       "    this.root.attr('style', 'display: inline-block');\n",
       "\n",
       "    $(parent_element).append(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen =  function () {\n",
       "            fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n",
       "            fig.send_message(\"send_image_mode\", {});\n",
       "            if (mpl.ratio != 1) {\n",
       "                fig.send_message(\"set_dpi_ratio\", {'dpi_ratio': mpl.ratio});\n",
       "            }\n",
       "            fig.send_message(\"refresh\", {});\n",
       "        }\n",
       "\n",
       "    this.imageObj.onload = function() {\n",
       "            if (fig.image_mode == 'full') {\n",
       "                // Full images could contain transparency (where diff images\n",
       "                // almost always do), so we need to clear the canvas so that\n",
       "                // there is no ghosting.\n",
       "                fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "            }\n",
       "            fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "        };\n",
       "\n",
       "    this.imageObj.onunload = function() {\n",
       "        fig.ws.close();\n",
       "    }\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_header = function() {\n",
       "    var titlebar = $(\n",
       "        '<div class=\"ui-dialog-titlebar ui-widget-header ui-corner-all ' +\n",
       "        'ui-helper-clearfix\"/>');\n",
       "    var titletext = $(\n",
       "        '<div class=\"ui-dialog-title\" style=\"width: 100%; ' +\n",
       "        'text-align: center; padding: 3px;\"/>');\n",
       "    titlebar.append(titletext)\n",
       "    this.root.append(titlebar);\n",
       "    this.header = titletext[0];\n",
       "}\n",
       "\n",
       "\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = $('<div/>');\n",
       "\n",
       "    canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n",
       "\n",
       "    function canvas_keyboard_event(event) {\n",
       "        return fig.key_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    canvas_div.keydown('key_press', canvas_keyboard_event);\n",
       "    canvas_div.keyup('key_release', canvas_keyboard_event);\n",
       "    this.canvas_div = canvas_div\n",
       "    this._canvas_extra_style(canvas_div)\n",
       "    this.root.append(canvas_div);\n",
       "\n",
       "    var canvas = $('<canvas/>');\n",
       "    canvas.addClass('mpl-canvas');\n",
       "    canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n",
       "\n",
       "    this.canvas = canvas[0];\n",
       "    this.context = canvas[0].getContext(\"2d\");\n",
       "\n",
       "    var backingStore = this.context.backingStorePixelRatio ||\n",
       "\tthis.context.webkitBackingStorePixelRatio ||\n",
       "\tthis.context.mozBackingStorePixelRatio ||\n",
       "\tthis.context.msBackingStorePixelRatio ||\n",
       "\tthis.context.oBackingStorePixelRatio ||\n",
       "\tthis.context.backingStorePixelRatio || 1;\n",
       "\n",
       "    mpl.ratio = (window.devicePixelRatio || 1) / backingStore;\n",
       "\n",
       "    var rubberband = $('<canvas/>');\n",
       "    rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n",
       "\n",
       "    var pass_mouse_events = true;\n",
       "\n",
       "    canvas_div.resizable({\n",
       "        start: function(event, ui) {\n",
       "            pass_mouse_events = false;\n",
       "        },\n",
       "        resize: function(event, ui) {\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "        stop: function(event, ui) {\n",
       "            pass_mouse_events = true;\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "    });\n",
       "\n",
       "    function mouse_event_fn(event) {\n",
       "        if (pass_mouse_events)\n",
       "            return fig.mouse_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    rubberband.mousedown('button_press', mouse_event_fn);\n",
       "    rubberband.mouseup('button_release', mouse_event_fn);\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband.mousemove('motion_notify', mouse_event_fn);\n",
       "\n",
       "    rubberband.mouseenter('figure_enter', mouse_event_fn);\n",
       "    rubberband.mouseleave('figure_leave', mouse_event_fn);\n",
       "\n",
       "    canvas_div.on(\"wheel\", function (event) {\n",
       "        event = event.originalEvent;\n",
       "        event['data'] = 'scroll'\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        mouse_event_fn(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.append(canvas);\n",
       "    canvas_div.append(rubberband);\n",
       "\n",
       "    this.rubberband = rubberband;\n",
       "    this.rubberband_canvas = rubberband[0];\n",
       "    this.rubberband_context = rubberband[0].getContext(\"2d\");\n",
       "    this.rubberband_context.strokeStyle = \"#000000\";\n",
       "\n",
       "    this._resize_canvas = function(width, height) {\n",
       "        // Keep the size of the canvas, canvas container, and rubber band\n",
       "        // canvas in synch.\n",
       "        canvas_div.css('width', width)\n",
       "        canvas_div.css('height', height)\n",
       "\n",
       "        canvas.attr('width', width * mpl.ratio);\n",
       "        canvas.attr('height', height * mpl.ratio);\n",
       "        canvas.attr('style', 'width: ' + width + 'px; height: ' + height + 'px;');\n",
       "\n",
       "        rubberband.attr('width', width);\n",
       "        rubberband.attr('height', height);\n",
       "    }\n",
       "\n",
       "    // Set the figure to an initial 600x600px, this will subsequently be updated\n",
       "    // upon first draw.\n",
       "    this._resize_canvas(600, 600);\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus () {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            // put a spacer in here.\n",
       "            continue;\n",
       "        }\n",
       "        var button = $('<button/>');\n",
       "        button.addClass('ui-button ui-widget ui-state-default ui-corner-all ' +\n",
       "                        'ui-button-icon-only');\n",
       "        button.attr('role', 'button');\n",
       "        button.attr('aria-disabled', 'false');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "\n",
       "        var icon_img = $('<span/>');\n",
       "        icon_img.addClass('ui-button-icon-primary ui-icon');\n",
       "        icon_img.addClass(image);\n",
       "        icon_img.addClass('ui-corner-all');\n",
       "\n",
       "        var tooltip_span = $('<span/>');\n",
       "        tooltip_span.addClass('ui-button-text');\n",
       "        tooltip_span.html(tooltip);\n",
       "\n",
       "        button.append(icon_img);\n",
       "        button.append(tooltip_span);\n",
       "\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    var fmt_picker_span = $('<span/>');\n",
       "\n",
       "    var fmt_picker = $('<select/>');\n",
       "    fmt_picker.addClass('mpl-toolbar-option ui-widget ui-widget-content');\n",
       "    fmt_picker_span.append(fmt_picker);\n",
       "    nav_element.append(fmt_picker_span);\n",
       "    this.format_dropdown = fmt_picker[0];\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = $(\n",
       "            '<option/>', {selected: fmt === mpl.default_extension}).html(fmt);\n",
       "        fmt_picker.append(option)\n",
       "    }\n",
       "\n",
       "    // Add hover states to the ui-buttons\n",
       "    $( \".ui-button\" ).hover(\n",
       "        function() { $(this).addClass(\"ui-state-hover\");},\n",
       "        function() { $(this).removeClass(\"ui-state-hover\");}\n",
       "    );\n",
       "\n",
       "    var status_bar = $('<span class=\"mpl-message\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.request_resize = function(x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', {'width': x_pixels, 'height': y_pixels});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_message = function(type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function() {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({type: \"draw\", figure_id: this.id}));\n",
       "    }\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function(fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] != fig.canvas.width || size[1] != fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1]);\n",
       "        fig.send_message(\"refresh\", {});\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function(fig, msg) {\n",
       "    var x0 = msg['x0'] / mpl.ratio;\n",
       "    var y0 = (fig.canvas.height - msg['y0']) / mpl.ratio;\n",
       "    var x1 = msg['x1'] / mpl.ratio;\n",
       "    var y1 = (fig.canvas.height - msg['y1']) / mpl.ratio;\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0, 0, fig.canvas.width, fig.canvas.height);\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function(fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function(fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch(cursor)\n",
       "    {\n",
       "    case 0:\n",
       "        cursor = 'pointer';\n",
       "        break;\n",
       "    case 1:\n",
       "        cursor = 'default';\n",
       "        break;\n",
       "    case 2:\n",
       "        cursor = 'crosshair';\n",
       "        break;\n",
       "    case 3:\n",
       "        cursor = 'move';\n",
       "        break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_message = function(fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function(fig, msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function(fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message(\"ack\", {});\n",
       "}\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function(fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            /* FIXME: We get \"Resource interpreted as Image but\n",
       "             * transferred with MIME type text/plain:\" errors on\n",
       "             * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "             * to be part of the websocket stream */\n",
       "            evt.data.type = \"image/png\";\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src);\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                evt.data);\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "        else if (typeof evt.data === 'string' && evt.data.slice(0, 21) == \"data:image/png;base64\") {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig[\"handle_\" + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\"No handler for the '\" + msg_type + \"' message type: \", msg);\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\"Exception inside the 'handler_\" + msg_type + \"' callback:\", e, e.stack, msg);\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "}\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function(e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e)\n",
       "        e = window.event;\n",
       "    if (e.target)\n",
       "        targ = e.target;\n",
       "    else if (e.srcElement)\n",
       "        targ = e.srcElement;\n",
       "    if (targ.nodeType == 3) // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "\n",
       "    // jQuery normalizes the pageX and pageY\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    // offset() returns the position of the element relative to the document\n",
       "    var x = e.pageX - $(targ).offset().left;\n",
       "    var y = e.pageY - $(targ).offset().top;\n",
       "\n",
       "    return {\"x\": x, \"y\": y};\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys (original) {\n",
       "  return Object.keys(original).reduce(function (obj, key) {\n",
       "    if (typeof original[key] !== 'object')\n",
       "        obj[key] = original[key]\n",
       "    return obj;\n",
       "  }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function(event, name) {\n",
       "    var canvas_pos = mpl.findpos(event)\n",
       "\n",
       "    if (name === 'button_press')\n",
       "    {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x * mpl.ratio;\n",
       "    var y = canvas_pos.y * mpl.ratio;\n",
       "\n",
       "    this.send_message(name, {x: x, y: y, button: event.button,\n",
       "                             step: event.step,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.key_event = function(event, name) {\n",
       "\n",
       "    // Prevent repeat events\n",
       "    if (name == 'key_press')\n",
       "    {\n",
       "        if (event.which === this._key)\n",
       "            return;\n",
       "        else\n",
       "            this._key = event.which;\n",
       "    }\n",
       "    if (name == 'key_release')\n",
       "        this._key = null;\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.which != 17)\n",
       "        value += \"ctrl+\";\n",
       "    if (event.altKey && event.which != 18)\n",
       "        value += \"alt+\";\n",
       "    if (event.shiftKey && event.which != 16)\n",
       "        value += \"shift+\";\n",
       "\n",
       "    value += 'k';\n",
       "    value += event.which.toString();\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, {key: value,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function(name) {\n",
       "    if (name == 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message(\"toolbar_button\", {name: name});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function(tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to  previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Pan axes with left mouse, zoom with right\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"jpeg\", \"pdf\", \"png\", \"ps\", \"raw\", \"svg\", \"tif\"];\n",
       "\n",
       "mpl.default_extension = \"png\";var comm_websocket_adapter = function(comm) {\n",
       "    // Create a \"websocket\"-like object which calls the given IPython comm\n",
       "    // object with the appropriate methods. Currently this is a non binary\n",
       "    // socket, so there is still some room for performance tuning.\n",
       "    var ws = {};\n",
       "\n",
       "    ws.close = function() {\n",
       "        comm.close()\n",
       "    };\n",
       "    ws.send = function(m) {\n",
       "        //console.log('sending', m);\n",
       "        comm.send(m);\n",
       "    };\n",
       "    // Register the callback with on_msg.\n",
       "    comm.on_msg(function(msg) {\n",
       "        //console.log('receiving', msg['content']['data'], msg);\n",
       "        // Pass the mpl event to the overridden (by mpl) onmessage function.\n",
       "        ws.onmessage(msg['content']['data'])\n",
       "    });\n",
       "    return ws;\n",
       "}\n",
       "\n",
       "mpl.mpl_figure_comm = function(comm, msg) {\n",
       "    // This is the function which gets called when the mpl process\n",
       "    // starts-up an IPython Comm through the \"matplotlib\" channel.\n",
       "\n",
       "    var id = msg.content.data.id;\n",
       "    // Get hold of the div created by the display call when the Comm\n",
       "    // socket was opened in Python.\n",
       "    var element = $(\"#\" + id);\n",
       "    var ws_proxy = comm_websocket_adapter(comm)\n",
       "\n",
       "    function ondownload(figure, format) {\n",
       "        window.open(figure.imageObj.src);\n",
       "    }\n",
       "\n",
       "    var fig = new mpl.figure(id, ws_proxy,\n",
       "                           ondownload,\n",
       "                           element.get(0));\n",
       "\n",
       "    // Call onopen now - mpl needs it, as it is assuming we've passed it a real\n",
       "    // web socket which is closed, not our websocket->open comm proxy.\n",
       "    ws_proxy.onopen();\n",
       "\n",
       "    fig.parent_element = element.get(0);\n",
       "    fig.cell_info = mpl.find_output_cell(\"<div id='\" + id + \"'></div>\");\n",
       "    if (!fig.cell_info) {\n",
       "        console.error(\"Failed to find cell for figure\", id, fig);\n",
       "        return;\n",
       "    }\n",
       "\n",
       "    var output_index = fig.cell_info[2]\n",
       "    var cell = fig.cell_info[0];\n",
       "\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_close = function(fig, msg) {\n",
       "    var width = fig.canvas.width/mpl.ratio\n",
       "    fig.root.unbind('remove')\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable()\n",
       "    $(fig.parent_element).html('<img src=\"' + dataURL + '\" width=\"' + width + '\">');\n",
       "    fig.close_ws(fig, msg);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.close_ws = function(fig, msg){\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function(remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var width = this.canvas.width/mpl.ratio\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\" width=\"' + width + '\">';\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message(\"ack\", {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () { fig.push_to_output() }, 1000);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items){\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) { continue; };\n",
       "\n",
       "        var button = $('<button class=\"btn btn-default\" href=\"#\" title=\"' + name + '\"><i class=\"fa ' + image + ' fa-lg\"></i></button>');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = $('<span class=\"mpl-message\" style=\"text-align:right; float: right;\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = $('<div class=\"btn-group inline pull-right\"></div>');\n",
       "    var button = $('<button class=\"btn btn-mini btn-primary\" href=\"#\" title=\"Stop Interaction\"><i class=\"fa fa-power-off icon-remove icon-large\"></i></button>');\n",
       "    button.click(function (evt) { fig.handle_close(fig, {}); } );\n",
       "    button.mouseover('Stop Interaction', toolbar_mouse_event);\n",
       "    buttongrp.append(button);\n",
       "    var titlebar = this.root.find($('.ui-dialog-titlebar'));\n",
       "    titlebar.prepend(buttongrp);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(el){\n",
       "    var fig = this\n",
       "    el.on(\"remove\", function(){\n",
       "\tfig.close_ws(fig, {});\n",
       "    });\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(el){\n",
       "    // this is important to make the div 'focusable\n",
       "    el.attr('tabindex', 0)\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    }\n",
       "    else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager)\n",
       "        manager = IPython.keyboard_manager;\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which == 13) {\n",
       "        this.canvas_div.blur();\n",
       "        event.shiftKey = false;\n",
       "        // Send a \"J\" for go to next cell\n",
       "        event.which = 74;\n",
       "        event.keyCode = 74;\n",
       "        manager.command_mode();\n",
       "        manager.handle_keydown(event);\n",
       "    }\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.find_output_cell = function(html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i=0; i<ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code'){\n",
       "            for (var j=0; j<cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] == html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "}\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel != null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\" width=\"638.888905813665\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x7f994397b810>"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Plot classified data - for test data\n",
    "\n",
    "X_test = np.concatenate((classifier.X_test, classifier.X_test_outliers), axis=0)\n",
    "y_test = np.concatenate((classifier.y_test, classifier.y_test_outliers), axis=0)\n",
    "\n",
    "y_pred = classifier.model.predict(X_test)\n",
    "y_max = y_pred.max(axis = -1)\n",
    "#plt.scatter(X_test[:, 0], X_test[:, 1], s=4, c=y_max.squeeze())\n",
    "\n",
    "# Plot classified data - for test data\n",
    "\n",
    "y_labels = y_pred.argmax(axis = -1)\n",
    "\n",
    "fig = plt.figure()\n",
    "#ax = fig.add_subplot(111, projection='3d')\n",
    "#ax = Axes3D(fig)\n",
    "\n",
    "#ax.scatter(X_test[:, 0], X_test[:, 1], y_max, c=y_test.squeeze())\n",
    "plt.scatter(X_test[:, 0], X_test[:, 1], s=4, c=y_max.squeeze())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(400, 1)\n",
      "(800, 1)\n",
      "(800, 1)\n",
      "(300, 2)\n",
      "(300, 1)\n"
     ]
    }
   ],
   "source": [
    "# Create image for paper\n",
    "classifier.n = 400\n",
    "data, labels, data_out, labels_out = classifier.get_data()\n",
    "ind_0 = np.where(labels==0)[0][0:50]\n",
    "ind_1 = np.where(labels==1)[0][0:50]\n",
    "ind_2 = np.where(labels_out==2)[0]\n",
    "ind = np.random.randint(0, ind_2.shape[0], 200)\n",
    "ind_2 = ind_2[ind]\n",
    "data = np.concatenate((data[ind_0,:], data[ind_1,:], data_out[ind_2,:]), axis=0)\n",
    "labels = np.concatenate((labels[ind_0,:], labels[ind_1,:], labels_out[ind_2,:]), axis=0)\n",
    "#data_outlier, labels_outlier = classifier.get_outlier_data()\n",
    "#data = np.concatenate((data, data_outlier), axis=0)\n",
    "#labels = np.concatenate((labels, labels_outlier))\n",
    "print(data.shape)\n",
    "print(labels.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\n",
       "\n",
       "mpl.get_websocket_type = function() {\n",
       "    if (typeof(WebSocket) !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof(MozWebSocket) !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert('Your browser does not have WebSocket support.' +\n",
       "              'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "              'Firefox 4 and 5 are also supported but you ' +\n",
       "              'have to enable WebSockets in about:config.');\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = (this.ws.binaryType != undefined);\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById(\"mpl-warnings\");\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent = (\n",
       "                \"This browser does not support binary websocket messages. \" +\n",
       "                    \"Performance may be slow.\");\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = $('<div/>');\n",
       "    this._root_extra_style(this.root)\n",
       "    this.root.attr('style', 'display: inline-block');\n",
       "\n",
       "    $(parent_element).append(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen =  function () {\n",
       "            fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n",
       "            fig.send_message(\"send_image_mode\", {});\n",
       "            if (mpl.ratio != 1) {\n",
       "                fig.send_message(\"set_dpi_ratio\", {'dpi_ratio': mpl.ratio});\n",
       "            }\n",
       "            fig.send_message(\"refresh\", {});\n",
       "        }\n",
       "\n",
       "    this.imageObj.onload = function() {\n",
       "            if (fig.image_mode == 'full') {\n",
       "                // Full images could contain transparency (where diff images\n",
       "                // almost always do), so we need to clear the canvas so that\n",
       "                // there is no ghosting.\n",
       "                fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "            }\n",
       "            fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "        };\n",
       "\n",
       "    this.imageObj.onunload = function() {\n",
       "        fig.ws.close();\n",
       "    }\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_header = function() {\n",
       "    var titlebar = $(\n",
       "        '<div class=\"ui-dialog-titlebar ui-widget-header ui-corner-all ' +\n",
       "        'ui-helper-clearfix\"/>');\n",
       "    var titletext = $(\n",
       "        '<div class=\"ui-dialog-title\" style=\"width: 100%; ' +\n",
       "        'text-align: center; padding: 3px;\"/>');\n",
       "    titlebar.append(titletext)\n",
       "    this.root.append(titlebar);\n",
       "    this.header = titletext[0];\n",
       "}\n",
       "\n",
       "\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = $('<div/>');\n",
       "\n",
       "    canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n",
       "\n",
       "    function canvas_keyboard_event(event) {\n",
       "        return fig.key_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    canvas_div.keydown('key_press', canvas_keyboard_event);\n",
       "    canvas_div.keyup('key_release', canvas_keyboard_event);\n",
       "    this.canvas_div = canvas_div\n",
       "    this._canvas_extra_style(canvas_div)\n",
       "    this.root.append(canvas_div);\n",
       "\n",
       "    var canvas = $('<canvas/>');\n",
       "    canvas.addClass('mpl-canvas');\n",
       "    canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n",
       "\n",
       "    this.canvas = canvas[0];\n",
       "    this.context = canvas[0].getContext(\"2d\");\n",
       "\n",
       "    var backingStore = this.context.backingStorePixelRatio ||\n",
       "\tthis.context.webkitBackingStorePixelRatio ||\n",
       "\tthis.context.mozBackingStorePixelRatio ||\n",
       "\tthis.context.msBackingStorePixelRatio ||\n",
       "\tthis.context.oBackingStorePixelRatio ||\n",
       "\tthis.context.backingStorePixelRatio || 1;\n",
       "\n",
       "    mpl.ratio = (window.devicePixelRatio || 1) / backingStore;\n",
       "\n",
       "    var rubberband = $('<canvas/>');\n",
       "    rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n",
       "\n",
       "    var pass_mouse_events = true;\n",
       "\n",
       "    canvas_div.resizable({\n",
       "        start: function(event, ui) {\n",
       "            pass_mouse_events = false;\n",
       "        },\n",
       "        resize: function(event, ui) {\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "        stop: function(event, ui) {\n",
       "            pass_mouse_events = true;\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "    });\n",
       "\n",
       "    function mouse_event_fn(event) {\n",
       "        if (pass_mouse_events)\n",
       "            return fig.mouse_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    rubberband.mousedown('button_press', mouse_event_fn);\n",
       "    rubberband.mouseup('button_release', mouse_event_fn);\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband.mousemove('motion_notify', mouse_event_fn);\n",
       "\n",
       "    rubberband.mouseenter('figure_enter', mouse_event_fn);\n",
       "    rubberband.mouseleave('figure_leave', mouse_event_fn);\n",
       "\n",
       "    canvas_div.on(\"wheel\", function (event) {\n",
       "        event = event.originalEvent;\n",
       "        event['data'] = 'scroll'\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        mouse_event_fn(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.append(canvas);\n",
       "    canvas_div.append(rubberband);\n",
       "\n",
       "    this.rubberband = rubberband;\n",
       "    this.rubberband_canvas = rubberband[0];\n",
       "    this.rubberband_context = rubberband[0].getContext(\"2d\");\n",
       "    this.rubberband_context.strokeStyle = \"#000000\";\n",
       "\n",
       "    this._resize_canvas = function(width, height) {\n",
       "        // Keep the size of the canvas, canvas container, and rubber band\n",
       "        // canvas in synch.\n",
       "        canvas_div.css('width', width)\n",
       "        canvas_div.css('height', height)\n",
       "\n",
       "        canvas.attr('width', width * mpl.ratio);\n",
       "        canvas.attr('height', height * mpl.ratio);\n",
       "        canvas.attr('style', 'width: ' + width + 'px; height: ' + height + 'px;');\n",
       "\n",
       "        rubberband.attr('width', width);\n",
       "        rubberband.attr('height', height);\n",
       "    }\n",
       "\n",
       "    // Set the figure to an initial 600x600px, this will subsequently be updated\n",
       "    // upon first draw.\n",
       "    this._resize_canvas(600, 600);\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus () {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            // put a spacer in here.\n",
       "            continue;\n",
       "        }\n",
       "        var button = $('<button/>');\n",
       "        button.addClass('ui-button ui-widget ui-state-default ui-corner-all ' +\n",
       "                        'ui-button-icon-only');\n",
       "        button.attr('role', 'button');\n",
       "        button.attr('aria-disabled', 'false');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "\n",
       "        var icon_img = $('<span/>');\n",
       "        icon_img.addClass('ui-button-icon-primary ui-icon');\n",
       "        icon_img.addClass(image);\n",
       "        icon_img.addClass('ui-corner-all');\n",
       "\n",
       "        var tooltip_span = $('<span/>');\n",
       "        tooltip_span.addClass('ui-button-text');\n",
       "        tooltip_span.html(tooltip);\n",
       "\n",
       "        button.append(icon_img);\n",
       "        button.append(tooltip_span);\n",
       "\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    var fmt_picker_span = $('<span/>');\n",
       "\n",
       "    var fmt_picker = $('<select/>');\n",
       "    fmt_picker.addClass('mpl-toolbar-option ui-widget ui-widget-content');\n",
       "    fmt_picker_span.append(fmt_picker);\n",
       "    nav_element.append(fmt_picker_span);\n",
       "    this.format_dropdown = fmt_picker[0];\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = $(\n",
       "            '<option/>', {selected: fmt === mpl.default_extension}).html(fmt);\n",
       "        fmt_picker.append(option)\n",
       "    }\n",
       "\n",
       "    // Add hover states to the ui-buttons\n",
       "    $( \".ui-button\" ).hover(\n",
       "        function() { $(this).addClass(\"ui-state-hover\");},\n",
       "        function() { $(this).removeClass(\"ui-state-hover\");}\n",
       "    );\n",
       "\n",
       "    var status_bar = $('<span class=\"mpl-message\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.request_resize = function(x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', {'width': x_pixels, 'height': y_pixels});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_message = function(type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function() {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({type: \"draw\", figure_id: this.id}));\n",
       "    }\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function(fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] != fig.canvas.width || size[1] != fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1]);\n",
       "        fig.send_message(\"refresh\", {});\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function(fig, msg) {\n",
       "    var x0 = msg['x0'] / mpl.ratio;\n",
       "    var y0 = (fig.canvas.height - msg['y0']) / mpl.ratio;\n",
       "    var x1 = msg['x1'] / mpl.ratio;\n",
       "    var y1 = (fig.canvas.height - msg['y1']) / mpl.ratio;\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0, 0, fig.canvas.width, fig.canvas.height);\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function(fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function(fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch(cursor)\n",
       "    {\n",
       "    case 0:\n",
       "        cursor = 'pointer';\n",
       "        break;\n",
       "    case 1:\n",
       "        cursor = 'default';\n",
       "        break;\n",
       "    case 2:\n",
       "        cursor = 'crosshair';\n",
       "        break;\n",
       "    case 3:\n",
       "        cursor = 'move';\n",
       "        break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_message = function(fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function(fig, msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function(fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message(\"ack\", {});\n",
       "}\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function(fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            /* FIXME: We get \"Resource interpreted as Image but\n",
       "             * transferred with MIME type text/plain:\" errors on\n",
       "             * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "             * to be part of the websocket stream */\n",
       "            evt.data.type = \"image/png\";\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src);\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                evt.data);\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "        else if (typeof evt.data === 'string' && evt.data.slice(0, 21) == \"data:image/png;base64\") {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig[\"handle_\" + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\"No handler for the '\" + msg_type + \"' message type: \", msg);\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\"Exception inside the 'handler_\" + msg_type + \"' callback:\", e, e.stack, msg);\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "}\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function(e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e)\n",
       "        e = window.event;\n",
       "    if (e.target)\n",
       "        targ = e.target;\n",
       "    else if (e.srcElement)\n",
       "        targ = e.srcElement;\n",
       "    if (targ.nodeType == 3) // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "\n",
       "    // jQuery normalizes the pageX and pageY\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    // offset() returns the position of the element relative to the document\n",
       "    var x = e.pageX - $(targ).offset().left;\n",
       "    var y = e.pageY - $(targ).offset().top;\n",
       "\n",
       "    return {\"x\": x, \"y\": y};\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys (original) {\n",
       "  return Object.keys(original).reduce(function (obj, key) {\n",
       "    if (typeof original[key] !== 'object')\n",
       "        obj[key] = original[key]\n",
       "    return obj;\n",
       "  }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function(event, name) {\n",
       "    var canvas_pos = mpl.findpos(event)\n",
       "\n",
       "    if (name === 'button_press')\n",
       "    {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x * mpl.ratio;\n",
       "    var y = canvas_pos.y * mpl.ratio;\n",
       "\n",
       "    this.send_message(name, {x: x, y: y, button: event.button,\n",
       "                             step: event.step,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.key_event = function(event, name) {\n",
       "\n",
       "    // Prevent repeat events\n",
       "    if (name == 'key_press')\n",
       "    {\n",
       "        if (event.which === this._key)\n",
       "            return;\n",
       "        else\n",
       "            this._key = event.which;\n",
       "    }\n",
       "    if (name == 'key_release')\n",
       "        this._key = null;\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.which != 17)\n",
       "        value += \"ctrl+\";\n",
       "    if (event.altKey && event.which != 18)\n",
       "        value += \"alt+\";\n",
       "    if (event.shiftKey && event.which != 16)\n",
       "        value += \"shift+\";\n",
       "\n",
       "    value += 'k';\n",
       "    value += event.which.toString();\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, {key: value,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function(name) {\n",
       "    if (name == 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message(\"toolbar_button\", {name: name});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function(tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to  previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Pan axes with left mouse, zoom with right\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"jpeg\", \"pdf\", \"png\", \"ps\", \"raw\", \"svg\", \"tif\"];\n",
       "\n",
       "mpl.default_extension = \"png\";var comm_websocket_adapter = function(comm) {\n",
       "    // Create a \"websocket\"-like object which calls the given IPython comm\n",
       "    // object with the appropriate methods. Currently this is a non binary\n",
       "    // socket, so there is still some room for performance tuning.\n",
       "    var ws = {};\n",
       "\n",
       "    ws.close = function() {\n",
       "        comm.close()\n",
       "    };\n",
       "    ws.send = function(m) {\n",
       "        //console.log('sending', m);\n",
       "        comm.send(m);\n",
       "    };\n",
       "    // Register the callback with on_msg.\n",
       "    comm.on_msg(function(msg) {\n",
       "        //console.log('receiving', msg['content']['data'], msg);\n",
       "        // Pass the mpl event to the overridden (by mpl) onmessage function.\n",
       "        ws.onmessage(msg['content']['data'])\n",
       "    });\n",
       "    return ws;\n",
       "}\n",
       "\n",
       "mpl.mpl_figure_comm = function(comm, msg) {\n",
       "    // This is the function which gets called when the mpl process\n",
       "    // starts-up an IPython Comm through the \"matplotlib\" channel.\n",
       "\n",
       "    var id = msg.content.data.id;\n",
       "    // Get hold of the div created by the display call when the Comm\n",
       "    // socket was opened in Python.\n",
       "    var element = $(\"#\" + id);\n",
       "    var ws_proxy = comm_websocket_adapter(comm)\n",
       "\n",
       "    function ondownload(figure, format) {\n",
       "        window.open(figure.imageObj.src);\n",
       "    }\n",
       "\n",
       "    var fig = new mpl.figure(id, ws_proxy,\n",
       "                           ondownload,\n",
       "                           element.get(0));\n",
       "\n",
       "    // Call onopen now - mpl needs it, as it is assuming we've passed it a real\n",
       "    // web socket which is closed, not our websocket->open comm proxy.\n",
       "    ws_proxy.onopen();\n",
       "\n",
       "    fig.parent_element = element.get(0);\n",
       "    fig.cell_info = mpl.find_output_cell(\"<div id='\" + id + \"'></div>\");\n",
       "    if (!fig.cell_info) {\n",
       "        console.error(\"Failed to find cell for figure\", id, fig);\n",
       "        return;\n",
       "    }\n",
       "\n",
       "    var output_index = fig.cell_info[2]\n",
       "    var cell = fig.cell_info[0];\n",
       "\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_close = function(fig, msg) {\n",
       "    var width = fig.canvas.width/mpl.ratio\n",
       "    fig.root.unbind('remove')\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable()\n",
       "    $(fig.parent_element).html('<img src=\"' + dataURL + '\" width=\"' + width + '\">');\n",
       "    fig.close_ws(fig, msg);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.close_ws = function(fig, msg){\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function(remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var width = this.canvas.width/mpl.ratio\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\" width=\"' + width + '\">';\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message(\"ack\", {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () { fig.push_to_output() }, 1000);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items){\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) { continue; };\n",
       "\n",
       "        var button = $('<button class=\"btn btn-default\" href=\"#\" title=\"' + name + '\"><i class=\"fa ' + image + ' fa-lg\"></i></button>');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = $('<span class=\"mpl-message\" style=\"text-align:right; float: right;\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = $('<div class=\"btn-group inline pull-right\"></div>');\n",
       "    var button = $('<button class=\"btn btn-mini btn-primary\" href=\"#\" title=\"Stop Interaction\"><i class=\"fa fa-power-off icon-remove icon-large\"></i></button>');\n",
       "    button.click(function (evt) { fig.handle_close(fig, {}); } );\n",
       "    button.mouseover('Stop Interaction', toolbar_mouse_event);\n",
       "    buttongrp.append(button);\n",
       "    var titlebar = this.root.find($('.ui-dialog-titlebar'));\n",
       "    titlebar.prepend(buttongrp);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(el){\n",
       "    var fig = this\n",
       "    el.on(\"remove\", function(){\n",
       "\tfig.close_ws(fig, {});\n",
       "    });\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(el){\n",
       "    // this is important to make the div 'focusable\n",
       "    el.attr('tabindex', 0)\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    }\n",
       "    else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager)\n",
       "        manager = IPython.keyboard_manager;\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which == 13) {\n",
       "        this.canvas_div.blur();\n",
       "        event.shiftKey = false;\n",
       "        // Send a \"J\" for go to next cell\n",
       "        event.which = 74;\n",
       "        event.keyCode = 74;\n",
       "        manager.command_mode();\n",
       "        manager.handle_keydown(event);\n",
       "    }\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.find_output_cell = function(html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i=0; i<ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code'){\n",
       "            for (var j=0; j<cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] == html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "}\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel != null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\" width=\"498.88890210493145\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_pred = classifier.model.predict(data)\n",
    "y_labels = y_pred.argmax(axis = -1)\n",
    "y_max = y_pred.max(axis = -1)\n",
    "\n",
    "ind = np.where(y_max <= 0.6)[0]\n",
    "y_labels[ind] = 2\n",
    "\n",
    "ind = np.where(y_max >= 0.6)[0]\n",
    "y_labels[ind] = 0\n",
    "\n",
    "ind = np.where(((y_max > 0.6) & (y_max < 0.9)))[0]\n",
    "y_labels[ind] = 1\n",
    "\n",
    "\n",
    "c_list = ['g', '#ADD8E6', 'y']\n",
    "marker_list = ['D', 'o', '*']\n",
    "plt.figure(figsize=(5,5))\n",
    "for i in range(0, data.shape[0]):\n",
    "    marker = marker_list[int(labels[i])]\n",
    "    c = c_list[y_labels[i]]\n",
    "    plt.scatter(data[i, 0], data[i, 1], s=10, c=c, marker=marker)\n",
    "\n",
    "plt.legend(('$class\\ 0$','$class\\ 1$','$OOD\\ sample$'), fontsize='x-small');\n",
    "import matplotlib.lines as lines\n",
    "dimond_line = lines.Line2D([0], [0], color='w', marker='D', markerfacecolor='#000000',\n",
    "                          markersize=5, label='TC 0')\n",
    "\n",
    "circle_line = lines.Line2D([0], [0], marker='o', color='w', label='TC 1',\n",
    "                          markerfacecolor='#000000', markersize=5)\n",
    "\n",
    "star_line = lines.Line2D([0], [0], color='w', marker='*', markerfacecolor='#000000',\n",
    "                          markersize=10, label='TC 2')\n",
    "\n",
    "\n",
    "green_line = lines.Line2D([0], [0], marker='s', color='w', label='[0.9, 1]',\n",
    "                          markerfacecolor='g', markersize=5)\n",
    "\n",
    "\n",
    "lightblue_line = lines.Line2D([0], [0], marker='s', color='w', label='(0.6, 0.9)',\n",
    "                          markerfacecolor='#ADD8E6', markersize=5)\n",
    "\n",
    "\n",
    "yellow_line = lines.Line2D([0], [0], marker='s', color='w', label='[0.5, 0.6]',\n",
    "                          markerfacecolor='y', markersize=5)\n",
    "\n",
    "# Create a legend for the first line.\n",
    "first_legend = plt.legend(handles=[dimond_line, circle_line, star_line], loc=1)\n",
    "\n",
    "\n",
    "# Add the legend manually to the current Axes.\n",
    "ax = plt.gca().add_artist(first_legend)\n",
    "\n",
    "# Create another legend for the second line.\n",
    "plt.legend(handles=[green_line, lightblue_line, yellow_line], loc=4)\n",
    "\n",
    "plt.show()\n",
    "plt.savefig(\"exp2_b\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(300,)\n",
      "(98,)\n",
      "(196,)\n"
     ]
    }
   ],
   "source": [
    "print(y_max.shape)\n",
    "print(np.where(y_max>0.95)[0].shape)\n",
    "print(np.where(y_max<0.55)[0].shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0., 1., 2.], dtype=float32)"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.unique(y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(500,)\n"
     ]
    }
   ],
   "source": [
    "print(np.where(y_test==0)[0].shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}