4_0_random_weight_try_densenet.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%load_ext autotime"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "time: 1.9 s\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from matplotlib import pyplot as plt\n",
    "%matplotlib inline\n",
    "from keras.applications.densenet import DenseNet169\n",
    "from keras.utils import np_utils\n",
    "from keras.utils import to_categorical\n",
    "from keras.models import load_model\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "from keras import layers\n",
    "from keras import optimizers\n",
    "from keras import metrics\n",
    "from keras import models\n",
    "\n",
    "from keras.utils import plot_model\n",
    "from IPython.display import SVG\n",
    "from keras.utils.vis_utils import model_to_dot\n",
    "from keras_tqdm import TQDMNotebookCallback\n",
    "from ipywidgets import IntProgress, HTML, HBox\n",
    "\n",
    "from pathlib import Path\n",
    "import pickle\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "time: 709 µs\n"
     ]
    }
   ],
   "source": [
    "import logging\n",
    "\n",
    "log = logging.getLogger(__name__)\n",
    "logging.basicConfig(level=logging.INFO)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "time: 2.35 s\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "from keras import backend as K\n",
    "config = tf.ConfigProto()\n",
    "config.gpu_options.allow_growth=True\n",
    "sess = tf.Session(config=config)\n",
    "K.set_session(sess)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "time: 4.38 s\n"
     ]
    }
   ],
   "source": [
    "import utils # importing from our utils functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:utils.utils:loading data/MURA-v1.1/x_train.pkl\n",
      "INFO:utils.utils:loading data/MURA-v1.1/y_train.pkl\n",
      "INFO:utils.utils:loading data/MURA-v1.1/x_valid.pkl\n",
      "INFO:utils.utils:loading data/MURA-v1.1/y_valid.pkl\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "time: 15.8 s\n"
     ]
    }
   ],
   "source": [
    "x_train, y_train, x_test, y_test = utils.read_mura_pickle()\n",
    "x_train = utils.normalize_pixels(x_train)\n",
    "x_test = utils.normalize_pixels(x_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(36808, 256, 256)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "time: 2.23 ms\n"
     ]
    }
   ],
   "source": [
    "x_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "time: 923 µs\n"
     ]
    }
   ],
   "source": [
    "size = x_train.shape[1]\n",
    "x_train = x_train.reshape(x_train.shape[0], size, size, 1)\n",
    "x_test = x_test.reshape(x_test.shape[0], size, size, 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Train Submodel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "time: 1.44 ms\n"
     ]
    }
   ],
   "source": [
    "datagen = ImageDataGenerator(\n",
    "    rotation_range=45,\n",
    "    width_shift_range=0.2,\n",
    "    height_shift_range=0.2,\n",
    "    zoom_range=[1.0, 1.2],\n",
    "    fill_mode = 'constant',\n",
    "    cval=0,\n",
    "    horizontal_flip=True,\n",
    "    vertical_flip=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "time: 5.78 s\n"
     ]
    }
   ],
   "source": [
    "datagen.fit(x_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "time: 29.8 s\n"
     ]
    }
   ],
   "source": [
    "with tf.device('/GPU:2'):\n",
    "    densenet = DenseNet169(input_shape=(256, 256, 1), classes=1, weights=None, include_top=False)\n",
    "    for layer in densenet.layers:\n",
    "        if isinstance(layer, layers.Convolution2D):\n",
    "            layer.trainable = False\n",
    "    model = models.Model(input=densenet.input, output=layers.Dense(1, activation='sigmoid')(layers.Flatten()(densenet.output)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "time: 1.77 ms\n"
     ]
    }
   ],
   "source": [
    "starting_epoch = 0\n",
    "train_round = 0\n",
    "batch_size = 16\n",
    "epochs_per_eval = 10\n",
    "epochs_per_save = 30\n",
    "total_epochs = 90\n",
    "optimizer = 'adam'\n",
    "lr = 0.001\n",
    "train_history = {'loss': [], 'binary_accuracy': [], 'learn_rate': []}\n",
    "val_per_image_history = {'accuracy': [], 'kappa': [], 'precision': [], 'recall': []}\n",
    "val_per_study_history = {'accuracy': [], 'kappa': [], 'precision': [], 'recall': []}\n",
    "model_name='4_0_try_densenet169'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "time: 2.16 ms\n"
     ]
    }
   ],
   "source": [
    "def weighted_binary_crossentropy(p1):\n",
    "    '''\n",
    "    p1 is the proportion of ones in population\n",
    "    '''\n",
    "    w0 = K.variable(p1)\n",
    "        \n",
    "    def loss(y_true, y_pred):\n",
    "        # clip to prevent NaN's and Inf's\n",
    "        y_pred = K.clip(y_pred, K.epsilon(), 1 - K.epsilon())\n",
    "        # calc\n",
    "        loss = -(1 - w0) * y_true * K.log(y_pred) - w0 * (1 - y_true) * K.log(1 - y_pred)\n",
    "        loss = K.sum(loss, -1)\n",
    "        return loss\n",
    "    \n",
    "    return loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "time: 995 µs\n"
     ]
    }
   ],
   "source": [
    "p1 = y_train.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "time: 12.3 ms\n"
     ]
    }
   ],
   "source": [
    "# clr = utils.clr.CyclicLR(mode='triangular2')\n",
    "# opt = optimizers.SGD(nesterov=True)\n",
    "opt = optimizers.Adam()"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "with tf.device('/GPU:0'):\n",
    "    model = models.load_model(f'./trained_models/{model_name}_{starting_epoch}.h5')"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "with open(f'./trained_models/{model_name}_train_history.pkl', 'rb') as pkl_file:\n",
    "    train_history = pickle.load(pkl_file)\n",
    "with open(f'./trained_models/{model_name}_val_per_image_history.pkl', 'rb') as pkl_file:\n",
    "    val_per_image_history = pickle.load(pkl_file)\n",
    "with open(f'./trained_models/{model_name}_val_per_study_history.pkl', 'rb') as pkl_file:\n",
    "    val_per_study_history = pickle.load(pkl_file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "time: 895 µs\n"
     ]
    }
   ],
   "source": [
    "progress = TQDMNotebookCallback(leave_inner=True, leave_outer=True) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:__main__:executing epochs 1-10\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c97389471b444323a2fafe4ff25cc654",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Training', max=10), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ed40ab841b5b427882b984592c7e0192",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 0', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "037504d9e554401d9cbfae7b3fce4285",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 1', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d2cc4cc05dbc4b7985c706877ec3dd3a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 2', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7a0bb75f1bb54b8db0bffb8f453ae6a1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 3', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cd4ef990b87a4e7986325873b8bbfc55",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 4', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0a9d49fdc241454392aff93dcd9ab3af",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 5', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b26f44cc25624347bff3e49c365056c4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 6', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2227772b27c04f18a3660c72aed0c4fd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 7', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7cedcf4d09d144cb933c0aa48cd4ef85",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 8', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e9d7d38a5c0c4dc8ab86af57fc3152ed",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 9', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:utils.mura_metrics:Accuracy: 0.6183922427275571\n",
      "INFO:utils.mura_metrics:Kappa: 0.22619456216945025\n",
      "INFO:utils.mura_metrics:Precision: 0.6424632352941176\n",
      "INFO:utils.mura_metrics:Recall: 0.4568627450980392\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:utils.mura_metrics:Accuracy: 0.640533778148457\n",
      "INFO:utils.mura_metrics:Kappa: 0.2434600311242349\n",
      "INFO:utils.mura_metrics:Precision: 0.6731391585760518\n",
      "INFO:utils.mura_metrics:Recall: 0.38661710037174724\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:__main__:Saving evaluation metrics history...\n",
      "INFO:__main__:executing epochs 11-20\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f32aa0c43c754626a95d08c77ebf9c37",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Training', max=10), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0b198cb747bb4a1fa487424bfbdd052e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 0', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "61151f0942ca446b88cb14bfd998d4c6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 1', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "97e6c22478b94c5e907932f8d0b111d7",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 2', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7cd70ebeac904f40b0e82078814fa2bd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 3', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5261bddab373463786f62a03a70432cb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 4', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ab357f7891c342b5b0b2779e98496892",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 5', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cc08b525bf1c494c906e6efdd37efaf6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 6', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "da8df4bd1fe9445f998cab222771c64a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 7', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bbd610a941974743bfb38f21bc894083",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 8', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e80edab5829d4a79a43917ae6fee930c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 9', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:utils.mura_metrics:Accuracy: 0.6359086643728495\n",
      "INFO:utils.mura_metrics:Kappa: 0.26996789380118486\n",
      "INFO:utils.mura_metrics:Precision: 0.6216755319148937\n",
      "INFO:utils.mura_metrics:Recall: 0.6111111111111112\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:utils.mura_metrics:Accuracy: 0.6705587989991659\n",
      "INFO:utils.mura_metrics:Kappa: 0.3260680555931064\n",
      "INFO:utils.mura_metrics:Precision: 0.652452025586354\n",
      "INFO:utils.mura_metrics:Recall: 0.5687732342007435\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:__main__:Saving evaluation metrics history...\n",
      "INFO:__main__:executing epochs 21-30\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d59d24bf57754c218a013ffb205def84",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Training', max=10), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "eaceb1c40f1f44aba92126dec231379f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 0', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2d22474676e84cac809448c1c927f290",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 1', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0becb8daec72485391fd619288f9243a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 2', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b7868d2b03de43a0b2735d566602ced5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 3', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "18c907a6b7354f72894e81367dd4018b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 4', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0dffc3a12a3d43899a7a63df63e54342",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 5', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2d6abda05fd84c3785d3be897d38ff24",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 6', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1f95de0efd7a406ab5b6f83a6ba62f10",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 7', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "76f977fce64b4e28813baa6060252f9f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 8', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "94c07ed42895449a81dbf8e9387c722d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 9', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:utils.mura_metrics:Accuracy: 0.6371598373475132\n",
      "INFO:utils.mura_metrics:Kappa: 0.2797933055803919\n",
      "INFO:utils.mura_metrics:Precision: 0.5982996811902231\n",
      "INFO:utils.mura_metrics:Recall: 0.7359477124183007\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:utils.mura_metrics:Accuracy: 0.6538782318598833\n",
      "INFO:utils.mura_metrics:Kappa: 0.31123335303538485\n",
      "INFO:utils.mura_metrics:Precision: 0.5977742448330684\n",
      "INFO:utils.mura_metrics:Recall: 0.6988847583643123\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:__main__:Saving Model to trained_models/4_0_try_densenet169_30.h5...\n",
      "INFO:__main__:Saving evaluation metrics history...\n",
      "INFO:__main__:executing epochs 31-40\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "62d00662bb394e00baaa4dda8fa4a3bd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Training', max=10), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ad391513eca241afbaec97d3e9395ca6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 0', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "19c1221c8fcd4e719a4a9a2f93f91778",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 1', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ccfc619a06d9442ab434686863018c5f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 2', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "feeb95c2e94f471c80b3b97cda9f2b94",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 3', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c5de3558ba7c46d6b74c3f449d96c5a5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 4', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bd0c00115071446ba703daa0af048ad2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 5', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4ddbdee731ce4caaa7e143bb5a490d54",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 6', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0ca420136463478a9c60fda108731693",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 7', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4aee43ab349e481da614807d4c43a23e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 8', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ddf7bee909cf4cb69e7e0f4cf016231e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 9', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:utils.mura_metrics:Accuracy: 0.653112292774476\n",
      "INFO:utils.mura_metrics:Kappa: 0.31348501515456323\n",
      "INFO:utils.mura_metrics:Precision: 0.6056196688409433\n",
      "INFO:utils.mura_metrics:Recall: 0.7888888888888889\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:utils.mura_metrics:Accuracy: 0.6713928273561302\n",
      "INFO:utils.mura_metrics:Kappa: 0.34929944517600664\n",
      "INFO:utils.mura_metrics:Precision: 0.6094224924012158\n",
      "INFO:utils.mura_metrics:Recall: 0.7453531598513011\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUsAAAEmCAYAAADr3bIaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3Xm8VVXdx/HP914ElNkJEUTNHFJKEIfUVNIy9cEwG5xyytIGy9J8ciq1smzS1EbNnDKHBpXIR7OyzFIUFBGcwBwAUUAGGRQVfs8fe1084L3n7APnsO+59/vmtV+cvfY6e/8uyM+119prbUUEZmZWXlPRAZiZNQInSzOzHJwszcxycLI0M8vBydLMLAcnSzOzHJwsOxFJ60r6k6QFkn63Buc5StJfahlbUSTtJenJouOw9k9+zrL9kXQkcCqwHbAQmABcEBH3ruF5jwa+COwREW+ucaDtnKQAto6IqUXHYo3PLct2RtKpwI+B7wD9gcHAz4BRNTj95sBTnSFR5iGpS9ExWAOJCG/tZAP6AIuAj5ep040smb6Qth8D3dKxEcB04DRgFjATOD4dOx94HXgjXeME4DzgNyXn3gIIoEvaPw74L1nr9hngqJLye0u+twfwILAg/b5HybF/AN8C/p3O8xdgwzZ+tpb4/7ck/kOAg4CngLnAWSX1dwXuA+anuj8BuqZj96SfZXH6eQ8rOf/XgBeB61rK0ne2StfYKe1vCswGRhT934a34je3LNuX3YHuwC1l6pwNvBcYCuxIljDOKTm+CVnSHUiWEH8qqV9EnEvWWr0pInpGxJXlApHUA7gUODAiepElxAmt1Fsf+HOquwFwEfBnSRuUVDsSOB7YGOgKfLXMpTch+zMYCHwDuAL4JDAc2Av4uqQtU91lwFeADcn+7PYDPg8QEXunOjumn/emkvOvT9bKPrH0whHxNFki/Y2k9YCrgGsi4h9l4rVOwsmyfdkAmBPlb5OPAr4ZEbMiYjZZi/HokuNvpONvRMTtZK2qbVcznuXAEEnrRsTMiJjcSp3/AaZExHUR8WZE3AA8ARxcUueqiHgqIl4FbiZL9G15g6x/9g3gRrJEeElELEzXf4zsfxJExPiIuD9d91ngl8A+OX6mcyNiaYpnJRFxBTAVGAsMIPufk5mTZTvzMrBhhb60TYHnSvafS2UrzrFKsl0C9Kw2kIhYTHbr+llgpqQ/S9ouRzwtMQ0s2X+xinhejohl6XNLMnup5PirLd+XtI2kMZJelPQKWct5wzLnBpgdEa9VqHMFMAS4LCKWVqhrnYSTZftyH7CUrJ+uLS+Q3UK2GJzKVsdiYL2S/U1KD0bEnRHxQbIW1hNkSaRSPC0xzVjNmKrxc7K4to6I3sBZgCp8p+zjH5J6kvUDXwmcl7oZzJws25OIWEDWT/dTSYdIWk/SOpIOlPT9VO0G4BxJG0naMNX/zWpecgKwt6TBkvoAZ7YckNRf0qjUd7mU7HZ+eSvnuB3YRtKRkrpIOgzYHhizmjFVoxfwCrAotXo/t8rxl4B3VHnOS4BxEfFpsr7YX6xxlNYhOFm2MxHxI7JnLM8hG4mdBpwM3JqqfBsYB0wEHgUeSmWrc627gJvSucazcoJrSnG8QDZCvA9vT0ZExMvASLIR+JfJRrJHRsSc1YmpSl8lGzxaSNbqvWmV4+cB10iaL+kTlU4maRRwAG/9nKcCO0k6qmYRW8PyQ+lmZjm4ZWlmloOTpZlZDk6WZmY5OFmameXQrhYSUJd1Q117FR2G1cgmg/oXHYLVyPyXZrBkwdxKz7BWpbn35hFvvm0SVZvi1dl3RsQBtYyhGu0rWXbtRbdtKz7hYQ3ixAtPKToEq5HLv3hozc8Zb75a1b/31yb8tNLsLCQ1kz1aNyMiRqZ1BG4km0o8Hjg6Il6X1A24lmzNgZeBw9KU2Tb5NtzMCiJQU/4tn1OAx0v2vwdcHBHvBOaRLS5D+n1eKr841SvLydLMiiFAyr9VOp00iGxhl1+lfQH7Ar9PVa7hranEo9I+6fh+qX6b2tVtuJl1MvlbjJAtMjOuZP/yiLi8ZP/HZDPIWgY+NgDmlywsM523FngZSDY7joh4U9KCVL/NmWdOlmZWEEFTczVfmBMRO7d6JmkkMCsixksaUYvoVuVkaWbFyXF7ndOewIclHUS2eHRvskVR+krqklqXg3hrNawZwGbA9LQkYh+ygZ42uc/SzIohajbAExFnRsSgiNgCOBz4e0QcBdwNfCxVOxa4LX0enfZJx/8eFRbKcLI0s4JUMbiz+i3QrwGnSppK1ifZ8jqVK4ENUvmpwBmVTuTbcDMrTnUDPLmkdyb9I33+L9l7qlat8xrw8WrO62RpZsWpXZ9l3TlZmllBVJeWZb04WZpZMVoeSm8QTpZmVhy3LM3MKhE0V/VQeqGcLM2sGC3PWTYIJ0szK477LM3MKvFouJlZPm5Zmpnl4JalmVkFazbne61zsjSz4rhlaWaWg1uWZmaVeDTczKwyUe1rJQrlZGlmBXHL0swsH/dZmpnl4JalmVkOblmamVUg91mameXjlqWZWWVysjQzKy97BY+TpZlZeRJqcrI0M6vILUszsxycLM3McnCyNDOrRGlrEE6WZlYIIbcszczycLI0M8vBydLMLAcnSzOzSjzAY2ZWmRBNTY2z6lDjRGpmHY6k3FuF83SX9ICkRyRNlnR+Kr9a0jOSJqRtaCqXpEslTZU0UdJOlWJ1y9LMilO72/ClwL4RsUjSOsC9kv4vHTs9In6/Sv0Dga3Tthvw8/R7m5wszawYqt0AT0QEsCjtrpO2KPOVUcC16Xv3S+oraUBEzGzrC74NN7PCVHkbvqGkcSXbiaucq1nSBGAWcFdEjE2HLki32hdL6pbKBgLTSr4+PZW1yS1LMytMlS3LORGxc1sHI2IZMFRSX+AWSUOAM4EXga7A5cDXgG+uTqxuWZpZIVqmO9ZigKdURMwH7gYOiIiZkVkKXAXsmqrNADYr+dqgVNYmJ0szK46q2MqdRtootSiRtC7wQeAJSQNSmYBDgEnpK6OBY9Ko+HuBBeX6K8G34XXT1CT+ff3/8sKsBXz0lF+w+aYbcN2Fx7N+nx48/PjzfOqca3njzWUr6h+y31Bu+OGn2fOo7/PQY88XGLmV6t29Cx95d396dm0mgPHTXmHs8/MZsdX67DSoD0tez/4O/zZlDlPmLOHdA3qx5xb9Vny/f6+u/PK+53lx4esF/QTtWA0HeIABwDWSmskagTdHxBhJf5e0UXY1JgCfTfVvBw4CpgJLgOMrXcDJsk5OPvL9PPnMS/Tq0R2AC04ZxWXX383v7hzPpWcfznEf2Z0rfncvAD3X68YXjhzBAxOfKTJka8Xy5cFfnpjDzIVL6dosTtp9MP99eQkA9z83j/88O3+l+o/OXMijMxcCsHHPrhw+bIATZRk1HA2fCAxrpXzfNuoH8IVqruHb8DoYuHFfDnjfDlx1y39WlO2zyzb88a8PA3D9n8Zy8IgdVxw79/Mj+dFVd/Ha62+u9VitvEWvL2PmwqUAvL4smL34dXp1z9fGePeAXkyauahyxU5MTcq9Fc3Jsg5+cPpHOfuSW1m+PHvMa4O+PViw8FWWLVsOwIyX5rHpxn0AGLrdIAZt0o877p1cWLyWT9/uXRjQqxsz5r8GwK6D+/K5PQYzaoeN6d7l7f+UdtikJ5NeXLi2w2wo9RjgqZe6JktJB0h6Mk0pOqOe12ovDtxrCLPmLuThx6dVrCuJ7532Ub72oz+uhchsTXRtFp8YOoA7npjN0mXLeXDaAi6551l+8Z/nWbh0GR/adsOV6g/s0403lgWzFvkWvC3VJMr2kCzr1meZOlp/SjYqNR14UNLoiHisXtdsD3Yf+g5G7vNuDnjfDnTrug69e3Tnh6d/jD691qW5uYlly5YzsH8/Xpi1gF49urH9VgP4y69OAaD/Br35/Y9P4mNf/qUHedqRJsEnhg7g0ZkLeXzWYgAWv/7W4NxD0xdw5E6brvSdIZv0YtJMtyoraQ9JMK96DvDsCkyNiP8CSLqRbIpRh06W37hsNN+4bDQAew3fmi8fsx/Hn30N13//Uxz6gWH87s7xHHXwboz5x0ReWfQam+37VoP7zitO4cyLb3GibGdG7dCfOYtf577n3hrM6dm1mUUpYW7Xv+dKLUgBO2zSi6seqHx30dk5WWZam070tonqacpSNm1pnZ51DKdYZ19yG9ddeDznfn4kjzw5jatvva/okCyHwX27s+PA3ry0cCmf3X0wkD0mNGRALzbplc2cm//qG/xp8qwV39m837q88tobzHvVA3YVNU6uLP7RoYi4nGwaEk3rbVxu4nvD+df4Kfxr/BQAnp3xMnsd/cOy9T/0mUvWRlhWhefnv8Z5d055W/mUOUva/M6z817lV2On1zOsDsMty0zV04nMrBOp7UPpdVfP0fAHga0lbSmpK3A42RQjM7NsFqPyb0WrW8syIt6UdDJwJ9AM/Doi/DChmSWiqR08bJ5XXfssI+J2sjmYZmZv00i34YUP8JhZJ9VObq/zcrI0s0IIfBtuZpaHW5ZmZjm4z9LMrBL3WZqZVZY9Z9k42dLJ0swK0j6WXsvLydLMCtNAudLJ0swKIj86ZGZWkfsszcxyaqBc6WRpZsVxy9LMLIcGypVOlmZWkAZb/NfJ0swK0bL4b6NwsjSzgvihdDOzXBooVzpZmllB/FC6mVllfijdzCwnJ0szsxwaKFc6WZpZcdyyNDOrpMFWSm8qOgAz65yUnrPMu5U9l9Rd0gOSHpE0WdL5qXxLSWMlTZV0k6Suqbxb2p+ajm9RKV4nSzMrjJR/q2ApsG9E7AgMBQ6Q9F7ge8DFEfFOYB5wQqp/AjAvlV+c6pXlZGlmhWmScm/lRGZR2l0nbQHsC/w+lV8DHJI+j0r7pOP7qULz1cnSzApTw5YlkpolTQBmAXcBTwPzI+LNVGU6MDB9HghMA0jHFwAblDu/B3jMrBASNFc3g2dDSeNK9i+PiMtbdiJiGTBUUl/gFmC72kSacbI0s8JU+ejQnIjYuVKliJgv6W5gd6CvpC6p9TgImJGqzQA2A6ZL6gL0AV4ud942b8Ml9S635fzhzMzaVKvbcEkbpRYlktYFPgg8DtwNfCxVOxa4LX0enfZJx/8eEVHuGuValpPJOkhLw2zZD2Bw+fDNzNomsseHamQAcI2kZrJG4M0RMUbSY8CNkr4NPAxcmepfCVwnaSowFzi80gXaTJYRsdmaRm9mVk6tFh2KiInAsFbK/wvs2kr5a8DHq7lGrtFwSYdLOit9HiRpeDUXMTN7myoeSG8P0yIrJktJPwHeDxydipYAv6hnUGbWOdTy0aF6yzMavkdE7CTpYYCImNsyZcjMbHUJKj5s3p7kSZZvSGoiG9RB0gbA8rpGZWadQgPlylx9lj8F/gBslCan30uOeZRmZpU0Up9lxZZlRFwraTzwgVT08YiYVN+wzKyjW40ZPIXKO4OnGXiD7Fbc88nNrCYaJ1XmGw0/G7gB2JRsutBvJZ1Z78DMrOPrULfhwDHAsIhYAiDpArIn4b9bz8DMrGPLRsOLjiK/PMly5ir1uqQyM7PV105ajHm1mSwlXUzWRzkXmCzpzrS/P/Dg2gnPzDqyBsqVZVuWLSPek4E/l5TfX79wzKwz6RAty4i4sq1jZmZrqsP1WUraCrgA2B7o3lIeEdvUMS4z6wQaqWWZ55nJq4GryP5HcCBwM3BTHWMys05AgmYp91a0PMlyvYi4EyAino6Ic8iSppnZGuloqw4tTQtpPC3ps2TvruhV37DMrDNopNvwPMnyK0AP4EtkfZd9gE/VMygz6xwaKFfmWkhjbPq4kLcWADYzWyNCHWM9S0m3kNawbE1EHFqXiMysc2gnfZF5lWtZ/mStRZEMe9dg/j12rV/W6qTfbqcUHYLVyNIZs+py3g7RZxkRf1ubgZhZ59NI6z3mXc/SzKymRAdpWZqZ1VuHmu7YQlK3iFhaz2DMrPNotNdK5FkpfVdJjwJT0v6Oki6re2Rm1uE1Kf9WtDz9q5cCI4GXASLiEeD99QzKzDqHjjbdsSkinlulI3ZZneIxs04iW6KtHWTBnPIky2mSdgVCUjPwReCp+oZlZp1BR3t06HNkt+KDgZeAv6YyM7M10kANy1xzw2cBh6+FWMysE5E6yNzwFpKuoJU54hFxYl0iMrNOo4FyZa7b8L+WfO4OfASYVp9wzKwzaQ+PBOWV5zZ8pVdISLoOuLduEZlZpyAa66H01ZnuuCXQv9aBmFkn004eNs8rzwyeeZLmpm0+cBdwZv1DM7OOTlX8KnseaTNJd0t6TNJkSaek8vMkzZA0IW0HlXznTElTJT0p6UOVYi3bslT2JPqOZO/dAVgeEW0uCGxmlleN3xv+JnBaRDwkqRcwXtJd6djFEfHDla4tbU/2lM8OwKbAXyVtExFtTrgp27JMifH2iFiWNidKM6uZWs0Nj4iZEfFQ+rwQeBwYWOYro4AbI2JpRDwDTAV2LRtrjp9ngqRhOeqZmVVFUu4N2FDSuJKt1ccXJW0BDANa3h92sqSJkn4tqV8qG8jKT/VMp3xyLfsOni4R8Wa66IOSngYWk7WeIyJ2qvQHYWbWltW4DZ8TETuXPafUE/gD8OWIeEXSz4FvkT0r/i3gR6zm22nL9Vk+AOwEfHh1TmxmVlaNVxOStA5Zorw+Iv4IEBEvlRy/AhiTdmcAm5V8fRBvjc20qlyyVLrY09WHbWZWWa2mO6bB6CuBxyPiopLyARExM+1+BJiUPo8GfivpIrIBnq3JGohtKpcsN5J0alsHSwMyM6tWjUfD9wSOBh6VNCGVnQUcIWko2W34s8BJABExWdLNwGNkI+lfKDcSDuWTZTPQEyo84GRmtlpEc41alhFxL63nqtvLfOcC4IK81yiXLGdGxDfznsjMrBrZ2x2LjiK/in2WZmZ10WDTHcsly/3WWhRm1il1iPUsI2Lu2gzEzDqXjnQbbmZWVx2iZWlmVm8NlCudLM2sGKLjvd3RzKz2RMsCGQ3BydLMCtM4qdLJ0swKIqjZDJ61wcnSzArTQLnSydLMiiL3WZqZVeLRcDOznNyyNDPLoXFSpZOlmRXFz1mamVXmPkszs5zcsjQzy6GjLP5rZlY32W1442RLJ0szK0wD3YU7WZpZUYTcsjQzq8wtSzOzCtxnaWaWh9yyNDPLxcnSzCyHRhrgaaTZRg3hpE9/isGbbszwoUNWlJ1/7tfZZdh72G34UEYeuD8vvPACABf96AfsNnwouw0fyvChQ+jRrZm5c/269vamqUncd/3p/OHHJwKw+abrc881X2HSredw3XePZZ0uzQDsOWwr/nP9V1k49iI+st+ORYbcEET2UHrerWhOljV29LHHcduYO1Yq+8ppp/PgwxMZO34CBx40ku9++5sAnHra6YwdP4Gx4yfwzW9/l7323of111+/iLCtjJOP2Icnn31pxf4FX/owl13/D4Yc8m3mvfIqxx3yXgCmvTiPE8/9LTfdMb6oUBtOk5R7K5qTZY29b6+935bwevfuveLzkiWLW50Pe/NNN/CJw46oe3xWnYEb9+GA9+3AVbfet6Jsn1225o9/ewSA68c8wMEj3g3A8zPnMmnqCyyPKCTWRqQqfhXNfZZryblfP5vrf3Mtffr04Y677l7p2JIlS7jrzju4+JKfFBSdteUHpx3K2ZfcRs8e3QHYoG8PFix8lWXLlgMwY9Z8Nt2ob5EhNqyW2/BGUbeWpaRfS5olaVK9rtFIzv/WBUx9ZhqHH3EUv/jZyknxz2P+xO577Olb8HbmwL12YNa8RTz8xPSiQ+mgqmlXFp9V63kbfjVwQB3P35AOO+Iobr3lDyuV/e7mG/m4b8Hbnd133JKRew/hiT99g2u/cywjdtmaH371UPr0Wpfm5uyfzsCN+/LC7PkFR9qg0nOWebei1S1ZRsQ9gId2galTpqz4PGb0bWyz7XYr9hcsWMC99/yTgz88qojQrIxv/GQM7zzoXLY7+Jscc9Y1/OPBKRx/znXcM24Kh6bR7qNG7sqYf/rmaXWpiq1ohQ/wSDpR0jhJ42bPmV10OGvsmE8ewYi9duepJ59kqy0GcfWvr+Scs89g+NAh7DLsPfztr3/hhxddsqL+6FtvYb8P7k+PHj0KjNqqcfalf+JLR72fSbeewwZ91+PqNPgzfPvBTL39fA79wFAuO+swxt98RsGRtm9Zn2VtRsMlbSbpbkmPSZos6ZRUvr6kuyRNSb/3S+WSdKmkqZImStqpYrxRx5E7SVsAYyJiSIWqAAwfvnP8e+y4usVja1e/3U4pOgSrkaVP3MjyxS/VtIH3rncPi6tuubtyxWT3rfuNj4idWzsmaQAwICIektQLGA8cAhwHzI2ICyWdAfSLiK9JOgj4InAQsBtwSUTsVu76hbcszawTq9F9eETMjIiH0ueFwOPAQGAUcE2qdg1ZAiWVXxuZ+4G+KeG2yY8OmVlh6vGwebqjHQaMBfpHxMx06EWgf/o8EJhW8rXpqWwmbajno0M3APcB20qaLumEel3LzBpTlQ3LDVvGN9J24tvOJ/UE/gB8OSJeKT0WWZ/javc71q1lGRF+FsbMyquuYTmnrT5LAEnrkCXK6yPij6n4JUkDImJmus2elcpnAJuVfH1QKmuT+yzNrBBZi7E2D6Urm0N8JfB4RFxUcmg0cGz6fCxwW0n5MWlU/L3AgpLb9Va5z9LMilHbh833BI4GHpU0IZWdBVwI3Jy6AZ8DPpGO3U42Ej4VWAIcX+kCTpZmVpha5cqIuLfM6fZrpX4AX6jmGk6WZlac9jA1JycnSzMrSPtYICMvJ0szK0x7WCAjLydLMytEe1kgIy8nSzMrTGtvDWivnCzNrDANlCudLM2sOA2UK50szawgDdZp6WRpZoXxo0NmZhUI91mameXSQLnSydLMCtRA2dLJ0swK4z5LM7McmhonVzpZmlmBnCzNzMprWSm9UThZmlkxartSet05WZpZYRooVzpZmlmBGihbOlmaWUG8UrqZWS7uszQzq6DBFh1ysjSzAjVQtnSyNLPCNDXQfbiTpZkVpnFSpZOlmRXFD6WbmeXVONnSydLMCuGV0s3McmqgXOlkaWbFccvSzCwHT3c0M8ujcXKlk6WZFaeBcqWTpZkVQ/IMHjOzfBonV9JUdABm1nmpiq3iuaRfS5olaVJJ2XmSZkiakLaDSo6dKWmqpCclfajS+Z0szawwUv4th6uBA1opvzgihqbt9uy62h44HNghfednkprLndzJ0swKoqp+VRIR9wBzc158FHBjRCyNiGeAqcCu5b7gZGlmhWiZ7lhFy3JDSeNKthNzXupkSRPTbXq/VDYQmFZSZ3oqa5OTpZk1ijkRsXPJdnmO7/wc2AoYCswEfrS6F/douJkVpt5PDkXES29dS1cAY9LuDGCzkqqDUlmb3LI0s8LUss+y1fNLA0p2PwK0jJSPBg6X1E3SlsDWwAPlzuWWpZkVInsovZbn0w3ACLK+zenAucAISUOBAJ4FTgKIiMmSbgYeA94EvhARy8qd38nSzIpTw2QZEUe0UnxlmfoXABfkPb+TpZkVxqsOmZnl0EBTw50szaw4DZQrnSzNrEANlC2dLM2sMI3UZ6mIKDqGFSTNBp4rOo61YENgTtFBWE10lr/LzSNio1qeUNIdZH9+ec2JiNYWylgr2lWy7CwkjYuInYuOw9ac/y47D8/gMTPLwcnSzCwHJ8ti5FktxRqD/y47CfdZmpnl4JalmVkOTpZmZjk4WZqZ5eBkuRZI2lbS7pLWqfQGOWsM/nvsfDzAU2eSDgW+Q7Zk/QxgHHB1RLxSaGC2WiRtExFPpc/NlRaMtY7DLcs6krQOcBhwQkTsB9xG9t6Pr0nqXWhwVjVJI4EJkn4LEBHL3MLsPJws66832fs9AG4he2HSOsCRUiOt5te5SeoBnAx8GXhd0m/ACbMzcbKso4h4A7gIOFTSXhGxHLgXmAC8r9DgrCoRsRj4FPBb4KtA99KEWWRstnY4Wdbfv4C/AEdL2jsilkXEb4FNgR2LDc2qEREvRMSiiJhD9uKrdVsSpqSdJG1XbIRWT17Pss4i4jVJ15O9Xe7M9A9qKdCf7KXv1oAi4mVJJwE/kPQE0Ay8v+CwrI6cLNeCiJiXXvD+GFmL5DXgk6UvgLfGExFzJE0EDgQ+GBHTi47J6sePDq1laTAgUv+lNTBJ/YCbgdMiYmLR8Vh9OVmarQFJ3SPitaLjsPpzsjQzy8Gj4WZmOThZmpnl4GRpZpaDk6WZWQ5Olh2EpGWSJkiaJOl3ktZbg3ONkDQmff6wpDPK1O0r6fOrcY3zJH01b/kqda6W9LEqrrWFpEnVxmhWysmy43g1IoZGxBDgdeCzpQeVqfrvOyJGR8SFZar0BapOlmaNxsmyY/oX8M7UonpS0rXAJGAzSftLuk/SQ6kF2hNA0gGSnpD0EHBoy4kkHSfpJ+lzf0m3SHokbXsAFwJbpVbtD1K90yU9KGmipPNLznW2pKck3QtsW+mHkPSZdJ5HJP1hldbyBySNS+cbmeo3S/pBybVPWtM/SLMWTpYdjKQuZNPvHk1FWwM/i4gdgMXAOcAHImInsoWIT5XUHbgCOBgYDmzSxukvBf4ZETsCOwGTgTOAp1Or9nRJ+6dr7goMBYZL2lvScODwVHYQsEuOH+ePEbFLut7jwAklx7ZI1/gf4BfpZzgBWBARu6Tzf0bSljmuY1aR54Z3HOtKmpA+/wu4kmxlo+ci4v5U/l5ge+DfaSnNrsB9wHbAMxExBSCtpHNiK9fYFzgGVixLtiBN+Su1f9oeTvs9yZJnL+CWiFiSrjE6x880RNK3yW71ewJ3lhy7OU0ZnSLpv+ln2B94T0l/Zp907adyXMusLCfLjuPViBhaWpAS4uLSIuCuiDhilXorfW8NCfhuRPxylWt8eTXOdTVwSEQ8Iuk4YETJsVWnnkW69hcjojSpImmL1bi22Up8G9653A/sKemdkK3+LWkb4AlgC0lbpXpHtPH9vwGfS99tltQHWEjWamxxJ/Cpkr7QgZI2Bu4BDpG0rqReZLf8lfQCZqbXcxy1yrGPS2pKMb8DeDJd+3OpPpK2SSucm63Io0CSAAAAoklEQVQxtyw7kYiYnVpoN0jqlorPiYinJJ0I/FnSErLb+F6tnOIU4HJJJwDLgM9FxH2S/p0ezfm/1G/5LuC+1LJdRLYc3UOSbgIeAWYBD+YI+evAWGB2+r00pueBB8he2/HZtG7or8j6Mh9SdvHZwCH5/nTMyvNCGmZmOfg23MwsBydLM7McnCzNzHJwsjQzy8HJ0swsBydLM7McnCzNzHL4fyRGxkfaJmnIAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:__main__:Saving evaluation metrics history...\n",
      "INFO:__main__:executing epochs 41-50\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fbd7b5b37b514cdea16a47429e151b9b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Training', max=10), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "788c17022883440e976c4d445356ab8d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 0', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "33957387adea4d3a89d67a9b0f38bcc8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 1', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a2395b4ac30840bf834a06f28cfa2009",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 2', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "89b50af2cebe4d70850da2fd23e5ff12",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 3', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "162a044d0ff14804a27c4a307df006c3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 4', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "be89aa4fb1004f48bfc71f682bdf7959",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 5', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cf3e795d13b344a2920b9e932dc25129",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 6', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7f0c95663670411780ee432e33977080",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 7', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "34b644a5133d42a7a1dffe50fabb8311",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 8', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f2ad7d37e7324e9bb51c80da9df5b5e5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 9', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:utils.mura_metrics:Accuracy: 0.6778229590240851\n",
      "INFO:utils.mura_metrics:Kappa: 0.3556718278248585\n",
      "INFO:utils.mura_metrics:Precision: 0.65625\n",
      "INFO:utils.mura_metrics:Recall: 0.6862745098039216\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAVEAAAEmCAYAAADbUaM7AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3Xm8ndO9x/HP95wQQ5AQgpDSiph6kZjKpTSGIBWlMZOiTeuio3tL615tUXrbW62hlFIxT62KUERUXSq5Ig0VQ8QQMg9ISEiT+N0/nnViizPsMc85+3zfXs9r772etZ9n7YSv9UxrKSIwM7PyNOTdADOzjswhamZWAYeomVkFHKJmZhVwiJqZVcAhamZWAYdoJyJpTUn3Slog6c4KtnO8pIeq2ba8SNpb0kt5t8M6Lvk+0fZH0nHAd4FtgHeBicCFEfF4hds9ETgT2DMillXc0HZOUgB9I2JK3m2x+uWeaDsj6bvAr4CfAr2APsBvgCFV2PyngMmdIUCLIalL3m2wOhARXtrJAqwHvAcMbaVOV7KQnZGWXwFd07p9gWnA94A5wEzg5LTux8A/gaVpH6cCPwJuKtj2FkAAXdLnrwCvkvWGXwOOLyh/vOB7ewJPAQvS654F6x4FzgeeSNt5COjZwm9rav9/FLT/cOAQYDLwFvCDgvq7AU8C76S6lwOrp3WPpd+yKP3eowu2/31gFnBjU1n6zmfSPvqnz5sCc4F98/53w0v7XdwTbV8+B6wB3N1KnR8CewA7ATuSBcm5Bes3Jgvj3mRBeYWkHhFxHlnv9vaI6BYR17bWEElrA5cCB0fEOmRBObGZeusD96W6GwC/BO6TtEFBteOAk4GNgNWBs1rZ9cZkfwa9gf8CrgFOAAYAewP/KWnLVHc58B2gJ9mf3UDg3wAiYp9UZ8f0e28v2P76ZL3y4YU7johXyAL2JklrAb8HRkTEo6201zo5h2j7sgEwL1o/3D4e+ElEzImIuWQ9zBML1i9N65dGxP1kvbB+ZbbnQ2AHSWtGxMyImNRMnUOBlyPixohYFhG3Ai8CXyyo8/uImBwR7wN3kP0PoCVLyc7/LgVuIwvIX0fEu2n/z5P9z4OIeDoixqb9vg78Fvh8Eb/pvIhYktrzMRFxDTAFGAdsQvY/LbMWOUTbl/lAzzbO1W0KTC34PDWVrdjGSiG8GOhWakMiYhHZIfA3gJmS7pO0TRHtaWpT74LPs0poz/yIWJ7eN4Xc7IL17zd9X9LWkkZJmiVpIVlPu2cr2waYGxEftFHnGmAH4LKIWNJGXevkHKLty5PAErLzgC2ZQXYo2qRPKivHImCtgs8bF66MiAcj4gCyHtmLZOHSVnua2jS9zDaV4kqydvWNiHWBHwBq4zut3o4iqRvZeeZrgR+l0xVmLXKItiMRsYDsPOAVkg6XtJak1SQdLOm/U7VbgXMlbSipZ6p/U5m7nAjsI6mPpPWAc5pWSOolaUg6N7qE7LTAh81s435ga0nHSeoi6WhgO2BUmW0qxTrAQuC91Es+baX1s4FPl7jNXwPjI+KrZOd6r6q4lVbXHKLtTET8D9k9oueSXRl+EzgD+FOqcgEwHngW+AcwIZWVs6/RwO1pW0/z8eBrSO2YQXbF+vN8MqSIiPnAYLI7AuaTXVkfHBHzymlTic4iu2j1Llkv+faV1v8IGCHpHUlHtbUxSUOAQXz0O78L9Jd0fNVabHXHN9ubmVXAPVEzswo4RM3MKuAQNTOrgEPUzKwC7WoABnVZM7T6Onk3w6pkh603z7sJViXT3pzKW/PntXUPbkka1/1UxLJPPDTWonh/7oMRMaiabaiG9hWiq69D135t3oliHcSoh3+RdxOsSgYP3Kvq24xl75f03/sHE69o62m0XLSrEDWzzkSgjn9GseP/AjPrmARIxS9tbU66TtIcSc8VlA2VNEnSh5J2Wan+OZKmSHpJ0kEF5YNS2RRJZ7e1X4eomeVHDcUvbbue7ImzQs8BR5CNL/vRbqXtgGOA7dN3fiOpUVIjcAVwMNnjy8emui3y4byZ5UTQ0Fi1rUXEY5K2WKnsBQB9sic7BLgtjdL1mqQpZGPzAkyJiFfT925LdZ9vab/uiZpZfqp4OF+i3mTjUjSZlspaKm+Re6Jmlg9R6oWlnpLGF3y+OiKurm6jSucQNbOclNzDnBcRu7RdrSjTgcIbmTfjozFwWypvlg/nzSw/1b2wVIqRwDGSuqY5u/oC/0c20WJfSVtKWp3s4tPI1jbknqiZ5aeK5zol3Uo2e2tPSdOA88jGwr0M2JBsAsWJEXFQREySdAfZBaNlwOlN09JIOgN4EGgErmthbrEVHKJmlpPq3mwfEce2sKrZ2XMj4kLgwmbK7yebsaEoDlEzy0fTzfYdnEPUzPJTB499OkTNLCeCxurdbJ8Xh6iZ5aP0+0TbJYeomeXH50TNzMpVH0PhOUTNLD/uiZqZVcA9UTOzMtVmdKZVziFqZvlxT9TMrALuiZqZlctX583MyieqOj1IXhyiZpYT90TNzCrjc6JmZhVwT9TMrALuiZqZlUk+J2pmVhn3RM3MyieHqJlZebIplhyiZmblkVCDQ9TMrGzuiZqZVcAhamZWAYeomVm5lJYOziFqZrkQck/UzKwS9RCiHf+ZKzPrsCQVvRSxreskzZH0XEHZ+pJGS3o5vfZI5ZJ0qaQpkp6V1L/gO8NS/ZclDWtrvw5RM8tNNUMUuB4YtFLZ2cCYiOgLjEmfAQ4G+qZlOHBlas/6wHnA7sBuwHlNwdsSh6iZ5UMlLm2IiMeAt1YqHgKMSO9HAIcXlN8QmbFAd0mbAAcBoyPirYh4GxjNJ4P5Y3xO1MxyIURDQ0n9uJ6Sxhd8vjoirm7jO70iYmZ6Pwvold73Bt4sqDctlbVU3iKHqJnlpsQLS/MiYpdy9xURISnK/X5LfDhvZvmp4uF8C2anw3TS65xUPh3YvKDeZqmspfIWOUTNLB+q+oWl5owEmq6wDwPuKSg/KV2l3wNYkA77HwQOlNQjXVA6MJW1yIfzZpabat4nKulWYF+yc6fTyK6yXwzcIelUYCpwVKp+P3AIMAVYDJwMEBFvSTofeCrV+0lErHyx6mMcomaWm2qGaEQc28Kqgc3UDeD0FrZzHXBdsft1iJpZLvzYp5lZpTp+hvrCUjVcdd7xTB1zEePv/MGKsiP235mn7/ohi56+lP7b9VlRvv56a/PA1d9k7hP/wyXfH7qivNtaXRl729krljcfuZifn3XkKv0d1ry9du7HgXvvwsH77s7ggXsBcMnPLmC3HT7NwfvuzsH77s4jox9YUf+KX/2cfXbdnv12/xf++sjovJrd/q2aC0s1555oFdx471iuuv2v/O78k1aUTXplBsd87xouP/fjp2k+WLKUn/xmFNtttSnbf2aTFeXvLV7CHsdcvOLzEzf/B396ZGLtG29Fue1PD7D+Bj0/VnbqN87k62d852Nlk196gXvvvpPRj09g9qyZHH/kITw67h80NjauyuZ2GO05HIvlnmgVPDHhFd5asPhjZS+9NpuXp875RN3FH/yTv018lQ+WLG1xe1v12YiN1l+HJya8UvW2Wm2N/vMovviloXTt2pU+n9qCLbb8DBMnPNX2FzspNajopb1yiLZDQwf1566HJuTdDGsiccKXv8ihX9iTW0Zcu6L4hmuv4qB9duWsb36dBe+8DcCsmdPZZNPNVtTZeNPezJo5Y5U3uaOoh8P5moaopEGSXkrDTZ3d9jcMYOhBA7jjgfFtV7RV4g/3jeH+vzzJiNv/xA3X/ZZxf3ucE07+Go+Nf54/PzqOjXptzPn/5X+9S1VKgHbKEJXUCFxBNuTUdsCxkrar1f7qxWe37k2Xxkb+/sKbbVe2VWLjTbLxJ3puuBEHHXIYEyc8xYYb9aKxsZGGhgaOPfEUnpkwfkXdmTOmrfjurBnT2XiTTXNpd0fgEG3dbsCUiHg1Iv4J3EY2/JS14qhB7oW2J4sXLeK9d99d8f6xRx+m37bbM3vWzBV1HrzvHvptk/UPDhh0KPfefSdLlizhjamv89qrU9ip/665tL0jqIcQreXV+eaGlNp95UqShpMNigqrdathc2pnxEVfYe8BfenZvRtTHjif86+6n7cXLOKX3x9Kzx7d+OOl3+DZl6Zz2OlXAPDifT9mnbXXYPXVuvDF/f6Fwf92BS++OguAIw/oz+FnXpnnz7EC8+bOYfiwowFYtmwZQ448mn0HHsi3TzuF5597Fklstvmn+On/XAbA1ttsx6FDjmT/vXamS2MXzv/Zr3xlvjXtNxuLpuzppxpsWPoyMCgivpo+nwjsHhFntPSdhrU2iq79jmpptXUwLz38i7ybYFUyeOBePDvx6apGXtdefaP38b8uuv5rlxz6dCVD4dVKLXuiJQ8pZWadiHyfaFueAvpK2lLS6sAxZMNPmZllw4Sq+KW9qllPNCKWSTqDbCy+RuC6iJhUq/2ZWUcjGtrxTfTFquljnxFxP9m4fWZmn1APh/N+dt7M8tHOD9OL5RA1s1wIfDhvZlYJ90TNzCrgc6JmZuXyOVEzs/Jl94l2/BR1iJpZTtr3wCLFcoiaWW7qIEMdomaWE/kWJzOzsvmcqJlZheogQx2iZpafeuiJerZPM8tNNYfCk/QtSc9JmiTp26lsfUmjJb2cXnukckm6NE2i+ayk/uX+BoeomeVD1ZtjSdIOwNfI5nbbERgsaSvgbGBMRPQFxqTPkE2g2Tctw4Gy5+RxiJpZLqo8KPO2wLiIWBwRy4C/AkeQTY45ItUZARye3g8BbojMWKC7pE3K+R0OUTPLScnzzveUNL5gGV6wseeAvSVtIGkt4BCy6Yl6RUTT1KyzgF7pfXMTafYu51f4wpKZ5abE60rzWpqoLiJekPQz4CFgETARWL5SnZBU9Zk53RM1s3ykm+2LXdoSEddGxICI2Ad4G5gMzG46TE+vc1L1qk2k6RA1s1w03WxfjQtLZNvaKL32ITsfegvZ5JjDUpVhwD3p/UjgpHSVfg9gQcFhf0l8OG9muanyfaJ/kLQBsBQ4PSLekXQxcIekU4GpwFGp7v1k502nAIuBk8vdqUPUzHJTzQyNiL2bKZsPDGymPIDTq7Ffh6iZ5aYenlhyiJpZPjyyvZlZ+eRBmc3MKlMHGeoQNbP8NNRBijpEzSw3dZChDlEzy4cEjZ4exMysfHV9YUnSuq19MSIWVr85ZtaZ1EGGttoTnQQE2SOuTZo+B9Cnhu0yszonstucOroWQzQiNm9pnZlZNdTBKdHiRnGSdIykH6T3m0kaUNtmmVndK2EEp/Z87rTNEJV0ObAfcGIqWgxcVctGmVnnUM2J6vJSzNX5PSOiv6S/A0TEW5JWr3G7zKzOic5zs/1SSQ1kF5NI4/V9WNNWmVmnUAcZWtQ50SuAPwAbSvox8Djws5q2ysw6hXo4J9pmTzQibpD0NLB/KhoaEc/VtllmVu862xNLjWRD7geel8nMqqTjR2hxV+d/CNwKbEo2I94tks6pdcPMrP51isN54CRg54hYDCDpQuDvwEW1bJiZ1bfs6nzerahcMSE6c6V6XVKZmVn52nkPs1itDUByCdk50LeASZIeTJ8PBJ5aNc0zs3pWBxnaak+06Qr8JOC+gvKxtWuOmXUmdd0TjYhrV2VDzKxz6TTnRCV9BrgQ2A5Yo6k8IrauYbvMrBOoh55oMfd8Xg/8nux/HAcDdwC317BNZtYJSNAoFb20V8WE6FoR8SBARLwSEeeShamZWUU6yyhOS9IAJK9I+gYwHVints0ys86gsxzOfwdYG/gmsBfwNeCUWjbKzDqHavZEJX1H0iRJz0m6VdIakraUNE7SFEm3Nw3jKalr+jwlrd+i3N/QZohGxLiIeDci3oiIEyPisIh4otwdmplBNr9Sg4pfWt2W1Juso7dLROxANt7HMWQjzl0SEVsBbwOnpq+cCrydyi+hgpHpWrvZ/m7SGKLNiYgjyt2pmRnVP9fZBVhT0lJgLbInK78AHJfWjwB+BFwJDEnvAe4CLpekiGgx81rbaUsuL3Vjldp52z48MW6V79ZqpMeQy/JuglXJktfm1mS7JZ4T7SlpfMHnqyPiaoCImC7pF8AbwPvAQ8DTwDsRsSzVnwb0Tu97A2+m7y6TtADYAJhX6m9o7Wb7MaVuzMysFCWOqzkvInZpboWkHmS9yy2Bd4A7gUEVNq8oHhvUzHIhqjoU3v7AaxExNyKWAn8kuxDeXVJTZ3EzsruLSK+bk7WhC7AeML+c3+EQNbPcNKj4pQ1vAHtIWktZ4g4Engf+Anw51RkG3JPej0yfSesfKed8KBQ/sj2SukbEknJ2Yma2smpODxIR4yTdBUwAlpGNeXw12eBJt0m6IJU1jQlyLXCjpClkI9UdU+6+i3l2fre0w/WAPpJ2BL4aEWeWu1MzM6juACQRcR5w3krFrwK7NVP3A2BoNfZbzOH8pcBg0vmCiHgG2K8aOzezzq2zPPbZEBFTVzqxu7xG7TGzTiIbCq8dp2ORignRN9MhfUhqBM4EJte2WWbWGdTDle1iQvQ0skP6PsBs4OFUZmZWkTroiLYdohExhwquXJmZNUdFPBPfERRzdf4amnmGPiKG16RFZtZp1EGGFnU4/3DB+zWAL5GeOTUzq0SnmGMpIj42FYikG4HHa9YiM+sURPVuts9T0U8sFdgS6FXthphZJ1Pc45ztXjHnRN/mo3OiDWSPSJ1dy0aZWecgOn6Kthqi6UH+Hflo5JMPy31I38ysUL3MO9/qva4pMO+PiOVpcYCaWdVUcRSn3BTzwMBESTvXvCVm1ulUcTzR3LQ2x1KXNKz+zsBTkl4BFpH1wiMi+q+iNppZHaqXw/nWzon+H9AfOGwVtcXMOpN2PjpTsVoLUQFExCurqC1m1snU+2OfG0r6bksrI+KXNWiPmXUSneFwvhHoBnVwI5eZtUOisc57ojMj4ierrCVm1qlks33m3YrKtXlO1MysJtr5/Z/Fai1EB66yVphZp1TXF5Yi4q1V2RAz61w6w+G8mVlN1XVP1Mys1uogQx2iZpYP0Xlm+zQzqz7RrgcWKZZD1Mxy0/EjtD5602bWAQlolIpeWt2W1E/SxIJloaRvS1pf0mhJL6fXHqm+JF0qaYqkZyWVPSqdQ9TMciMVv7QmIl6KiJ0iYidgALAYuJtsKqMxEdEXGMNHUxsdDPRNy3DgynJ/g0PUzHJS/IDMJZ47HQi8EhFTgSHAiFQ+Ajg8vR8C3BCZsUB3SZuU8yt8TtTMclHG1fmeksYXfL46Iq5upt4xwK3pfa+ImJnez+KjmYp7A28WfGdaKptJiRyiZpabEnuY8yJilza2tzrZQPLnrLwuIkJS1eeJ8+G8meVGJSxFOhiYEBGz0+fZTYfp6XVOKp8ObF7wvc34aFbjkjhEzSwfqslEdcfy0aE8wEhgWHo/DLinoPykdJV+D2BBwWF/SXw4b2a5qPYTS5LWBg4Avl5QfDFwh6RTganAUan8fuAQYArZlfyTy92vQ9TMclPNJ5YiYhGwwUpl82lmWM+ICOD0auzXIWpmuan3QZnNzGomO5zv+CnqEDWz3NTB+CMOUTPLi5B7omZm5XNP1MysTD4namZWiSJGZ+oIHKJmlhuHqJlZBerhwpKfna+R5cuXs8cuO3PEkMEA/OWRMXxu1/7sPmAnvvD5f+WVKVMAWLJkCSccdzTbb7MVe++5O1Nffz3HVhvAVd8ayNSbT2X8FcetKOvRrSujLhjCP64+kVEXDKF7t64A7P3Z3sy6YzhjLzuGsZcdwznH7rriOwcM6MMzvz2B5645kbOGDljlv6O9E9nN9sUu7ZVDtEYuv/TX9Nt22xWfv3nGafz+hpsZ9/REjj7mOC7+6QUAXH/dtfTo3oNJL07hzG99hx/+4Pt5NdmSGx9+gSH/NfJjZWcNHcCjz0zjs8Nv5NFnpn0sFJ+YNIM9zryNPc68jYtufQqAhgbxq9P2Zch5I9n5tJsZus/WbLN5j1X6OzqCBqnopb1yiNbAtGnTeODP93HyKV9dUSaJhQsXArBw4QI22XRTAEbdew/Hn5gNMnPEkV/m0UfGkD3Wa3l5YtIM3nr3g4+VDd7j09z08AsA3PTwC3xxj0+3uo1dt+7FKzPe4fVZC1m67EPufGwyg9v4TmekEv5pr3xOtAb+/Xvf5sKL/pv33nt3Rdlvfvs7vnTYIayx5pqsu+66/PXxsQDMmDGdzTbPhjXs0qUL6663HvPnz6dnz565tN2at1H3tZj19mIAZr29mI26r7Vi3e7bbMy4y45l5luLOOfax3nhjbfYdIO1mTbvvRV1ps97j936bbzK292eNR3Od3Q164lKuk7SHEnP1Wof7dH9941iow03ov+Aj58Du+zXl3D3yPt55fVpnDjsZL5/1ndzaqFVQ5AdLUycMod+J49g9zNv5cp7n+GOcw/NuWUdSSn90PabtrU8nL8eGFTD7bdLT/7tCUaNGkm/rbbgpOOP4dG/PMKXDjuUfzz7DLvtvjsAXx56NGPH/g2ATTftzbQ3s6leli1bxsIFC9hggw1a3L7lY847i9m4R9b73LjHWsx9530A3n1/KYs+WArAg+OnslqXBjZYdw1mzF/EZj27rfh+757dmD7/vU9uuDMrYabPdnxKtHYhGhGPAW/Vavvt1fkXXsQrr0/jpSmvc8PNt7Hvfl/gzj/ew8IFC3h58mQAHnl4NP22yS46HTr4MG6+MZuM8I9/uIvP7/eFqo6xaNVx37jXOGH/7O/shP23ZdTYVwHo1eOjw/pdtu5Fg8T8hR8wfvJsturdnU/1WpfVujQwdJ+tuW/ca7m0vT2rwfQgq1zu50QlDSeb95nN+/TJuTW10aVLF6646hqOPepIGhoa6N6jB7+95joAvnLKqZzylRPZfput6NFjfW68+bacW2sj/uMg9v5sb3quuwZTRpzM+TeP4xd3Ps1NZw9i2AHb8cbcdznhoj8D8KW9tuJrh+zAsuXBB/9cxkn//QAAyz8MvnPlX7n3/MNobGhgxOjneeGNTtenaFV2TrQ9x2NxVMsrwZK2AEZFxA7F1B8wYJd4Ytz4titah9BjyGV5N8GqZMnjP+fDBW9UNfG2/ezO8fu7/1J0/c/17fF0W7N95iH3nqiZdWIdvyPqEDWz/NTD4Xwtb3G6FXgS6CdpWpptz8xsBV9YakVEHFurbZtZnWjP6VgkH86bWS6yHmbHT1GHqJnlo53fRF8sh6iZ5aYOMtQhamY5qoMUdYiaWU7a98AixfJ4omaWm2oOQCKpu6S7JL0o6QVJn5O0vqTRkl5Orz1SXUm6VNIUSc9K6l/ub3CImlkuSrlHtMj+6q+BByJiG2BH4AXgbGBMRPQFxqTPAAcDfdMyHLiy3N/hEDWz3EgqemljO+sB+wDXAkTEPyPiHWAIMCJVGwEcnt4PAW6IzFigu6RNyvkNDlEzy00VD+e3BOYCv5f0d0m/k7Q20CsiZqY6s4Be6X1v4M2C709LZSVziJpZbko8nO8paXzBMrxgU12A/sCVEbEzsIiPDt0BiGzIuqoPW+er82aWj9Ifip/XylB404BpETEufb6LLERnS9okImamw/U5af10YPOC72+WykrmnqiZ5aZacyxFxCzgTUn9UtFA4HlgJDAslQ0D7knvRwInpav0ewALCg77S+KeqJnlQlT9sc8zgZslrQ68CpxM1lG8I40iNxU4KtW9HzgEmAIsTnXL4hA1s9xUM0MjYiLQ3OH+wGbqBnB6NfbrEDWz/HT8B5YcomaWn3p47NMhama5aej4GeoQNbMcOUTNzMrjke3NzCrhke3NzCpTBxnqEDWzHNVBijpEzSwn9TGyvUPUzHLjc6JmZmUqfRCn9skhamb5qYMUdYiaWW4a6uB43iFqZrnp+BHqEDWzvPhmezOzSnX8FHWImlkuajCyfS4comaWmzrIUIeomeXHPVEzswr4sU8zs0p0/Ax1iJpZfuogQx2iZpYPyU8smZlVpuNnqEPUzPJTBxnqEDWz/NTB0bxD1Mzy4pHtzczKVi+PfTbk3QAzs2qQ9Lqkf0iaKGl8Kltf0mhJL6fXHqlcki6VNEXSs5L6l7tfh6iZ5UYqfinSfhGxU0Tskj6fDYyJiL7AmPQZ4GCgb1qGA1eW+xscomaWG5XwT5mGACPS+xHA4QXlN0RmLNBd0ibl7MAhama5yG62L34BekoaX7AMX2mTATwk6emCdb0iYmZ6Pwvold73Bt4s+O60VFYyX1gys/yU1sGcV3CY3px/jYjpkjYCRkt6sXBlRISkKKOVrXJP1MxyU83D+YiYnl7nAHcDuwGzmw7T0+ucVH06sHnB1zdLZSVziJpZbqp1YUnS2pLWaXoPHAg8B4wEhqVqw4B70vuRwEnpKv0ewIKCw/6S+HDezHJTxdtEewF3K0vbLsAtEfGApKeAOySdCkwFjkr17wcOAaYAi4GTy92xQ9TM8lOlFI2IV4EdmymfDwxspjyA06uxb4eomeWmHh77VBbI7YOkuWRd7nrXE5iXdyOsKjrL3+WnImLDam5Q0gNkf37FmhcRg6rZhmpoVyHaWUga38atGtZB+O/SfHXezKwCDlEzswo4RPNxdd4NsKrx32Un53OiZmYVcE/UzKwCDlEzswo4RM3MKuAQXQUk9ZP0OUmrSWrMuz1WOf89WhNfWKoxSUcAPyUbZms6MB64PiIW5towK4ukrSNicnrfGBHL826T5cs90RqStBpwNHBqRAwkG4Zrc+D7ktbNtXFWMkmDgYmSbgGIiOXukZpDtPbWJZsMC7KBYkcBqwHHSfUwYWznkMaoPAP4NvBPSTeBg9QcojUVEUuBXwJHSNo7Ij4EHgcmAv+aa+OsJBGxCDgFuAU4C1ijMEjzbJvlyyFae/8LPAScKGmfiFgeEbcAm9LM+IfWfkXEjIh4LyLmAV8H1mwKUkn9JW2TbwstDx5PtMYi4gNJN5PNRHhO+g9tCdlI3GVNR2D5i4j5kr4O/DxNiNYI7JdzsywHDtFVICLelnQN8DxZD+YD4ISImJ1vy6wSETFP0rPAwcABETEt7zbZqudbnFaxdBEi0vlR68Ak9QDuAL4XEc/m3R7Lh0PUrAKS1oiID/Juh+XHIWpmVgFfnTd1nZ9yAAADIklEQVQzq4BD1MysAg5RM7MKOETNzCrgEK0TkpZLmijpOUl3Slqrgm3tK2lUen+YpLNbqdtd0r+VsY8fSTqr2PKV6lwv6csl7GsLSc+V2kazYjhE68f7EbFTROwA/BP4RuFKZUr++46IkRFxcStVugMlh6hZvXCI1qf/BbZKPbCXJN0APAdsLulASU9KmpB6rN0AJA2S9KKkCcARTRuS9BVJl6f3vSTdLemZtOwJXAx8JvWCf57q/bukpyQ9K+nHBdv6oaTJkh4H+rX1IyR9LW3nGUl/WKl3vb+k8Wl7g1P9Rkk/L9j31yv9gzRri0O0zkjqQvYY4j9SUV/gNxGxPbAIOBfYPyL6kw0Q/V1JawDXAF8EBgAbt7D5S4G/RsSOQH9gEnA28ErqBf+7pAPTPncDdgIGSNpH0gDgmFR2CLBrET/njxGxa9rfC8CpBeu2SPs4FLgq/YZTgQURsWva/tckbVnEfszK5mfn68eakiam9/8LXEs2UtTUiBibyvcAtgOeSEOZrg48CWwDvBYRLwOkkYmGN7OPLwAnwYrh3xakRx8LHZiWv6fP3chCdR3g7ohYnPYxsojftIOkC8hOGXQDHixYd0d6dPZlSa+m33Ag8C8F50vXS/ueXMS+zMriEK0f70fEToUFKSgXFRYBoyPi2JXqfex7FRJwUUT8dqV9fLuMbV0PHB4Rz0j6CrBvwbqVH7WLtO8zI6IwbJG0RRn7NiuKD+c7l7HAXpK2gmy0dklbAy8CW0j6TKp3bAvfHwOclr7bKGk94F2yXmaTB4FTCs619pa0EfAYcLikNSWtQ3bqoC3rADPTNCvHr7RuqKSG1OZPAy+lfZ+W6iNp6zQivVnNuCfaiUTE3NSju1VS11R8bkRMljQcuE/SYrLTAes0s4lvAVdLOhVYDpwWEU9KeiLdQvTndF50W+DJ1BN+j2zYvwmSbgeeAeYATxXR5P8ExgFz02thm94A/o9s+pVvpHFbf0d2rnSCsp3PBQ4v7k/HrDwegMTMrAI+nDczq4BD1MysAg5RM7MKOETNzCrgEDUzq4BD1MysAg5RM7MK/D+yt5a+Yd4+ngAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:utils.mura_metrics:Accuracy: 0.6914095079232694\n",
      "INFO:utils.mura_metrics:Kappa: 0.3738726659421619\n",
      "INFO:utils.mura_metrics:Precision: 0.6627906976744186\n",
      "INFO:utils.mura_metrics:Recall: 0.6356877323420075\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:__main__:Saving evaluation metrics history...\n",
      "INFO:__main__:executing epochs 51-60\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1eaa7b24c333422691aef16b154bf1a4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Training', max=10), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "80c77ed342d1463f84e5a6a2e8656ede",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 0', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "15a0a61f6dcd4f379d77da9dbf8e3670",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 1', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c885ea23c28a4c7eb753ad07476c5263",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 2', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "13812b9c534740a2b5694954696fe6be",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 3', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "804913f6f04546d0b074da24555b86ca",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 4', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7bb00125ac4a40238b86f178016a8b7d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch 5', max=2300), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-41-af3298155ec1>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     27\u001b[0m                            \u001b[0msteps_per_epoch\u001b[0m \u001b[0;34m=\u001b[0m  \u001b[0mint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mround\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx_train\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m/\u001b[0m \u001b[0mbatch_size\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     28\u001b[0m                            \u001b[0mepochs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mepochs_per_eval\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mverbose\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 29\u001b[0;31m                            callbacks=[progress])\n\u001b[0m\u001b[1;32m     30\u001b[0m                            \u001b[0;31m# callbacks=[clr, TQDMNotebookCallback()])\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     31\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/mura/lib/python3.6/site-packages/keras/legacy/interfaces.py\u001b[0m in \u001b[0;36mwrapper\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m     89\u001b[0m                 warnings.warn('Update your `' + object_name +\n\u001b[1;32m     90\u001b[0m                               '` call to the Keras 2 API: ' + signature, stacklevel=2)\n\u001b[0;32m---> 91\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     92\u001b[0m         \u001b[0mwrapper\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_original_function\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     93\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mwrapper\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/mura/lib/python3.6/site-packages/keras/engine/training.py\u001b[0m in \u001b[0;36mfit_generator\u001b[0;34m(self, generator, steps_per_epoch, epochs, verbose, callbacks, validation_data, validation_steps, class_weight, max_queue_size, workers, use_multiprocessing, shuffle, initial_epoch)\u001b[0m\n\u001b[1;32m   1424\u001b[0m             \u001b[0muse_multiprocessing\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0muse_multiprocessing\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1425\u001b[0m             \u001b[0mshuffle\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mshuffle\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1426\u001b[0;31m             initial_epoch=initial_epoch)\n\u001b[0m\u001b[1;32m   1427\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1428\u001b[0m     \u001b[0;34m@\u001b[0m\u001b[0minterfaces\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlegacy_generator_methods_support\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/mura/lib/python3.6/site-packages/keras/engine/training_generator.py\u001b[0m in \u001b[0;36mfit_generator\u001b[0;34m(model, generator, steps_per_epoch, epochs, verbose, callbacks, validation_data, validation_steps, class_weight, max_queue_size, workers, use_multiprocessing, shuffle, initial_epoch)\u001b[0m\n\u001b[1;32m    189\u001b[0m                 outs = model.train_on_batch(x, y,\n\u001b[1;32m    190\u001b[0m                                             \u001b[0msample_weight\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0msample_weight\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 191\u001b[0;31m                                             class_weight=class_weight)\n\u001b[0m\u001b[1;32m    192\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    193\u001b[0m                 \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mouts\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlist\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/mura/lib/python3.6/site-packages/keras/engine/training.py\u001b[0m in \u001b[0;36mtrain_on_batch\u001b[0;34m(self, x, y, sample_weight, class_weight)\u001b[0m\n\u001b[1;32m   1218\u001b[0m             \u001b[0mins\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mx\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0my\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0msample_weights\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1219\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_make_train_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1220\u001b[0;31m         \u001b[0moutputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mins\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1221\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1222\u001b[0m             \u001b[0;32mreturn\u001b[0m \u001b[0moutputs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/mura/lib/python3.6/site-packages/keras/backend/tensorflow_backend.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, inputs)\u001b[0m\n\u001b[1;32m   2659\u001b[0m                 \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_legacy_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2660\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2661\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2662\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2663\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mpy_any\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mis_tensor\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mx\u001b[0m \u001b[0;32min\u001b[0m \u001b[0minputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/mura/lib/python3.6/site-packages/keras/backend/tensorflow_backend.py\u001b[0m in \u001b[0;36m_call\u001b[0;34m(self, inputs)\u001b[0m\n\u001b[1;32m   2629\u001b[0m                                 \u001b[0msymbol_vals\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2630\u001b[0m                                 session)\n\u001b[0;32m-> 2631\u001b[0;31m         \u001b[0mfetched\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_callable_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0marray_vals\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2632\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mfetched\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2633\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/mura/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, *args)\u001b[0m\n\u001b[1;32m   1449\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_session\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_created_with_new_api\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1450\u001b[0m           return tf_session.TF_SessionRunCallable(\n\u001b[0;32m-> 1451\u001b[0;31m               self._session._session, self._handle, args, status, None)\n\u001b[0m\u001b[1;32m   1452\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1453\u001b[0m           return tf_session.TF_DeprecatedSessionRunCallable(\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "time: 7h 11min 48s\n"
     ]
    }
   ],
   "source": [
    "while (train_round+1) * epochs_per_eval + starting_epoch <= total_epochs:\n",
    "    log.info(f'executing epochs {train_round * epochs_per_eval + 1 + starting_epoch}-{(train_round+1) * epochs_per_eval + starting_epoch}')\n",
    "    \n",
    "#     if train_round * epochs_per_eval + 1 + starting_epoch >= 10:\n",
    "#         if np.mean(train_history['loss'][-10:-5]) < np.mean(train_history['loss'][-5:]): # if no improvements in 5 epochs\n",
    "#             if optimizer == 'adam':\n",
    "#                 log.info(f'switching to SGD with lr={lr}')\n",
    "#                 optimizer = 'sgd'\n",
    "#             else:\n",
    "#                 log.info(f'reducing SGD learn rate from {lr} to {lr * 0.8}')\n",
    "#                 lr = lr * 0.8\n",
    "#             opt = optimizers.SGD(lr=lr, momentum=0.9, nesterov=True)\n",
    "#         else:\n",
    "#             if optimizer == 'adam':\n",
    "#                 opt = optimizers.Adam(lr=lr)\n",
    "#             else:\n",
    "#                 opt = optimizers.SGD(lr=lr, momentum=0.9, nesterov=True)\n",
    "#     else:\n",
    "#         opt = optimizers.Adam(lr=lr)\n",
    "\n",
    "    model.compile(\n",
    "        loss=weighted_binary_crossentropy(p1),\n",
    "        optimizer=opt,\n",
    "        metrics=[metrics.binary_accuracy])\n",
    "    \n",
    "    model.fit_generator(datagen.flow(x_train, y_train, batch_size=batch_size),\n",
    "                           steps_per_epoch =  int(np.round(len(x_train) / batch_size)), \n",
    "                           epochs=epochs_per_eval, verbose=0,\n",
    "                           callbacks=[progress])\n",
    "                           # callbacks=[clr, TQDMNotebookCallback()])\n",
    "    \n",
    "    # train_history['learn_rate'].extend(clr.history['lr'])\n",
    "    train_history['loss'].extend(model.history.history['loss'])\n",
    "    train_history['binary_accuracy'].extend(model.history.history['binary_accuracy'])\n",
    "    \n",
    "    y_valid_hat = model.predict(x_test)\n",
    "    true_label = np.round(y_test)\n",
    "    pred_label = np.round(y_valid_hat)\n",
    "    evaluate = utils.MURAMetrics(true_label, pred_label)\n",
    "    \n",
    "    per_image_metrics = evaluate.report_by_image()\n",
    "    for key in per_image_metrics:\n",
    "        val_per_image_history[key].append(per_image_metrics[key])\n",
    "    \n",
    "    per_study_metrics = evaluate.report_by_study()\n",
    "    for key in per_image_metrics:\n",
    "        val_per_study_history[key].append(per_study_metrics[key])\n",
    "    \n",
    "    cur_epoch = (train_round + 1) * epochs_per_eval\n",
    "    if cur_epoch % epochs_per_save == 0:\n",
    "        log.info(f'Saving Model to trained_models/{model_name}_{cur_epoch + starting_epoch}.h5...')\n",
    "        model.save(f'./trained_models/{model_name}_{cur_epoch + starting_epoch}.h5')\n",
    "    \n",
    "    log.info('Saving evaluation metrics history...')\n",
    "    with open(f'./trained_models/{model_name}_train_history.pkl', 'wb') as pkl_file:\n",
    "        pickle.dump(train_history, pkl_file)\n",
    "    with open(f'./trained_models/{model_name}_val_per_image_history.pkl', 'wb') as pkl_file:\n",
    "        pickle.dump(val_per_image_history, pkl_file)\n",
    "    with open(f'./trained_models/{model_name}_val_per_study_history.pkl', 'wb') as pkl_file:\n",
    "        pickle.dump(val_per_study_history, pkl_file)\n",
    "    \n",
    "    train_round += 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}