main.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/aamomeni/anaconda3/lib/python3.6/site-packages/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.\n",
      "  from ._conv import register_converters as _register_converters\n",
      "Using TensorFlow backend.\n",
      "/home/aamomeni/anaconda3/lib/python3.6/site-packages/sklearn/cross_validation.py:41: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.\n",
      "  \"This module will be removed in 0.20.\", DeprecationWarning)\n",
      "/home/aamomeni/Seven-Segment-OCR/Model.py:18: UserWarning: \n",
      "This call to matplotlib.use() has no effect because the backend has already\n",
      "been chosen; matplotlib.use() must be called *before* pylab, matplotlib.pyplot,\n",
      "or matplotlib.backends is imported for the first time.\n",
      "\n",
      "The backend was *originally* set to 'module://ipykernel.pylab.backend_inline' by the following code:\n",
      "  File \"/home/aamomeni/anaconda3/lib/python3.6/runpy.py\", line 193, in _run_module_as_main\n",
      "    \"__main__\", mod_spec)\n",
      "  File \"/home/aamomeni/anaconda3/lib/python3.6/runpy.py\", line 85, in _run_code\n",
      "    exec(code, run_globals)\n",
      "  File \"/home/aamomeni/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py\", line 16, in <module>\n",
      "    app.launch_new_instance()\n",
      "  File \"/home/aamomeni/anaconda3/lib/python3.6/site-packages/traitlets/config/application.py\", line 658, in launch_instance\n",
      "    app.start()\n",
      "  File \"/home/aamomeni/anaconda3/lib/python3.6/site-packages/ipykernel/kernelapp.py\", line 478, in start\n",
      "    self.io_loop.start()\n",
      "  File \"/home/aamomeni/anaconda3/lib/python3.6/site-packages/zmq/eventloop/ioloop.py\", line 177, in start\n",
      "    super(ZMQIOLoop, self).start()\n",
      "  File \"/home/aamomeni/anaconda3/lib/python3.6/site-packages/tornado/ioloop.py\", line 888, in start\n",
      "    handler_func(fd_obj, events)\n",
      "  File \"/home/aamomeni/anaconda3/lib/python3.6/site-packages/tornado/stack_context.py\", line 277, in null_wrapper\n",
      "    return fn(*args, **kwargs)\n",
      "  File \"/home/aamomeni/anaconda3/lib/python3.6/site-packages/zmq/eventloop/zmqstream.py\", line 440, in _handle_events\n",
      "    self._handle_recv()\n",
      "  File \"/home/aamomeni/anaconda3/lib/python3.6/site-packages/zmq/eventloop/zmqstream.py\", line 472, in _handle_recv\n",
      "    self._run_callback(callback, msg)\n",
      "  File \"/home/aamomeni/anaconda3/lib/python3.6/site-packages/zmq/eventloop/zmqstream.py\", line 414, in _run_callback\n",
      "    callback(*args, **kwargs)\n",
      "  File \"/home/aamomeni/anaconda3/lib/python3.6/site-packages/tornado/stack_context.py\", line 277, in null_wrapper\n",
      "    return fn(*args, **kwargs)\n",
      "  File \"/home/aamomeni/anaconda3/lib/python3.6/site-packages/ipykernel/kernelbase.py\", line 283, in dispatcher\n",
      "    return self.dispatch_shell(stream, msg)\n",
      "  File \"/home/aamomeni/anaconda3/lib/python3.6/site-packages/ipykernel/kernelbase.py\", line 233, in dispatch_shell\n",
      "    handler(stream, idents, msg)\n",
      "  File \"/home/aamomeni/anaconda3/lib/python3.6/site-packages/ipykernel/kernelbase.py\", line 399, in execute_request\n",
      "    user_expressions, allow_stdin)\n",
      "  File \"/home/aamomeni/anaconda3/lib/python3.6/site-packages/ipykernel/ipkernel.py\", line 208, in do_execute\n",
      "    res = shell.run_cell(code, store_history=store_history, silent=silent)\n",
      "  File \"/home/aamomeni/anaconda3/lib/python3.6/site-packages/ipykernel/zmqshell.py\", line 537, in run_cell\n",
      "    return super(ZMQInteractiveShell, self).run_cell(*args, **kwargs)\n",
      "  File \"/home/aamomeni/anaconda3/lib/python3.6/site-packages/IPython/core/interactiveshell.py\", line 2728, in run_cell\n",
      "    interactivity=interactivity, compiler=compiler, result=result)\n",
      "  File \"/home/aamomeni/anaconda3/lib/python3.6/site-packages/IPython/core/interactiveshell.py\", line 2850, in run_ast_nodes\n",
      "    if self.run_code(code, result):\n",
      "  File \"/home/aamomeni/anaconda3/lib/python3.6/site-packages/IPython/core/interactiveshell.py\", line 2910, in run_code\n",
      "    exec(code_obj, self.user_global_ns, self.user_ns)\n",
      "  File \"<ipython-input-1-886a08fc224d>\", line 2, in <module>\n",
      "    from Model import Model_Multi, Model_Single\n",
      "  File \"/home/aamomeni/Seven-Segment-OCR/Model.py\", line 16, in <module>\n",
      "    from Datasets import Dataset_Multi, Dataset_Single\n",
      "  File \"/home/aamomeni/Seven-Segment-OCR/Datasets.py\", line 4, in <module>\n",
      "    import matplotlib.pyplot as plt\n",
      "  File \"/home/aamomeni/anaconda3/lib/python3.6/site-packages/matplotlib/pyplot.py\", line 71, in <module>\n",
      "    from matplotlib.backends import pylab_setup\n",
      "  File \"/home/aamomeni/anaconda3/lib/python3.6/site-packages/matplotlib/backends/__init__.py\", line 16, in <module>\n",
      "    line for line in traceback.format_stack()\n",
      "\n",
      "\n",
      "  matplotlib.use('agg')\n",
      "/home/aamomeni/Seven-Segment-OCR/Model.py:81: UserWarning: Update your `Dense` call to the Keras 2 API: `Dense(activation=\"softmax\", name=\"digit_1\", units=11)`\n",
      "  digit1 = (Dense(output_dim =11,activation = 'softmax', name='digit_1'))(x)\n",
      "/home/aamomeni/Seven-Segment-OCR/Model.py:82: UserWarning: Update your `Dense` call to the Keras 2 API: `Dense(activation=\"softmax\", name=\"digit_2\", units=11)`\n",
      "  digit2 = (Dense(output_dim =11,activation = 'softmax', name='digit_2'))(x)\n",
      "/home/aamomeni/Seven-Segment-OCR/Model.py:83: UserWarning: Update your `Dense` call to the Keras 2 API: `Dense(activation=\"softmax\", name=\"digit_3\", units=11)`\n",
      "  digit3 = (Dense(output_dim =11,activation = 'softmax', name='digit_3'))(x)\n",
      "/home/aamomeni/Seven-Segment-OCR/Model.py:84: UserWarning: Update your `Dense` call to the Keras 2 API: `Dense(activation=\"softmax\", name=\"digit_4\", units=11)`\n",
      "  digit4 = (Dense(output_dim =11,activation = 'softmax', name='digit_4'))(x)\n",
      "/home/aamomeni/Seven-Segment-OCR/Model.py:88: UserWarning: Update your `Model` call to the Keras 2 API: `Model(inputs=Tensor(\"in..., outputs=[<tf.Tenso...)`\n",
      "  self.model = keras.models.Model(input = model_input , output = outputs)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /home/aamomeni/anaconda3/lib/python3.6/site-packages/tensorflow/python/util/tf_should_use.py:118: initialize_all_variables (from tensorflow.python.ops.variables) is deprecated and will be removed after 2017-03-02.\n",
      "Instructions for updating:\n",
      "Use `tf.global_variables_initializer` instead.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/aamomeni/Seven-Segment-OCR/Model.py:95: UserWarning: The `nb_epoch` argument in `fit` has been renamed `epochs`.\n",
      "  self.history = self.model.fit(self.X_train, self.y_train_vect, batch_size= 50, nb_epoch=epochs, verbose=1, validation_data=(self.X_val, self.y_val_vect))\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:Variable *= will be deprecated. Use variable.assign_mul if you want assignment to the variable value or 'x = x * y' if you want a new python Tensor object.\n",
      "Train on 516 samples, validate on 172 samples\n",
      "Epoch 1/50\n",
      "516/516 [==============================] - 10s 19ms/step - loss: 16.5210 - digit_1_loss: 0.3127 - digit_2_loss: 2.8536 - digit_3_loss: 3.6920 - digit_4_loss: 3.5457 - digit_1_acc: 0.9070 - digit_2_acc: 0.4593 - digit_3_acc: 0.1570 - digit_4_acc: 0.1047 - val_loss: 12.2617 - val_digit_1_loss: 1.0729e-06 - val_digit_2_loss: 0.9464 - val_digit_3_loss: 2.3587 - val_digit_4_loss: 2.9058 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.6395 - val_digit_3_acc: 0.1512 - val_digit_4_acc: 0.1163\n",
      "Epoch 2/50\n",
      "516/516 [==============================] - 0s 902us/step - loss: 12.4557 - digit_1_loss: 1.0729e-06 - digit_2_loss: 1.0329 - digit_3_loss: 2.6354 - digit_4_loss: 2.8207 - digit_1_acc: 1.0000 - digit_2_acc: 0.7171 - digit_3_acc: 0.1880 - digit_4_acc: 0.1512 - val_loss: 11.6260 - val_digit_1_loss: 1.0729e-06 - val_digit_2_loss: 0.8572 - val_digit_3_loss: 2.4984 - val_digit_4_loss: 2.4349 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.7616 - val_digit_3_acc: 0.1802 - val_digit_4_acc: 0.1395\n",
      "Epoch 3/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 11.1871 - digit_1_loss: 1.0729e-06 - digit_2_loss: 0.8058 - digit_3_loss: 2.3470 - digit_4_loss: 2.3099 - digit_1_acc: 1.0000 - digit_2_acc: 0.7636 - digit_3_acc: 0.2616 - digit_4_acc: 0.2248 - val_loss: 10.6026 - val_digit_1_loss: 1.0736e-06 - val_digit_2_loss: 0.8218 - val_digit_3_loss: 1.9626 - val_digit_4_loss: 2.2490 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.7791 - val_digit_3_acc: 0.3430 - val_digit_4_acc: 0.2267\n",
      "Epoch 4/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 10.3509 - digit_1_loss: 1.0729e-06 - digit_2_loss: 0.7577 - digit_3_loss: 1.9547 - digit_4_loss: 2.1854 - digit_1_acc: 1.0000 - digit_2_acc: 0.7422 - digit_3_acc: 0.3508 - digit_4_acc: 0.2636 - val_loss: 10.0480 - val_digit_1_loss: 1.1353e-06 - val_digit_2_loss: 0.7223 - val_digit_3_loss: 1.7827 - val_digit_4_loss: 2.2468 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.7965 - val_digit_3_acc: 0.4244 - val_digit_4_acc: 0.1919\n",
      "Epoch 5/50\n",
      "516/516 [==============================] - 0s 955us/step - loss: 9.5146 - digit_1_loss: 1.0729e-06 - digit_2_loss: 0.6234 - digit_3_loss: 1.6826 - digit_4_loss: 2.0249 - digit_1_acc: 1.0000 - digit_2_acc: 0.8081 - digit_3_acc: 0.4167 - digit_4_acc: 0.2868 - val_loss: 9.6984 - val_digit_1_loss: 2.0958e-06 - val_digit_2_loss: 0.7341 - val_digit_3_loss: 1.8033 - val_digit_4_loss: 2.1271 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.7965 - val_digit_3_acc: 0.4186 - val_digit_4_acc: 0.2558\n",
      "Epoch 6/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 8.9493 - digit_1_loss: 1.0729e-06 - digit_2_loss: 0.5566 - digit_3_loss: 1.6243 - digit_4_loss: 1.8404 - digit_1_acc: 1.0000 - digit_2_acc: 0.8198 - digit_3_acc: 0.4651 - digit_4_acc: 0.3314 - val_loss: 9.4409 - val_digit_1_loss: 5.6431e-06 - val_digit_2_loss: 0.7123 - val_digit_3_loss: 1.7257 - val_digit_4_loss: 2.2146 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8140 - val_digit_3_acc: 0.3837 - val_digit_4_acc: 0.2326\n",
      "Epoch 7/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 8.5011 - digit_1_loss: 1.0729e-06 - digit_2_loss: 0.5499 - digit_3_loss: 1.4567 - digit_4_loss: 1.8048 - digit_1_acc: 1.0000 - digit_2_acc: 0.8101 - digit_3_acc: 0.5019 - digit_4_acc: 0.4012 - val_loss: 9.0521 - val_digit_1_loss: 9.7810e-06 - val_digit_2_loss: 0.6641 - val_digit_3_loss: 1.6052 - val_digit_4_loss: 2.2237 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8198 - val_digit_3_acc: 0.4302 - val_digit_4_acc: 0.2093\n",
      "Epoch 8/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 7.9541 - digit_1_loss: 1.0752e-06 - digit_2_loss: 0.4906 - digit_3_loss: 1.3211 - digit_4_loss: 1.6748 - digit_1_acc: 1.0000 - digit_2_acc: 0.8450 - digit_3_acc: 0.5620 - digit_4_acc: 0.4380 - val_loss: 8.8649 - val_digit_1_loss: 1.0757e-05 - val_digit_2_loss: 0.7967 - val_digit_3_loss: 1.6308 - val_digit_4_loss: 2.0907 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.7674 - val_digit_3_acc: 0.4360 - val_digit_4_acc: 0.3081\n",
      "Epoch 9/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 7.5495 - digit_1_loss: 1.0740e-06 - digit_2_loss: 0.4623 - digit_3_loss: 1.2656 - digit_4_loss: 1.5596 - digit_1_acc: 1.0000 - digit_2_acc: 0.8391 - digit_3_acc: 0.5930 - digit_4_acc: 0.4806 - val_loss: 8.5327 - val_digit_1_loss: 1.2069e-05 - val_digit_2_loss: 0.6604 - val_digit_3_loss: 1.6362 - val_digit_4_loss: 2.0862 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8081 - val_digit_3_acc: 0.4419 - val_digit_4_acc: 0.3081\n",
      "Epoch 10/50\n",
      "516/516 [==============================] - 0s 919us/step - loss: 7.0758 - digit_1_loss: 1.0750e-06 - digit_2_loss: 0.4297 - digit_3_loss: 1.1462 - digit_4_loss: 1.4291 - digit_1_acc: 1.0000 - digit_2_acc: 0.8566 - digit_3_acc: 0.6240 - digit_4_acc: 0.5136 - val_loss: 8.2938 - val_digit_1_loss: 1.6559e-05 - val_digit_2_loss: 0.7455 - val_digit_3_loss: 1.4775 - val_digit_4_loss: 2.1040 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.7907 - val_digit_3_acc: 0.5465 - val_digit_4_acc: 0.2791\n",
      "Epoch 11/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 6.8161 - digit_1_loss: 1.0770e-06 - digit_2_loss: 0.3949 - digit_3_loss: 1.0927 - digit_4_loss: 1.4345 - digit_1_acc: 1.0000 - digit_2_acc: 0.8643 - digit_3_acc: 0.6415 - digit_4_acc: 0.5465 - val_loss: 8.1714 - val_digit_1_loss: 1.7901e-05 - val_digit_2_loss: 0.6946 - val_digit_3_loss: 1.5644 - val_digit_4_loss: 2.1144 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.7907 - val_digit_3_acc: 0.4419 - val_digit_4_acc: 0.2849\n",
      "Epoch 12/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 6.3089 - digit_1_loss: 1.1057e-06 - digit_2_loss: 0.3460 - digit_3_loss: 0.9931 - digit_4_loss: 1.2390 - digit_1_acc: 1.0000 - digit_2_acc: 0.8895 - digit_3_acc: 0.6667 - digit_4_acc: 0.6240 - val_loss: 7.8602 - val_digit_1_loss: 1.5979e-05 - val_digit_2_loss: 0.6383 - val_digit_3_loss: 1.4468 - val_digit_4_loss: 2.1338 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.7965 - val_digit_3_acc: 0.5058 - val_digit_4_acc: 0.3140\n",
      "Epoch 13/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 5.9985 - digit_1_loss: 1.0849e-06 - digit_2_loss: 0.3453 - digit_3_loss: 0.9335 - digit_4_loss: 1.1416 - digit_1_acc: 1.0000 - digit_2_acc: 0.8915 - digit_3_acc: 0.7112 - digit_4_acc: 0.6240 - val_loss: 7.6809 - val_digit_1_loss: 7.1409e-06 - val_digit_2_loss: 0.6173 - val_digit_3_loss: 1.4998 - val_digit_4_loss: 2.0693 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8256 - val_digit_3_acc: 0.5465 - val_digit_4_acc: 0.3023\n",
      "Epoch 14/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 5.6558 - digit_1_loss: 1.1205e-06 - digit_2_loss: 0.2897 - digit_3_loss: 0.8579 - digit_4_loss: 1.0729 - digit_1_acc: 1.0000 - digit_2_acc: 0.8992 - digit_3_acc: 0.7345 - digit_4_acc: 0.6531 - val_loss: 7.7670 - val_digit_1_loss: 1.3306e-05 - val_digit_2_loss: 0.6206 - val_digit_3_loss: 1.5397 - val_digit_4_loss: 2.2493 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8081 - val_digit_3_acc: 0.5058 - val_digit_4_acc: 0.3081\n",
      "Epoch 15/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 5.3927 - digit_1_loss: 1.1463e-06 - digit_2_loss: 0.2754 - digit_3_loss: 0.7918 - digit_4_loss: 1.0230 - digit_1_acc: 1.0000 - digit_2_acc: 0.9109 - digit_3_acc: 0.7345 - digit_4_acc: 0.6550 - val_loss: 7.3198 - val_digit_1_loss: 1.5880e-05 - val_digit_2_loss: 0.6336 - val_digit_3_loss: 1.5021 - val_digit_4_loss: 1.9545 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8256 - val_digit_3_acc: 0.4942 - val_digit_4_acc: 0.3779\n",
      "Epoch 16/50\n",
      "516/516 [==============================] - 0s 963us/step - loss: 5.0022 - digit_1_loss: 1.2697e-06 - digit_2_loss: 0.2458 - digit_3_loss: 0.6689 - digit_4_loss: 0.9090 - digit_1_acc: 1.0000 - digit_2_acc: 0.9225 - digit_3_acc: 0.8023 - digit_4_acc: 0.6996 - val_loss: 7.1431 - val_digit_1_loss: 2.0070e-05 - val_digit_2_loss: 0.6911 - val_digit_3_loss: 1.4475 - val_digit_4_loss: 1.8941 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8081 - val_digit_3_acc: 0.5058 - val_digit_4_acc: 0.3895\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 17/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 4.8441 - digit_1_loss: 1.3004e-06 - digit_2_loss: 0.2104 - digit_3_loss: 0.6459 - digit_4_loss: 0.9257 - digit_1_acc: 1.0000 - digit_2_acc: 0.9360 - digit_3_acc: 0.7984 - digit_4_acc: 0.7054 - val_loss: 6.9537 - val_digit_1_loss: 1.1224e-05 - val_digit_2_loss: 0.6151 - val_digit_3_loss: 1.3407 - val_digit_4_loss: 2.0000 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8198 - val_digit_3_acc: 0.5407 - val_digit_4_acc: 0.3547\n",
      "Epoch 18/50\n",
      "516/516 [==============================] - 1s 979us/step - loss: 4.5550 - digit_1_loss: 1.4268e-06 - digit_2_loss: 0.2025 - digit_3_loss: 0.5901 - digit_4_loss: 0.8097 - digit_1_acc: 1.0000 - digit_2_acc: 0.9244 - digit_3_acc: 0.8062 - digit_4_acc: 0.7345 - val_loss: 6.9456 - val_digit_1_loss: 7.0148e-06 - val_digit_2_loss: 0.6274 - val_digit_3_loss: 1.3633 - val_digit_4_loss: 2.0630 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8314 - val_digit_3_acc: 0.5698 - val_digit_4_acc: 0.3663\n",
      "Epoch 19/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 4.4110 - digit_1_loss: 1.4370e-06 - digit_2_loss: 0.1708 - digit_3_loss: 0.5609 - digit_4_loss: 0.8301 - digit_1_acc: 1.0000 - digit_2_acc: 0.9477 - digit_3_acc: 0.8295 - digit_4_acc: 0.7209 - val_loss: 6.8340 - val_digit_1_loss: 5.7747e-06 - val_digit_2_loss: 0.6275 - val_digit_3_loss: 1.3504 - val_digit_4_loss: 2.0627 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8256 - val_digit_3_acc: 0.5407 - val_digit_4_acc: 0.3314\n",
      "Epoch 20/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 4.0422 - digit_1_loss: 1.3224e-06 - digit_2_loss: 0.1778 - digit_3_loss: 0.4738 - digit_4_loss: 0.6353 - digit_1_acc: 1.0000 - digit_2_acc: 0.9302 - digit_3_acc: 0.8624 - digit_4_acc: 0.8140 - val_loss: 7.2158 - val_digit_1_loss: 3.7410e-06 - val_digit_2_loss: 0.7746 - val_digit_3_loss: 1.4701 - val_digit_4_loss: 2.2662 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.7733 - val_digit_3_acc: 0.4942 - val_digit_4_acc: 0.3081\n",
      "Epoch 21/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 3.9792 - digit_1_loss: 1.3492e-06 - digit_2_loss: 0.1788 - digit_3_loss: 0.4291 - digit_4_loss: 0.7021 - digit_1_acc: 1.0000 - digit_2_acc: 0.9341 - digit_3_acc: 0.8876 - digit_4_acc: 0.7771 - val_loss: 6.7440 - val_digit_1_loss: 5.6063e-06 - val_digit_2_loss: 0.5966 - val_digit_3_loss: 1.4051 - val_digit_4_loss: 2.1194 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8314 - val_digit_3_acc: 0.5988 - val_digit_4_acc: 0.3372\n",
      "Epoch 22/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 3.6120 - digit_1_loss: 1.5853e-06 - digit_2_loss: 0.1191 - digit_3_loss: 0.3711 - digit_4_loss: 0.5313 - digit_1_acc: 1.0000 - digit_2_acc: 0.9632 - digit_3_acc: 0.8973 - digit_4_acc: 0.8450 - val_loss: 6.5037 - val_digit_1_loss: 4.3639e-06 - val_digit_2_loss: 0.6276 - val_digit_3_loss: 1.3601 - val_digit_4_loss: 1.9691 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8023 - val_digit_3_acc: 0.5407 - val_digit_4_acc: 0.3779\n",
      "Epoch 23/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 3.4790 - digit_1_loss: 1.4596e-06 - digit_2_loss: 0.1195 - digit_3_loss: 0.3512 - digit_4_loss: 0.4936 - digit_1_acc: 1.0000 - digit_2_acc: 0.9671 - digit_3_acc: 0.9070 - digit_4_acc: 0.8624 - val_loss: 6.6657 - val_digit_1_loss: 2.5962e-06 - val_digit_2_loss: 0.6298 - val_digit_3_loss: 1.4125 - val_digit_4_loss: 2.1530 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8256 - val_digit_3_acc: 0.5523 - val_digit_4_acc: 0.3140\n",
      "Epoch 24/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 3.3087 - digit_1_loss: 1.5686e-06 - digit_2_loss: 0.1051 - digit_3_loss: 0.3225 - digit_4_loss: 0.4433 - digit_1_acc: 1.0000 - digit_2_acc: 0.9729 - digit_3_acc: 0.9167 - digit_4_acc: 0.8798 - val_loss: 6.5365 - val_digit_1_loss: 2.4118e-06 - val_digit_2_loss: 0.7079 - val_digit_3_loss: 1.3729 - val_digit_4_loss: 2.0620 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8140 - val_digit_3_acc: 0.5581 - val_digit_4_acc: 0.3430\n",
      "Epoch 25/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 3.1817 - digit_1_loss: 1.5448e-06 - digit_2_loss: 0.1027 - digit_3_loss: 0.3039 - digit_4_loss: 0.4133 - digit_1_acc: 1.0000 - digit_2_acc: 0.9671 - digit_3_acc: 0.9225 - digit_4_acc: 0.8915 - val_loss: 6.5433 - val_digit_1_loss: 2.5075e-06 - val_digit_2_loss: 0.6183 - val_digit_3_loss: 1.3754 - val_digit_4_loss: 2.2304 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8372 - val_digit_3_acc: 0.5814 - val_digit_4_acc: 0.3372\n",
      "Epoch 26/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 3.0590 - digit_1_loss: 1.3416e-06 - digit_2_loss: 0.0982 - digit_3_loss: 0.2706 - digit_4_loss: 0.4013 - digit_1_acc: 1.0000 - digit_2_acc: 0.9690 - digit_3_acc: 0.9419 - digit_4_acc: 0.9050 - val_loss: 6.3684 - val_digit_1_loss: 2.8755e-06 - val_digit_2_loss: 0.6020 - val_digit_3_loss: 1.3781 - val_digit_4_loss: 2.1398 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8256 - val_digit_3_acc: 0.5930 - val_digit_4_acc: 0.2907\n",
      "Epoch 27/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 2.9256 - digit_1_loss: 1.4642e-06 - digit_2_loss: 0.0737 - digit_3_loss: 0.2668 - digit_4_loss: 0.3648 - digit_1_acc: 1.0000 - digit_2_acc: 0.9864 - digit_3_acc: 0.9380 - digit_4_acc: 0.9089 - val_loss: 6.0638 - val_digit_1_loss: 3.5622e-06 - val_digit_2_loss: 0.5658 - val_digit_3_loss: 1.2401 - val_digit_4_loss: 2.0744 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8314 - val_digit_3_acc: 0.6279 - val_digit_4_acc: 0.3663\n",
      "Epoch 28/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 2.8089 - digit_1_loss: 1.6869e-06 - digit_2_loss: 0.0705 - digit_3_loss: 0.2322 - digit_4_loss: 0.3478 - digit_1_acc: 1.0000 - digit_2_acc: 0.9845 - digit_3_acc: 0.9419 - digit_4_acc: 0.9012 - val_loss: 6.0693 - val_digit_1_loss: 3.4756e-06 - val_digit_2_loss: 0.5866 - val_digit_3_loss: 1.2312 - val_digit_4_loss: 2.1269 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8256 - val_digit_3_acc: 0.6337 - val_digit_4_acc: 0.3605\n",
      "Epoch 29/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 2.7146 - digit_1_loss: 1.6236e-06 - digit_2_loss: 0.0791 - digit_3_loss: 0.2014 - digit_4_loss: 0.3349 - digit_1_acc: 1.0000 - digit_2_acc: 0.9787 - digit_3_acc: 0.9593 - digit_4_acc: 0.9225 - val_loss: 6.1502 - val_digit_1_loss: 2.8692e-06 - val_digit_2_loss: 0.6107 - val_digit_3_loss: 1.3569 - val_digit_4_loss: 2.1176 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8372 - val_digit_3_acc: 0.5872 - val_digit_4_acc: 0.3198\n",
      "Epoch 30/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 2.5999 - digit_1_loss: 1.3356e-06 - digit_2_loss: 0.0695 - digit_3_loss: 0.1884 - digit_4_loss: 0.3009 - digit_1_acc: 1.0000 - digit_2_acc: 0.9787 - digit_3_acc: 0.9651 - digit_4_acc: 0.9360 - val_loss: 5.8657 - val_digit_1_loss: 3.0903e-06 - val_digit_2_loss: 0.5962 - val_digit_3_loss: 1.2812 - val_digit_4_loss: 1.9799 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8488 - val_digit_3_acc: 0.6395 - val_digit_4_acc: 0.4070\n",
      "Epoch 31/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 2.5580 - digit_1_loss: 1.3319e-06 - digit_2_loss: 0.0692 - digit_3_loss: 0.1995 - digit_4_loss: 0.3048 - digit_1_acc: 1.0000 - digit_2_acc: 0.9806 - digit_3_acc: 0.9516 - digit_4_acc: 0.9380 - val_loss: 5.9396 - val_digit_1_loss: 2.6468e-06 - val_digit_2_loss: 0.5971 - val_digit_3_loss: 1.4018 - val_digit_4_loss: 1.9888 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8488 - val_digit_3_acc: 0.5988 - val_digit_4_acc: 0.3663\n",
      "Epoch 32/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 2.4187 - digit_1_loss: 1.3933e-06 - digit_2_loss: 0.0598 - digit_3_loss: 0.1666 - digit_4_loss: 0.2641 - digit_1_acc: 1.0000 - digit_2_acc: 0.9806 - digit_3_acc: 0.9671 - digit_4_acc: 0.9496 - val_loss: 5.7042 - val_digit_1_loss: 2.5137e-06 - val_digit_2_loss: 0.6182 - val_digit_3_loss: 1.2594 - val_digit_4_loss: 1.9305 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8430 - val_digit_3_acc: 0.6221 - val_digit_4_acc: 0.4360\n",
      "Epoch 33/50\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "516/516 [==============================] - 1s 1ms/step - loss: 2.3800 - digit_1_loss: 1.3039e-06 - digit_2_loss: 0.0509 - digit_3_loss: 0.1572 - digit_4_loss: 0.2973 - digit_1_acc: 1.0000 - digit_2_acc: 0.9942 - digit_3_acc: 0.9709 - digit_4_acc: 0.9225 - val_loss: 6.0049 - val_digit_1_loss: 2.2878e-06 - val_digit_2_loss: 0.6312 - val_digit_3_loss: 1.3758 - val_digit_4_loss: 2.1521 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8256 - val_digit_3_acc: 0.5872 - val_digit_4_acc: 0.3372\n",
      "Epoch 34/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 2.2881 - digit_1_loss: 1.4453e-06 - digit_2_loss: 0.0505 - digit_3_loss: 0.1493 - digit_4_loss: 0.2634 - digit_1_acc: 1.0000 - digit_2_acc: 0.9903 - digit_3_acc: 0.9767 - digit_4_acc: 0.9438 - val_loss: 6.2815 - val_digit_1_loss: 2.5477e-06 - val_digit_2_loss: 0.6468 - val_digit_3_loss: 1.3832 - val_digit_4_loss: 2.4549 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8081 - val_digit_3_acc: 0.5814 - val_digit_4_acc: 0.2965\n",
      "Epoch 35/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 2.2406 - digit_1_loss: 1.5372e-06 - digit_2_loss: 0.0574 - digit_3_loss: 0.1588 - digit_4_loss: 0.2483 - digit_1_acc: 1.0000 - digit_2_acc: 0.9942 - digit_3_acc: 0.9651 - digit_4_acc: 0.9380 - val_loss: 5.8016 - val_digit_1_loss: 3.0612e-06 - val_digit_2_loss: 0.6595 - val_digit_3_loss: 1.2765 - val_digit_4_loss: 2.1164 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8140 - val_digit_3_acc: 0.5988 - val_digit_4_acc: 0.3430\n",
      "Epoch 36/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 2.1369 - digit_1_loss: 1.2512e-06 - digit_2_loss: 0.0488 - digit_3_loss: 0.1285 - digit_4_loss: 0.2272 - digit_1_acc: 1.0000 - digit_2_acc: 0.9864 - digit_3_acc: 0.9787 - digit_4_acc: 0.9457 - val_loss: 5.9821 - val_digit_1_loss: 3.1256e-06 - val_digit_2_loss: 0.6488 - val_digit_3_loss: 1.3193 - val_digit_4_loss: 2.3040 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8314 - val_digit_3_acc: 0.6163 - val_digit_4_acc: 0.2849\n",
      "Epoch 37/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 2.0576 - digit_1_loss: 1.2066e-06 - digit_2_loss: 0.0452 - digit_3_loss: 0.1259 - digit_4_loss: 0.1938 - digit_1_acc: 1.0000 - digit_2_acc: 0.9922 - digit_3_acc: 0.9729 - digit_4_acc: 0.9554 - val_loss: 5.4383 - val_digit_1_loss: 4.2378e-06 - val_digit_2_loss: 0.5961 - val_digit_3_loss: 1.2576 - val_digit_4_loss: 1.9165 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8430 - val_digit_3_acc: 0.6163 - val_digit_4_acc: 0.4128\n",
      "Epoch 38/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 2.0166 - digit_1_loss: 1.4229e-06 - digit_2_loss: 0.0443 - digit_3_loss: 0.1333 - digit_4_loss: 0.1878 - digit_1_acc: 1.0000 - digit_2_acc: 0.9903 - digit_3_acc: 0.9806 - digit_4_acc: 0.9651 - val_loss: 6.1744 - val_digit_1_loss: 3.2698e-06 - val_digit_2_loss: 0.5980 - val_digit_3_loss: 1.4320 - val_digit_4_loss: 2.5156 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8547 - val_digit_3_acc: 0.5291 - val_digit_4_acc: 0.2616\n",
      "Epoch 39/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 1.9615 - digit_1_loss: 1.5987e-06 - digit_2_loss: 0.0556 - digit_3_loss: 0.1195 - digit_4_loss: 0.1742 - digit_1_acc: 1.0000 - digit_2_acc: 0.9903 - digit_3_acc: 0.9806 - digit_4_acc: 0.9496 - val_loss: 5.9199 - val_digit_1_loss: 3.6800e-06 - val_digit_2_loss: 0.5691 - val_digit_3_loss: 1.3992 - val_digit_4_loss: 2.3620 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8488 - val_digit_3_acc: 0.5756 - val_digit_4_acc: 0.3023\n",
      "Epoch 40/50\n",
      "516/516 [==============================] - 0s 938us/step - loss: 1.9057 - digit_1_loss: 2.0636e-06 - digit_2_loss: 0.0490 - digit_3_loss: 0.1131 - digit_4_loss: 0.1709 - digit_1_acc: 1.0000 - digit_2_acc: 0.9884 - digit_3_acc: 0.9748 - digit_4_acc: 0.9690 - val_loss: 5.5489 - val_digit_1_loss: 4.0507e-06 - val_digit_2_loss: 0.6624 - val_digit_3_loss: 1.3039 - val_digit_4_loss: 2.0328 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8140 - val_digit_3_acc: 0.5756 - val_digit_4_acc: 0.4128\n",
      "Epoch 41/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 1.8180 - digit_1_loss: 1.4298e-06 - digit_2_loss: 0.0368 - digit_3_loss: 0.0936 - digit_4_loss: 0.1541 - digit_1_acc: 1.0000 - digit_2_acc: 0.9961 - digit_3_acc: 0.9884 - digit_4_acc: 0.9729 - val_loss: 5.5295 - val_digit_1_loss: 3.8768e-06 - val_digit_2_loss: 0.6822 - val_digit_3_loss: 1.2527 - val_digit_4_loss: 2.0829 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.7965 - val_digit_3_acc: 0.6279 - val_digit_4_acc: 0.4012\n",
      "Epoch 42/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 1.7572 - digit_1_loss: 1.3778e-06 - digit_2_loss: 0.0329 - digit_3_loss: 0.0900 - digit_4_loss: 0.1385 - digit_1_acc: 1.0000 - digit_2_acc: 0.9942 - digit_3_acc: 0.9922 - digit_4_acc: 0.9787 - val_loss: 5.3476 - val_digit_1_loss: 3.3196e-06 - val_digit_2_loss: 0.6019 - val_digit_3_loss: 1.3577 - val_digit_4_loss: 1.9148 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8372 - val_digit_3_acc: 0.5698 - val_digit_4_acc: 0.4186\n",
      "Epoch 43/50\n",
      "516/516 [==============================] - 0s 940us/step - loss: 1.7445 - digit_1_loss: 1.4032e-06 - digit_2_loss: 0.0349 - digit_3_loss: 0.0899 - digit_4_loss: 0.1635 - digit_1_acc: 1.0000 - digit_2_acc: 0.9903 - digit_3_acc: 0.9922 - digit_4_acc: 0.9593 - val_loss: 5.4477 - val_digit_1_loss: 4.0881e-06 - val_digit_2_loss: 0.6967 - val_digit_3_loss: 1.2652 - val_digit_4_loss: 2.0525 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8081 - val_digit_3_acc: 0.6047 - val_digit_4_acc: 0.3837\n",
      "Epoch 44/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 1.6870 - digit_1_loss: 1.3559e-06 - digit_2_loss: 0.0331 - digit_3_loss: 0.1019 - digit_4_loss: 0.1353 - digit_1_acc: 1.0000 - digit_2_acc: 0.9961 - digit_3_acc: 0.9845 - digit_4_acc: 0.9806 - val_loss: 5.5106 - val_digit_1_loss: 3.4645e-06 - val_digit_2_loss: 0.5912 - val_digit_3_loss: 1.3156 - val_digit_4_loss: 2.2092 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8372 - val_digit_3_acc: 0.5930 - val_digit_4_acc: 0.3721\n",
      "Epoch 45/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 1.6075 - digit_1_loss: 1.5536e-06 - digit_2_loss: 0.0260 - digit_3_loss: 0.0751 - digit_4_loss: 0.1279 - digit_1_acc: 1.0000 - digit_2_acc: 0.9961 - digit_3_acc: 0.9922 - digit_4_acc: 0.9748 - val_loss: 5.4769 - val_digit_1_loss: 3.3841e-06 - val_digit_2_loss: 0.6283 - val_digit_3_loss: 1.3254 - val_digit_4_loss: 2.1665 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8256 - val_digit_3_acc: 0.6163 - val_digit_4_acc: 0.3314\n",
      "Epoch 46/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 1.5527 - digit_1_loss: 1.3508e-06 - digit_2_loss: 0.0205 - digit_3_loss: 0.0745 - digit_4_loss: 0.1166 - digit_1_acc: 1.0000 - digit_2_acc: 1.0000 - digit_3_acc: 0.9942 - digit_4_acc: 0.9826 - val_loss: 5.2726 - val_digit_1_loss: 4.6154e-06 - val_digit_2_loss: 0.6049 - val_digit_3_loss: 1.3034 - val_digit_4_loss: 2.0441 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8430 - val_digit_3_acc: 0.6105 - val_digit_4_acc: 0.3837\n",
      "Epoch 47/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 1.5008 - digit_1_loss: 1.4023e-06 - digit_2_loss: 0.0296 - digit_3_loss: 0.0686 - digit_4_loss: 0.0978 - digit_1_acc: 1.0000 - digit_2_acc: 0.9942 - digit_3_acc: 0.9903 - digit_4_acc: 0.9806 - val_loss: 5.4098 - val_digit_1_loss: 3.8331e-06 - val_digit_2_loss: 0.6622 - val_digit_3_loss: 1.3498 - val_digit_4_loss: 2.1133 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8256 - val_digit_3_acc: 0.5698 - val_digit_4_acc: 0.3314\n",
      "Epoch 48/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 1.4763 - digit_1_loss: 1.3053e-06 - digit_2_loss: 0.0251 - digit_3_loss: 0.0668 - digit_4_loss: 0.1143 - digit_1_acc: 1.0000 - digit_2_acc: 0.9961 - digit_3_acc: 0.9922 - digit_4_acc: 0.9767 - val_loss: 5.3683 - val_digit_1_loss: 4.4699e-06 - val_digit_2_loss: 0.6094 - val_digit_3_loss: 1.5111 - val_digit_4_loss: 1.9965 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8372 - val_digit_3_acc: 0.5174 - val_digit_4_acc: 0.3895\n",
      "Epoch 49/50\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "516/516 [==============================] - 1s 1ms/step - loss: 1.4602 - digit_1_loss: 1.5513e-06 - digit_2_loss: 0.0199 - digit_3_loss: 0.0895 - digit_4_loss: 0.1122 - digit_1_acc: 1.0000 - digit_2_acc: 1.0000 - digit_3_acc: 0.9884 - digit_4_acc: 0.9864 - val_loss: 5.2803 - val_digit_1_loss: 4.3175e-06 - val_digit_2_loss: 0.6253 - val_digit_3_loss: 1.2202 - val_digit_4_loss: 2.2127 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8372 - val_digit_3_acc: 0.6337 - val_digit_4_acc: 0.3721\n",
      "Epoch 50/50\n",
      "516/516 [==============================] - 1s 1ms/step - loss: 1.4065 - digit_1_loss: 1.4596e-06 - digit_2_loss: 0.0246 - digit_3_loss: 0.0712 - digit_4_loss: 0.1008 - digit_1_acc: 1.0000 - digit_2_acc: 0.9942 - digit_3_acc: 0.9884 - digit_4_acc: 0.9864 - val_loss: 5.3520 - val_digit_1_loss: 3.4977e-06 - val_digit_2_loss: 0.6662 - val_digit_3_loss: 1.4047 - val_digit_4_loss: 2.0884 - val_digit_1_acc: 1.0000 - val_digit_2_acc: 0.8140 - val_digit_3_acc: 0.5523 - val_digit_4_acc: 0.3953\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "exact accuracy 0.0\n",
      "exact accuracy 0.0\n",
      "exact accuracy 0.0\n",
      "exact accuracy 0.0\n",
      "exact accuracy 0.0\n",
      "exact accuracy 0.0\n",
      "exact accuracy 0.0\n",
      "exact accuracy 0.0\n",
      "exact accuracy 0.0\n",
      "exact accuracy 0.0\n",
      "exact accuracy 0.0\n",
      "exact accuracy 0.0\n",
      "exact accuracy 0.0\n",
      "exact accuracy 0.0\n",
      "exact accuracy 0.005813953488372093\n",
      "exact accuracy 0.005813953488372093\n",
      "exact accuracy 0.005813953488372093\n",
      "exact accuracy 0.005813953488372093\n",
      "exact accuracy 0.005813953488372093\n",
      "exact accuracy 0.011627906976744186\n",
      "exact accuracy 0.01744186046511628\n",
      "exact accuracy 0.01744186046511628\n",
      "exact accuracy 0.01744186046511628\n",
      "exact accuracy 0.01744186046511628\n",
      "exact accuracy 0.01744186046511628\n",
      "exact accuracy 0.01744186046511628\n",
      "exact accuracy 0.01744186046511628\n",
      "exact accuracy 0.023255813953488372\n",
      "exact accuracy 0.023255813953488372\n",
      "exact accuracy 0.023255813953488372\n",
      "exact accuracy 0.023255813953488372\n",
      "exact accuracy 0.023255813953488372\n",
      "exact accuracy 0.023255813953488372\n",
      "exact accuracy 0.023255813953488372\n",
      "exact accuracy 0.023255813953488372\n",
      "exact accuracy 0.023255813953488372\n",
      "exact accuracy 0.023255813953488372\n",
      "exact accuracy 0.023255813953488372\n",
      "exact accuracy 0.023255813953488372\n",
      "exact accuracy 0.023255813953488372\n",
      "exact accuracy 0.029069767441860465\n",
      "exact accuracy 0.029069767441860465\n",
      "exact accuracy 0.029069767441860465\n",
      "exact accuracy 0.029069767441860465\n",
      "exact accuracy 0.029069767441860465\n",
      "exact accuracy 0.03488372093023256\n",
      "exact accuracy 0.03488372093023256\n",
      "exact accuracy 0.03488372093023256\n",
      "exact accuracy 0.03488372093023256\n",
      "exact accuracy 0.03488372093023256\n",
      "exact accuracy 0.03488372093023256\n",
      "exact accuracy 0.03488372093023256\n",
      "exact accuracy 0.03488372093023256\n",
      "exact accuracy 0.03488372093023256\n",
      "exact accuracy 0.03488372093023256\n",
      "exact accuracy 0.03488372093023256\n",
      "exact accuracy 0.03488372093023256\n",
      "exact accuracy 0.03488372093023256\n",
      "exact accuracy 0.03488372093023256\n",
      "exact accuracy 0.03488372093023256\n",
      "exact accuracy 0.040697674418604654\n",
      "exact accuracy 0.040697674418604654\n",
      "exact accuracy 0.040697674418604654\n",
      "exact accuracy 0.040697674418604654\n",
      "exact accuracy 0.040697674418604654\n",
      "exact accuracy 0.040697674418604654\n",
      "exact accuracy 0.040697674418604654\n",
      "exact accuracy 0.046511627906976744\n",
      "exact accuracy 0.05232558139534884\n",
      "exact accuracy 0.05232558139534884\n",
      "exact accuracy 0.05813953488372093\n",
      "exact accuracy 0.06395348837209303\n",
      "exact accuracy 0.06395348837209303\n",
      "exact accuracy 0.06976744186046512\n",
      "exact accuracy 0.06976744186046512\n",
      "exact accuracy 0.06976744186046512\n",
      "exact accuracy 0.06976744186046512\n",
      "exact accuracy 0.0755813953488372\n",
      "exact accuracy 0.08139534883720931\n",
      "exact accuracy 0.0872093023255814\n",
      "exact accuracy 0.09302325581395349\n",
      "exact accuracy 0.09302325581395349\n",
      "exact accuracy 0.09302325581395349\n",
      "exact accuracy 0.09302325581395349\n",
      "exact accuracy 0.09302325581395349\n",
      "exact accuracy 0.09302325581395349\n",
      "exact accuracy 0.09302325581395349\n",
      "exact accuracy 0.09302325581395349\n",
      "exact accuracy 0.09883720930232558\n",
      "exact accuracy 0.09883720930232558\n",
      "exact accuracy 0.09883720930232558\n",
      "exact accuracy 0.09883720930232558\n",
      "exact accuracy 0.10465116279069768\n",
      "exact accuracy 0.11046511627906977\n",
      "exact accuracy 0.11046511627906977\n",
      "exact accuracy 0.11627906976744186\n",
      "exact accuracy 0.11627906976744186\n",
      "exact accuracy 0.11627906976744186\n",
      "exact accuracy 0.12209302325581395\n",
      "exact accuracy 0.12209302325581395\n",
      "exact accuracy 0.12209302325581395\n",
      "exact accuracy 0.12790697674418605\n",
      "exact accuracy 0.12790697674418605\n",
      "exact accuracy 0.12790697674418605\n",
      "exact accuracy 0.12790697674418605\n",
      "exact accuracy 0.12790697674418605\n",
      "exact accuracy 0.12790697674418605\n",
      "exact accuracy 0.12790697674418605\n",
      "exact accuracy 0.12790697674418605\n",
      "exact accuracy 0.13372093023255813\n",
      "exact accuracy 0.13372093023255813\n",
      "exact accuracy 0.13372093023255813\n",
      "exact accuracy 0.13372093023255813\n",
      "exact accuracy 0.13372093023255813\n",
      "exact accuracy 0.13953488372093023\n",
      "exact accuracy 0.13953488372093023\n",
      "exact accuracy 0.13953488372093023\n",
      "exact accuracy 0.13953488372093023\n",
      "exact accuracy 0.13953488372093023\n",
      "exact accuracy 0.14534883720930233\n",
      "exact accuracy 0.1511627906976744\n",
      "exact accuracy 0.1511627906976744\n",
      "exact accuracy 0.1511627906976744\n",
      "exact accuracy 0.1511627906976744\n",
      "exact accuracy 0.1511627906976744\n",
      "exact accuracy 0.1511627906976744\n",
      "exact accuracy 0.1511627906976744\n",
      "exact accuracy 0.1511627906976744\n",
      "exact accuracy 0.1511627906976744\n",
      "exact accuracy 0.1511627906976744\n",
      "exact accuracy 0.1511627906976744\n",
      "exact accuracy 0.1511627906976744\n",
      "exact accuracy 0.1511627906976744\n",
      "exact accuracy 0.1511627906976744\n",
      "exact accuracy 0.1511627906976744\n",
      "exact accuracy 0.1569767441860465\n",
      "exact accuracy 0.1569767441860465\n",
      "exact accuracy 0.1569767441860465\n",
      "exact accuracy 0.1569767441860465\n",
      "exact accuracy 0.1569767441860465\n",
      "exact accuracy 0.1569767441860465\n",
      "exact accuracy 0.1569767441860465\n",
      "exact accuracy 0.1569767441860465\n",
      "exact accuracy 0.1569767441860465\n",
      "exact accuracy 0.16279069767441862\n",
      "exact accuracy 0.1686046511627907\n",
      "exact accuracy 0.1744186046511628\n",
      "exact accuracy 0.18023255813953487\n",
      "exact accuracy 0.18023255813953487\n",
      "exact accuracy 0.18023255813953487\n",
      "exact accuracy 0.18023255813953487\n",
      "exact accuracy 0.18023255813953487\n",
      "exact accuracy 0.18023255813953487\n",
      "exact accuracy 0.18604651162790697\n",
      "exact accuracy 0.18604651162790697\n",
      "exact accuracy 0.18604651162790697\n",
      "exact accuracy 0.18604651162790697\n",
      "exact accuracy 0.19186046511627908\n",
      "exact accuracy 0.19186046511627908\n",
      "exact accuracy 0.19186046511627908\n",
      "exact accuracy 0.19767441860465115\n",
      "exact accuracy 0.19767441860465115\n",
      "exact accuracy 0.20348837209302326\n",
      "exact accuracy 0.20348837209302326\n",
      "exact accuracy 0.20348837209302326\n",
      "exact accuracy 0.20930232558139536\n",
      "exact accuracy 0.20930232558139536\n",
      "exact accuracy 0.21511627906976744\n",
      "exact accuracy 0.21511627906976744\n",
      "exact accuracy 0.21511627906976744\n",
      "exact accuracy 0.21511627906976744\n",
      "exact accuracy 0.21511627906976744\n",
      "difference label vs. prediction [1, -200, -2, 68, -11, 55, 201, 20, -21, 60, -9, -900, -3, 64, 0, -44, 24, -900, -700, 0, 0, 5, -10, -10, 2, -4, 15, 0, -8, 869, 2, -6, -4, 614, -2, -871, -1, -60, -1, 9, 0, 9, 9, 30, 921, 0, -1, 4, 4, -30, -2, -9, -79, 37, 21, 47, 16, -740, 2, 46, 0, -840, -4, -883, -21, 60, -940, 0, 0, 31, 0, 0, -9, 0, -900, -1, 8, 0, 0, 0, 0, 60, 6, -315, -906, 19, 3, 20, 0, 65, 1, 20, 0, 0, 50, 0, 60, -935, 0, -20, -87, 0, 48, 5, -2, -843, -1, -6, 60, 0, 96, 1, -805, 821, 0, 40, 11, 4, -19, 0, 0, -20, -20, -906, 54, 7, 21, -7, 34, -81, -33, 25, -925, -906, -899, 0, -1, -843, 30, -10, -6, 21, 8, 21, 0, 0, 0, 0, -891, 5, 57, 4, 892, 0, 4, -3, -861, 0, 80, 3, 0, 2, 0, 60, -5, 0, 1, 0, -10, -952, -1, -6]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/aamomeni/Seven-Segment-OCR/Model.py:198: UserWarning: Update your `Dense` call to the Keras 2 API: `Dense(activation=\"softmax\", name=\"output\", units=11)`\n",
      "  output = Dense(output_dim =11,activation = 'softmax', name='output')(x)\n",
      "/home/aamomeni/Seven-Segment-OCR/Model.py:200: UserWarning: Update your `Model` call to the Keras 2 API: `Model(inputs=Tensor(\"in..., outputs=Tensor(\"ou...)`\n",
      "  self.model = keras.models.Model(input = model_input , output = output)\n",
      "/home/aamomeni/Seven-Segment-OCR/Model.py:207: UserWarning: The `nb_epoch` argument in `fit` has been renamed `epochs`.\n",
      "  self.history = self.model.fit(self.X_train, self.y_train, batch_size= 32, nb_epoch=30, verbose=1, validation_data=(self.X_val, self.y_val))\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 2064 samples, validate on 688 samples\n",
      "Epoch 1/30\n",
      "2064/2064 [==============================] - 7s 3ms/step - loss: 18.6020 - acc: 0.3953 - val_loss: 12.9582 - val_acc: 0.4084\n",
      "Epoch 2/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 11.5059 - acc: 0.4016 - val_loss: 10.2261 - val_acc: 0.4084\n",
      "Epoch 3/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 8.0329 - acc: 0.3915 - val_loss: 6.0187 - val_acc: 0.2907\n",
      "Epoch 4/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 2.6648 - acc: 0.4520 - val_loss: 2.3791 - val_acc: 0.4564\n",
      "Epoch 5/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 2.0756 - acc: 0.5586 - val_loss: 2.0950 - val_acc: 0.5160\n",
      "Epoch 6/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 1.7416 - acc: 0.6323 - val_loss: 1.9161 - val_acc: 0.5814\n",
      "Epoch 7/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 1.5996 - acc: 0.6662 - val_loss: 1.6231 - val_acc: 0.6279\n",
      "Epoch 8/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 1.4174 - acc: 0.7040 - val_loss: 1.5471 - val_acc: 0.6483\n",
      "Epoch 9/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 1.3284 - acc: 0.7166 - val_loss: 1.4834 - val_acc: 0.6802\n",
      "Epoch 10/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 1.2886 - acc: 0.7248 - val_loss: 1.3061 - val_acc: 0.7369\n",
      "Epoch 11/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 1.2240 - acc: 0.7524 - val_loss: 1.2535 - val_acc: 0.7442\n",
      "Epoch 12/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 1.1711 - acc: 0.7529 - val_loss: 1.3851 - val_acc: 0.6831\n",
      "Epoch 13/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 1.1794 - acc: 0.7573 - val_loss: 1.2351 - val_acc: 0.7573\n",
      "Epoch 14/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 1.1661 - acc: 0.7505 - val_loss: 1.2331 - val_acc: 0.7384\n",
      "Epoch 15/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 1.1294 - acc: 0.7607 - val_loss: 1.3475 - val_acc: 0.7035\n",
      "Epoch 16/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 1.1237 - acc: 0.7689 - val_loss: 1.2531 - val_acc: 0.7529\n",
      "Epoch 17/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 1.0853 - acc: 0.7796 - val_loss: 1.2607 - val_acc: 0.7413\n",
      "Epoch 18/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 1.0879 - acc: 0.7766 - val_loss: 1.1968 - val_acc: 0.7529\n",
      "Epoch 19/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 1.0821 - acc: 0.7776 - val_loss: 1.5113 - val_acc: 0.6933\n",
      "Epoch 20/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 1.0901 - acc: 0.7897 - val_loss: 1.2322 - val_acc: 0.7573\n",
      "Epoch 21/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 1.0461 - acc: 0.7926 - val_loss: 1.2671 - val_acc: 0.7369\n",
      "Epoch 22/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 1.0444 - acc: 0.7897 - val_loss: 1.3538 - val_acc: 0.7209\n",
      "Epoch 23/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 1.0445 - acc: 0.7951 - val_loss: 1.2873 - val_acc: 0.7485\n",
      "Epoch 24/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 1.0483 - acc: 0.7970 - val_loss: 1.2014 - val_acc: 0.7733\n",
      "Epoch 25/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 0.9891 - acc: 0.8096 - val_loss: 1.2766 - val_acc: 0.7253\n",
      "Epoch 26/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 0.9911 - acc: 0.8130 - val_loss: 1.2209 - val_acc: 0.7674\n",
      "Epoch 27/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 0.9701 - acc: 0.8183 - val_loss: 1.2232 - val_acc: 0.7674\n",
      "Epoch 28/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 0.9810 - acc: 0.8246 - val_loss: 1.2355 - val_acc: 0.7718\n",
      "Epoch 29/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 0.9823 - acc: 0.8222 - val_loss: 1.2908 - val_acc: 0.7427\n",
      "Epoch 30/30\n",
      "2064/2064 [==============================] - 2s 1ms/step - loss: 0.9240 - acc: 0.8348 - val_loss: 1.3209 - val_acc: 0.7485\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "real_acc 0.748546511627907\n"
     ]
    }
   ],
   "source": [
    "from keras.backend.tensorflow_backend import set_session\n",
    "from Model import Model_Multi, Model_Single\n",
    "import tensorflow as tf\n",
    "\n",
    "session_config = tf.ConfigProto()\n",
    "session_config.gpu_options.visible_device_list = \"0\"\n",
    "session_config.gpu_options.allow_growth = True\n",
    "set_session(tf.Session(config=session_config))\n",
    "\n",
    "model_1 = Model_Multi()\n",
    "model_1.train_predict()\n",
    "\n",
    "model_2 = Model_Single()\n",
    "model_2.train_predict()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}