DOC_.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2.0.0\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow.keras import layers\n",
    "import time\n",
    "from pathlib import Path\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib import gridspec\n",
    "import matplotlib.patches as mpatches\n",
    "import numpy as np\n",
    "import os\n",
    "from sklearn.model_selection import train_test_split\n",
    "from skimage import data,segmentation,measure,morphology,color\n",
    "from tensorflow.python.keras.applications.vgg16 import VGG16\n",
    "#from tensorflow_core.python.keras.layers.convolutional import Conv2D\n",
    "#from tensorflow_core.python.keras.layers import Dense,Flatten\n",
    "#from tensorflow_core.python.keras.models import Model\n",
    "import numpy as np\n",
    "#from tensorflow_core.python.ops.init_ops_v2 import he_normal\n",
    "import os \n",
    "from PIL import Image \n",
    "import numpy as np \n",
    "import cv2\n",
    "import tensorflow_probability as tfp\n",
    "\n",
    "import math\n",
    "print(tf.__version__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "#import tensorflow_datasets as tfds\n",
    "#print(tfds.list_builders())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tf version: 2.0.0 \n",
      "use GPU True\n"
     ]
    }
   ],
   "source": [
    "\n",
    "config = tf.compat.v1.ConfigProto()\n",
    "config.gpu_options.allow_growth = True\n",
    "sess = tf.compat.v1.Session(config=config)\n",
    "config = tf.compat.v1.ConfigProto()\n",
    "config.gpu_options.per_process_gpu_memory_fraction = 0.95\n",
    "tf.compat.v1.Session(config=config)\n",
    "version = tf.__version__\n",
    "gpu_ok = tf.test.is_gpu_available()\n",
    "os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"0\"\n",
    "print(\"tf version:\",version,\"\\nuse GPU\",gpu_ok)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('64bit', 'WindowsPE')"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import sys, platform\n",
    "\n",
    "platform.architecture()\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 匯入資料"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'\\nSPLIT_WEIGHTS = (8, 1, 1)\\nsplits = tfds.Split.TRAIN.subsplit(weighted=SPLIT_WEIGHTS)\\n(raw_train, raw_validation, raw_test), metadata = tfds.load(name=\"caltech101\", \\n                                                            with_info=True,\\n                                                            split=list(splits),\\n# specifying batch_size=-1 will load full dataset in the memory\\n#                                                             batch_size=-1,\\n# as_supervised: `bool`, if `True`, the returned `tf.data.Dataset`\\n# will have a 2-tuple structure `(input, label)`                                                            \\n                                                            as_supervised=True)\\n'"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''\n",
    "SPLIT_WEIGHTS = (8, 1, 1)\n",
    "splits = tfds.Split.TRAIN.subsplit(weighted=SPLIT_WEIGHTS)\n",
    "(raw_train, raw_validation, raw_test), metadata = tfds.load(name=\"caltech101\", \n",
    "                                                            with_info=True,\n",
    "                                                            split=list(splits),\n",
    "# specifying batch_size=-1 will load full dataset in the memory\n",
    "#                                                             batch_size=-1,\n",
    "# as_supervised: `bool`, if `True`, the returned `tf.data.Dataset`\n",
    "# will have a 2-tuple structure `(input, label)`                                                            \n",
    "                                                            as_supervised=True)\n",
    "'''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "img_size = 224\n",
    "source_path = '101_ObjectCategories'\n",
    "target_name = 'airplanes'#negative'#'positive'\n",
    "category_dir = os.path.join(source_path, target_name)\n",
    "target_data = []\n",
    "for img in os.listdir(category_dir):\n",
    "    try:\n",
    "        img = cv2.imread(os.path.join(category_dir, img))\n",
    "        img = cv2.resize(img, (img_size, img_size))\n",
    "        #img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)   \n",
    "    except:\n",
    "        continue\n",
    "    target_data.append(img)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(800, 224, 224, 3)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "target_data = np.array(target_data)/255.\n",
    "target_data = target_data.astype('float32')\n",
    "target_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "讀取：類別 0 accordion\n",
      "讀取：類別 1 anchor\n",
      "讀取：類別 2 ant\n",
      "讀取：類別 3 BACKGROUND_Google\n",
      "讀取：類別 4 barrel\n",
      "讀取：類別 5 bass\n",
      "讀取：類別 6 beaver\n",
      "讀取：類別 7 binocular\n",
      "讀取：類別 8 bonsai\n",
      "讀取：類別 9 brain\n",
      "讀取：類別 10 brontosaurus\n",
      "讀取：類別 11 buddha\n",
      "讀取：類別 12 butterfly\n",
      "讀取：類別 13 camera\n",
      "讀取：類別 14 cannon\n",
      "讀取：類別 15 car_side\n",
      "讀取：類別 16 ceiling_fan\n",
      "讀取：類別 17 cellphone\n",
      "讀取：類別 18 chair\n",
      "讀取：類別 19 chandelier\n",
      "讀取：類別 20 cougar_body\n",
      "讀取：類別 21 cougar_face\n",
      "讀取：類別 22 crab\n",
      "讀取：類別 23 crayfish\n",
      "讀取：類別 24 crocodile\n",
      "讀取：類別 25 crocodile_head\n",
      "讀取：類別 26 cup\n",
      "讀取：類別 27 dalmatian\n",
      "讀取：類別 28 dollar_bill\n",
      "讀取：類別 29 dolphin\n",
      "讀取：類別 30 dragonfly\n",
      "讀取：類別 31 electric_guitar\n",
      "讀取：類別 32 elephant\n",
      "讀取：類別 33 emu\n",
      "讀取：類別 34 euphonium\n",
      "讀取：類別 35 ewer\n",
      "讀取：類別 36 Faces\n",
      "讀取：類別 37 Faces_easy\n",
      "讀取：類別 38 ferry\n",
      "讀取：類別 39 flamingo\n",
      "讀取：類別 40 flamingo_head\n",
      "讀取：類別 41 garfield\n",
      "讀取：類別 42 gerenuk\n",
      "讀取：類別 43 gramophone\n",
      "讀取：類別 44 grand_piano\n",
      "讀取：類別 45 hawksbill\n",
      "讀取：類別 46 headphone\n",
      "讀取：類別 47 hedgehog\n",
      "讀取：類別 48 helicopter\n",
      "讀取：類別 49 ibis\n",
      "讀取：類別 50 inline_skate\n",
      "讀取：類別 51 joshua_tree\n",
      "讀取：類別 52 kangaroo\n",
      "讀取：類別 53 ketch\n",
      "讀取：類別 54 lamp\n",
      "讀取：類別 55 laptop\n",
      "讀取：類別 56 Leopards\n",
      "讀取：類別 57 llama\n",
      "讀取：類別 58 lobster\n",
      "讀取：類別 59 lotus\n",
      "讀取：類別 60 mandolin\n",
      "讀取：類別 61 mayfly\n",
      "讀取：類別 62 menorah\n",
      "讀取：類別 63 metronome\n",
      "讀取：類別 64 minaret\n",
      "讀取：類別 65 Motorbikes\n",
      "讀取：類別 66 nautilus\n",
      "讀取：類別 67 octopus\n",
      "讀取：類別 68 okapi\n",
      "讀取：類別 69 pagoda\n",
      "讀取：類別 70 panda\n",
      "讀取：類別 71 pigeon\n",
      "讀取：類別 72 pizza\n",
      "讀取：類別 73 platypus\n",
      "讀取：類別 74 pyramid\n",
      "讀取：類別 75 revolver\n",
      "讀取：類別 76 rhino\n",
      "讀取：類別 77 rooster\n",
      "讀取：類別 78 saxophone\n",
      "讀取：類別 79 schooner\n",
      "讀取：類別 80 scissors\n",
      "讀取：類別 81 scorpion\n",
      "讀取：類別 82 sea_horse\n",
      "讀取：類別 83 snoopy\n",
      "讀取：類別 84 soccer_ball\n",
      "讀取：類別 85 stapler\n",
      "讀取：類別 86 starfish\n",
      "讀取：類別 87 stegosaurus\n",
      "讀取：類別 88 stop_sign\n",
      "讀取：類別 89 strawberry\n",
      "讀取：類別 90 sunflower\n",
      "讀取：類別 91 tick\n",
      "讀取：類別 92 trilobite\n",
      "讀取：類別 93 umbrella\n",
      "讀取：類別 94 watch\n",
      "讀取：類別 95 water_lilly\n",
      "讀取：類別 96 wheelchair\n",
      "讀取：類別 97 wild_cat\n",
      "讀取：類別 98 windsor_chair\n",
      "讀取：類別 99 wrench\n",
      "讀取：類別 100 yin_yang\n"
     ]
    }
   ],
   "source": [
    "reference_data = []\n",
    "reference_label = []\n",
    "label = 0\n",
    "source_path = os.walk(r\"101_ObjectCategories\")\n",
    "\n",
    "for path, dir_list, file_list in source_path:  \n",
    "    \n",
    "    for dir_name in dir_list:\n",
    "        if label<=100:\n",
    "            if dir_name == target_name:\n",
    "                continue\n",
    "            print('讀取：類別', label, dir_name)\n",
    "            category_dir = os.path.join(path, dir_name)\n",
    "\n",
    "            for img in os.listdir(category_dir):\n",
    "                try:\n",
    "                    img = cv2.imread(os.path.join(category_dir, img))\n",
    "                    img = cv2.resize(img, (img_size, img_size))\n",
    "                    reference_data.append(img)\n",
    "                    reference_label.append(label)\n",
    "                except:\n",
    "                    continue\n",
    "            label += 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(8345, 224, 224, 3)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reference_data = np.array(reference_data)/255.\n",
    "reference_data = reference_data.astype('float32')\n",
    "\n",
    "num_class = len(np.unique(reference_label))\n",
    "reference_label = np.array(reference_label)\n",
    "reference_label = reference_label.astype('int')\n",
    "reference_label = np.eye(num_class)[reference_label]\n",
    "reference_label = reference_label.astype('float32')\n",
    "reference_data.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 儲存資料"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"\\nnum_class = 101\\nnp.save('reference_data', reference_data)\\nnp.save('reference_label', reference_label)\\nnp.save('target_data', target_data)\\n\""
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''\n",
    "num_class = 101\n",
    "np.save('reference_data', reference_data)\n",
    "np.save('reference_label', reference_label)\n",
    "np.save('target_data', target_data)\n",
    "'''"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## -----------------------------------------------------------------------------------------------------------------------"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"\\nreference_data = np.load('reference_data.npy')\\nreference_label = np.load('reference_label.npy')\\ntarget_data = np.load('target_data.npy')\\n\""
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''\n",
    "reference_data = np.load('reference_data.npy')\n",
    "reference_label = np.load('reference_label.npy')\n",
    "target_data = np.load('target_data.npy')\n",
    "'''"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 切割data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "Rdata_train,Rdata_test, Rlabel_train, Rlabel_test = train_test_split(reference_data,reference_label,test_size = 0.3,\n",
    "                                                   random_state = 20)\n",
    "Tdata_train,Tdata_test = train_test_split(target_data,test_size = 0.3,\n",
    "                                                   random_state = 20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(5841, 224, 224, 3)\n",
      "(560, 224, 224, 3)\n"
     ]
    }
   ],
   "source": [
    "print(Rdata_train.shape)\n",
    "print(Tdata_train.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "batch_size = 10\n",
    "\n",
    "'''將data變成batch_train'''\n",
    "reference_dataset = tf.data.Dataset.from_tensor_slices((Rdata_train, Rlabel_train))\n",
    "reference_dataset = reference_dataset.shuffle(buffer_size = len(Rdata_train))\n",
    "reference_dataset = reference_dataset.batch(batch_size)\n",
    "\n",
    "target_dataset = tf.data.Dataset.from_tensor_slices(Tdata_train)\n",
    "target_dataset = target_dataset.shuffle(buffer_size=len(Tdata_train))\n",
    "target_dataset = target_dataset.batch(batch_size)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## pre_train model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"vgg16\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_1 (InputLayer)         [(None, 224, 224, 3)]     0         \n",
      "_________________________________________________________________\n",
      "block1_conv1 (Conv2D)        (None, 224, 224, 64)      1792      \n",
      "_________________________________________________________________\n",
      "block1_conv2 (Conv2D)        (None, 224, 224, 64)      36928     \n",
      "_________________________________________________________________\n",
      "block1_pool (MaxPooling2D)   (None, 112, 112, 64)      0         \n",
      "_________________________________________________________________\n",
      "block2_conv1 (Conv2D)        (None, 112, 112, 128)     73856     \n",
      "_________________________________________________________________\n",
      "block2_conv2 (Conv2D)        (None, 112, 112, 128)     147584    \n",
      "_________________________________________________________________\n",
      "block2_pool (MaxPooling2D)   (None, 56, 56, 128)       0         \n",
      "_________________________________________________________________\n",
      "block3_conv1 (Conv2D)        (None, 56, 56, 256)       295168    \n",
      "_________________________________________________________________\n",
      "block3_conv2 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_conv3 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_pool (MaxPooling2D)   (None, 28, 28, 256)       0         \n",
      "_________________________________________________________________\n",
      "block4_conv1 (Conv2D)        (None, 28, 28, 512)       1180160   \n",
      "_________________________________________________________________\n",
      "block4_conv2 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_conv3 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_pool (MaxPooling2D)   (None, 14, 14, 512)       0         \n",
      "_________________________________________________________________\n",
      "block5_conv1 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv2 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv3 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_pool (MaxPooling2D)   (None, 7, 7, 512)         0         \n",
      "=================================================================\n",
      "Total params: 14,714,688\n",
      "Trainable params: 14,714,688\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "input_shape = (224,224,3)\n",
    "pre_model = tf.keras.applications.VGG16(weights='imagenet', include_top = False, input_shape = input_shape) #include_top: 全盤接受VGG16\n",
    "pre_model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## transfer learning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "def block(x, layers):\n",
    "    \n",
    "    x = tf.keras.layers.Dense(layers, activation = 'linear')(x)\n",
    "    x = tf.keras.layers.ReLU()(x)\n",
    "    x = tf.keras.layers.Dropout(0.2)(x)\n",
    "    \n",
    "    return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_model(model, trainable = False):\n",
    "    if trainable:\n",
    "        model.trainable = True\n",
    "    else:\n",
    "        model.trainable = False\n",
    "    x = tf.keras.layers.Flatten()(model.output)\n",
    "    x = block(x,1000)\n",
    "    x = block(x,1000)\n",
    "\n",
    "    output_t = tf.keras.layers.Dense(1000)(x)\n",
    "    \n",
    "    target_model = tf.keras.Model(inputs = model.input, outputs = output_t)\n",
    "    \n",
    "    output_r = tf.keras.layers.Dense(num_class, activation = tf.nn.softmax, use_bias = True)(target_model.output)\n",
    "    \n",
    "    related_model = tf.keras.Model(inputs = model.input, outputs = output_r)    \n",
    "    \n",
    "    return related_model, target_model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'\\ndef make_model(model):\\n    model.trainable = False\\n    for layer in model.layers[:layer_num]:\\n        layer.trainable =  True\\n        \\n    target_model = tf.keras.Model(inputs = model.input, outputs=model.output)\\n    x = tf.keras.layers.Dense(num_class, activation = tf.nn.softmax, use_bias = True)(target_model.output)\\n    related_model = tf.keras.Model(inputs = model.input, outputs = x)    \\n    \\n    return related_model, target_model\\n'"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''\n",
    "def make_model(model):\n",
    "    model.trainable = False\n",
    "    for layer in model.layers[:layer_num]:\n",
    "        layer.trainable =  True\n",
    "        \n",
    "    target_model = tf.keras.Model(inputs = model.input, outputs=model.output)\n",
    "    x = tf.keras.layers.Dense(num_class, activation = tf.nn.softmax, use_bias = True)(target_model.output)\n",
    "    related_model = tf.keras.Model(inputs = model.input, outputs = x)    \n",
    "    \n",
    "    return related_model, target_model\n",
    "'''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "related_model, target_model =  make_model(model = pre_model, trainable = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_1 (InputLayer)         [(None, 224, 224, 3)]     0         \n",
      "_________________________________________________________________\n",
      "block1_conv1 (Conv2D)        (None, 224, 224, 64)      1792      \n",
      "_________________________________________________________________\n",
      "block1_conv2 (Conv2D)        (None, 224, 224, 64)      36928     \n",
      "_________________________________________________________________\n",
      "block1_pool (MaxPooling2D)   (None, 112, 112, 64)      0         \n",
      "_________________________________________________________________\n",
      "block2_conv1 (Conv2D)        (None, 112, 112, 128)     73856     \n",
      "_________________________________________________________________\n",
      "block2_conv2 (Conv2D)        (None, 112, 112, 128)     147584    \n",
      "_________________________________________________________________\n",
      "block2_pool (MaxPooling2D)   (None, 56, 56, 128)       0         \n",
      "_________________________________________________________________\n",
      "block3_conv1 (Conv2D)        (None, 56, 56, 256)       295168    \n",
      "_________________________________________________________________\n",
      "block3_conv2 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_conv3 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_pool (MaxPooling2D)   (None, 28, 28, 256)       0         \n",
      "_________________________________________________________________\n",
      "block4_conv1 (Conv2D)        (None, 28, 28, 512)       1180160   \n",
      "_________________________________________________________________\n",
      "block4_conv2 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_conv3 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_pool (MaxPooling2D)   (None, 14, 14, 512)       0         \n",
      "_________________________________________________________________\n",
      "block5_conv1 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv2 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv3 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_pool (MaxPooling2D)   (None, 7, 7, 512)         0         \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 25088)             0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 1000)              25089000  \n",
      "_________________________________________________________________\n",
      "re_lu (ReLU)                 (None, 1000)              0         \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 1000)              0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 1000)              1001000   \n",
      "_________________________________________________________________\n",
      "re_lu_1 (ReLU)               (None, 1000)              0         \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 1000)              0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 1000)              1001000   \n",
      "=================================================================\n",
      "Total params: 41,805,688\n",
      "Trainable params: 27,091,000\n",
      "Non-trainable params: 14,714,688\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "target_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model_1\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_1 (InputLayer)         [(None, 224, 224, 3)]     0         \n",
      "_________________________________________________________________\n",
      "block1_conv1 (Conv2D)        (None, 224, 224, 64)      1792      \n",
      "_________________________________________________________________\n",
      "block1_conv2 (Conv2D)        (None, 224, 224, 64)      36928     \n",
      "_________________________________________________________________\n",
      "block1_pool (MaxPooling2D)   (None, 112, 112, 64)      0         \n",
      "_________________________________________________________________\n",
      "block2_conv1 (Conv2D)        (None, 112, 112, 128)     73856     \n",
      "_________________________________________________________________\n",
      "block2_conv2 (Conv2D)        (None, 112, 112, 128)     147584    \n",
      "_________________________________________________________________\n",
      "block2_pool (MaxPooling2D)   (None, 56, 56, 128)       0         \n",
      "_________________________________________________________________\n",
      "block3_conv1 (Conv2D)        (None, 56, 56, 256)       295168    \n",
      "_________________________________________________________________\n",
      "block3_conv2 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_conv3 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_pool (MaxPooling2D)   (None, 28, 28, 256)       0         \n",
      "_________________________________________________________________\n",
      "block4_conv1 (Conv2D)        (None, 28, 28, 512)       1180160   \n",
      "_________________________________________________________________\n",
      "block4_conv2 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_conv3 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_pool (MaxPooling2D)   (None, 14, 14, 512)       0         \n",
      "_________________________________________________________________\n",
      "block5_conv1 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv2 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv3 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_pool (MaxPooling2D)   (None, 7, 7, 512)         0         \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 25088)             0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 1000)              25089000  \n",
      "_________________________________________________________________\n",
      "re_lu (ReLU)                 (None, 1000)              0         \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 1000)              0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 1000)              1001000   \n",
      "_________________________________________________________________\n",
      "re_lu_1 (ReLU)               (None, 1000)              0         \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 1000)              0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 1000)              1001000   \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 101)               101101    \n",
      "=================================================================\n",
      "Total params: 41,906,789\n",
      "Trainable params: 27,192,101\n",
      "Non-trainable params: 14,714,688\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "related_model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 訓練參數"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "base_lr = 1e-4\n",
    "max_lr = 1e-3\n",
    "\n",
    "com_loss_weight = 0.5\n",
    "\n",
    "n_samples = len(reference_data)\n",
    "step_size = 2 * np.ceil(n_samples / batch_size)\n",
    "global_step = 0   \n",
    "\n",
    "m_optimizer = tf.keras.optimizers.Adam(lr = base_lr)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 模型loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "Crossentropy = tf.keras.losses.CategoricalCrossentropy()\n",
    "def descriptiveness_loss(y_true, y_pred):\n",
    "    \n",
    "    loss = Crossentropy(y_true, y_pred)\n",
    "    \n",
    "    return  loss\n",
    "\n",
    "def compactness_loss(num_class, output):\n",
    "    \n",
    "    n = output.shape[0]\n",
    "    loss = 1/(num_class * n) * tf.reduce_sum(n**2/(n-1)**2 * tf.math.reduce_std(output,axis = 1) **2, axis = 0)\n",
    "\n",
    "    return loss\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "@tf.function\n",
    "def update_model(batch_traget_x, batch_related_x,  batch_related_y):\n",
    "\n",
    "    with tf.GradientTape() as m_tape:\n",
    "        \n",
    "        output_related = related_model(batch_related_x, training = True)\n",
    "        output_target = target_model(batch_traget_x, training = True)\n",
    "        \n",
    "        des_loss = descriptiveness_loss(batch_related_y, output_related)\n",
    "        com_loss = compactness_loss(num_class, output_target)  \n",
    "        total_loss =  des_loss + com_loss_weight * com_loss\n",
    "\n",
    "    m_grads = m_tape.gradient(total_loss, related_model.trainable_variables)\n",
    "    m_optimizer.apply_gradients(zip(m_grads, related_model.trainable_variables))\n",
    "    \n",
    "    return des_loss, com_loss, total_loss\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 訓練分類模型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   0: TIME: 24.01  DES_LOSS: 3.8419 COM_LOSS: 0.00453244 TOTAL_LOSS: 3.8441\n",
      "   1: TIME: 12.19  DES_LOSS: 3.3073 COM_LOSS: 0.00500688 TOTAL_LOSS: 3.3098\n",
      "   2: TIME: 11.67  DES_LOSS: 2.5531 COM_LOSS: 0.00628203 TOTAL_LOSS: 2.5562\n",
      "   3: TIME: 11.70  DES_LOSS: 2.1942 COM_LOSS: 0.00698666 TOTAL_LOSS: 2.1977\n",
      "   4: TIME: 11.84  DES_LOSS: 1.9491 COM_LOSS: 0.00784699 TOTAL_LOSS: 1.9531\n",
      "   5: TIME: 11.92  DES_LOSS: 1.7262 COM_LOSS: 0.00867618 TOTAL_LOSS: 1.7306\n",
      "   6: TIME: 12.03  DES_LOSS: 1.6557 COM_LOSS: 0.00734973 TOTAL_LOSS: 1.6594\n",
      "   7: TIME: 12.14  DES_LOSS: 1.5971 COM_LOSS: 0.00694848 TOTAL_LOSS: 1.6006\n",
      "   8: TIME: 11.72  DES_LOSS: 1.4337 COM_LOSS: 0.00789333 TOTAL_LOSS: 1.4377\n",
      "   9: TIME: 11.59  DES_LOSS: 1.5727 COM_LOSS: 0.00720503 TOTAL_LOSS: 1.5763\n",
      "  10: TIME: 11.60  DES_LOSS: 1.4756 COM_LOSS: 0.00530184 TOTAL_LOSS: 1.4783\n",
      "  11: TIME: 11.55  DES_LOSS: 1.1533 COM_LOSS: 0.00604976 TOTAL_LOSS: 1.1564\n",
      "  12: TIME: 11.68  DES_LOSS: 1.5555 COM_LOSS: 0.00640471 TOTAL_LOSS: 1.5587\n",
      "  13: TIME: 11.64  DES_LOSS: 1.5796 COM_LOSS: 0.00653238 TOTAL_LOSS: 1.5829\n",
      "  14: TIME: 11.80  DES_LOSS: 1.4578 COM_LOSS: 0.00439049 TOTAL_LOSS: 1.4600\n",
      "  15: TIME: 11.60  DES_LOSS: 1.4938 COM_LOSS: 0.00497868 TOTAL_LOSS: 1.4963\n",
      "  16: TIME: 11.54  DES_LOSS: 1.3747 COM_LOSS: 0.00484776 TOTAL_LOSS: 1.3771\n",
      "  17: TIME: 11.59  DES_LOSS: 1.5291 COM_LOSS: 0.00510596 TOTAL_LOSS: 1.5317\n",
      "  18: TIME: 11.65  DES_LOSS: 1.6336 COM_LOSS: 0.00430600 TOTAL_LOSS: 1.6358\n",
      "  19: TIME: 11.58  DES_LOSS: 1.6083 COM_LOSS: 0.00395783 TOTAL_LOSS: 1.6102\n",
      "  20: TIME: 11.57  DES_LOSS: 1.5826 COM_LOSS: 0.00391298 TOTAL_LOSS: 1.5845\n",
      "  21: TIME: 11.68  DES_LOSS: 1.8080 COM_LOSS: 0.00359014 TOTAL_LOSS: 1.8098\n",
      "  22: TIME: 11.54  DES_LOSS: 1.6626 COM_LOSS: 0.00313115 TOTAL_LOSS: 1.6642\n",
      "  23: TIME: 11.59  DES_LOSS: 1.9715 COM_LOSS: 0.00337466 TOTAL_LOSS: 1.9732\n",
      "  24: TIME: 11.49  DES_LOSS: 1.7319 COM_LOSS: 0.00321362 TOTAL_LOSS: 1.7335\n",
      "  25: TIME: 11.86  DES_LOSS: 1.6165 COM_LOSS: 0.00369412 TOTAL_LOSS: 1.6183\n",
      "  26: TIME: 11.95  DES_LOSS: 1.9793 COM_LOSS: 0.00360649 TOTAL_LOSS: 1.9811\n",
      "  27: TIME: 12.04  DES_LOSS: 1.8075 COM_LOSS: 0.00285402 TOTAL_LOSS: 1.8089\n",
      "  28: TIME: 12.01  DES_LOSS: 1.8177 COM_LOSS: 0.00292196 TOTAL_LOSS: 1.8191\n",
      "  29: TIME: 11.89  DES_LOSS: 1.7923 COM_LOSS: 0.00302096 TOTAL_LOSS: 1.7938\n",
      "  30: TIME: 11.93  DES_LOSS: 1.9650 COM_LOSS: 0.00270389 TOTAL_LOSS: 1.9664\n",
      "  31: TIME: 12.02  DES_LOSS: 1.9487 COM_LOSS: 0.00302096 TOTAL_LOSS: 1.9502\n",
      "  32: TIME: 12.12  DES_LOSS: 1.7581 COM_LOSS: 0.00229824 TOTAL_LOSS: 1.7592\n",
      "  33: TIME: 12.05  DES_LOSS: 1.8641 COM_LOSS: 0.00148967 TOTAL_LOSS: 1.8648\n",
      "  34: TIME: 11.89  DES_LOSS: 1.5886 COM_LOSS: 0.00158740 TOTAL_LOSS: 1.5894\n",
      "  35: TIME: 11.88  DES_LOSS: 1.3855 COM_LOSS: 0.00149726 TOTAL_LOSS: 1.3863\n",
      "  36: TIME: 12.04  DES_LOSS: 1.3417 COM_LOSS: 0.00177932 TOTAL_LOSS: 1.3426\n",
      "  37: TIME: 11.90  DES_LOSS: 1.0679 COM_LOSS: 0.00154202 TOTAL_LOSS: 1.0686\n",
      "  38: TIME: 11.98  DES_LOSS: 1.1512 COM_LOSS: 0.00195904 TOTAL_LOSS: 1.1522\n",
      "  39: TIME: 11.60  DES_LOSS: 1.1977 COM_LOSS: 0.00194069 TOTAL_LOSS: 1.1987\n",
      "  40: TIME: 11.75  DES_LOSS: 1.2108 COM_LOSS: 0.00203827 TOTAL_LOSS: 1.2118\n",
      "  41: TIME: 11.81  DES_LOSS: 1.0214 COM_LOSS: 0.00132305 TOTAL_LOSS: 1.0221\n",
      "  42: TIME: 11.85  DES_LOSS: 1.0156 COM_LOSS: 0.00156401 TOTAL_LOSS: 1.0164\n",
      "  43: TIME: 12.01  DES_LOSS: 0.8835 COM_LOSS: 0.00180887 TOTAL_LOSS: 0.8844\n",
      "  44: TIME: 11.90  DES_LOSS: 0.8179 COM_LOSS: 0.00184100 TOTAL_LOSS: 0.8189\n",
      "  45: TIME: 11.94  DES_LOSS: 0.9024 COM_LOSS: 0.00216820 TOTAL_LOSS: 0.9034\n",
      "  46: TIME: 12.02  DES_LOSS: 0.8730 COM_LOSS: 0.00152597 TOTAL_LOSS: 0.8738\n",
      "  47: TIME: 11.68  DES_LOSS: 0.7804 COM_LOSS: 0.00192249 TOTAL_LOSS: 0.7814\n",
      "  48: TIME: 11.86  DES_LOSS: 0.7214 COM_LOSS: 0.00172734 TOTAL_LOSS: 0.7222\n",
      "  49: TIME: 11.84  DES_LOSS: 0.7079 COM_LOSS: 0.00200374 TOTAL_LOSS: 0.7089\n",
      "  50: TIME: 11.85  DES_LOSS: 0.6745 COM_LOSS: 0.00221911 TOTAL_LOSS: 0.6756\n",
      "  51: TIME: 11.91  DES_LOSS: 0.5910 COM_LOSS: 0.00236125 TOTAL_LOSS: 0.5922\n",
      "  52: TIME: 11.66  DES_LOSS: 0.5764 COM_LOSS: 0.00200742 TOTAL_LOSS: 0.5774\n",
      "  53: TIME: 11.67  DES_LOSS: 0.5260 COM_LOSS: 0.00209299 TOTAL_LOSS: 0.5271\n",
      "  54: TIME: 11.67  DES_LOSS: 0.5114 COM_LOSS: 0.00178146 TOTAL_LOSS: 0.5123\n",
      "  55: TIME: 11.57  DES_LOSS: 0.4272 COM_LOSS: 0.00180545 TOTAL_LOSS: 0.4281\n",
      "  56: TIME: 11.65  DES_LOSS: 0.3520 COM_LOSS: 0.00174070 TOTAL_LOSS: 0.3529\n",
      "  57: TIME: 11.46  DES_LOSS: 0.3671 COM_LOSS: 0.00193378 TOTAL_LOSS: 0.3680\n",
      "  58: TIME: 11.68  DES_LOSS: 0.3921 COM_LOSS: 0.00188694 TOTAL_LOSS: 0.3930\n",
      "  59: TIME: 11.90  DES_LOSS: 0.4565 COM_LOSS: 0.00183845 TOTAL_LOSS: 0.4575\n",
      "learning rate changed!\n",
      "  60: TIME: 11.54  DES_LOSS: 0.2649 COM_LOSS: 0.00169956 TOTAL_LOSS: 0.2658\n",
      "  61: TIME: 11.67  DES_LOSS: 0.3178 COM_LOSS: 0.00186379 TOTAL_LOSS: 0.3187\n",
      "  62: TIME: 11.54  DES_LOSS: 0.2614 COM_LOSS: 0.00211536 TOTAL_LOSS: 0.2625\n",
      "  63: TIME: 11.57  DES_LOSS: 0.3186 COM_LOSS: 0.00188937 TOTAL_LOSS: 0.3195\n",
      "  64: TIME: 11.62  DES_LOSS: 0.3282 COM_LOSS: 0.00186434 TOTAL_LOSS: 0.3292\n",
      "  65: TIME: 11.78  DES_LOSS: 0.3613 COM_LOSS: 0.00194860 TOTAL_LOSS: 0.3623\n",
      "  66: TIME: 11.93  DES_LOSS: 0.3532 COM_LOSS: 0.00191260 TOTAL_LOSS: 0.3542\n",
      "  67: TIME: 11.87  DES_LOSS: 0.4222 COM_LOSS: 0.00185746 TOTAL_LOSS: 0.4231\n",
      "  68: TIME: 11.65  DES_LOSS: 0.4205 COM_LOSS: 0.00210465 TOTAL_LOSS: 0.4216\n",
      "  69: TIME: 11.62  DES_LOSS: 0.3505 COM_LOSS: 0.00218941 TOTAL_LOSS: 0.3516\n",
      "  70: TIME: 11.54  DES_LOSS: 0.4415 COM_LOSS: 0.00259402 TOTAL_LOSS: 0.4428\n",
      "  71: TIME: 11.62  DES_LOSS: 0.4683 COM_LOSS: 0.00200964 TOTAL_LOSS: 0.4693\n",
      "  72: TIME: 11.50  DES_LOSS: 0.3739 COM_LOSS: 0.00223021 TOTAL_LOSS: 0.3750\n",
      "  73: TIME: 11.55  DES_LOSS: 0.5262 COM_LOSS: 0.00205595 TOTAL_LOSS: 0.5273\n",
      "  74: TIME: 11.74  DES_LOSS: 0.6652 COM_LOSS: 0.00283575 TOTAL_LOSS: 0.6666\n",
      "  75: TIME: 11.66  DES_LOSS: 0.6592 COM_LOSS: 0.00246263 TOTAL_LOSS: 0.6604\n",
      "  76: TIME: 11.63  DES_LOSS: 0.6709 COM_LOSS: 0.00265588 TOTAL_LOSS: 0.6722\n",
      "  77: TIME: 11.60  DES_LOSS: 0.5417 COM_LOSS: 0.00287621 TOTAL_LOSS: 0.5431\n",
      "  78: TIME: 11.64  DES_LOSS: 0.5541 COM_LOSS: 0.00210857 TOTAL_LOSS: 0.5551\n",
      "  79: TIME: 11.73  DES_LOSS: 0.6143 COM_LOSS: 0.00163936 TOTAL_LOSS: 0.6152\n",
      "  80: TIME: 11.59  DES_LOSS: 0.4558 COM_LOSS: 0.00193025 TOTAL_LOSS: 0.4568\n",
      "  81: TIME: 11.64  DES_LOSS: 0.4844 COM_LOSS: 0.00230175 TOTAL_LOSS: 0.4855\n",
      "  82: TIME: 11.91  DES_LOSS: 0.3245 COM_LOSS: 0.00206667 TOTAL_LOSS: 0.3255\n",
      "  83: TIME: 11.86  DES_LOSS: 0.3939 COM_LOSS: 0.00198328 TOTAL_LOSS: 0.3948\n",
      "  84: TIME: 11.73  DES_LOSS: 0.2641 COM_LOSS: 0.00203801 TOTAL_LOSS: 0.2652\n",
      "  85: TIME: 11.59  DES_LOSS: 0.3212 COM_LOSS: 0.00227374 TOTAL_LOSS: 0.3223\n",
      "  86: TIME: 11.66  DES_LOSS: 0.2618 COM_LOSS: 0.00248885 TOTAL_LOSS: 0.2631\n",
      "  87: TIME: 11.75  DES_LOSS: 0.3029 COM_LOSS: 0.00227058 TOTAL_LOSS: 0.3040\n",
      "  88: TIME: 11.67  DES_LOSS: 0.2772 COM_LOSS: 0.00189435 TOTAL_LOSS: 0.2782\n",
      "  89: TIME: 11.74  DES_LOSS: 0.1560 COM_LOSS: 0.00177482 TOTAL_LOSS: 0.1569\n",
      "  90: TIME: 11.62  DES_LOSS: 0.1805 COM_LOSS: 0.00182056 TOTAL_LOSS: 0.1814\n",
      "  91: TIME: 11.72  DES_LOSS: 0.1791 COM_LOSS: 0.00200753 TOTAL_LOSS: 0.1801\n",
      "  92: TIME: 11.82  DES_LOSS: 0.2204 COM_LOSS: 0.00199121 TOTAL_LOSS: 0.2214\n",
      "  93: TIME: 11.64  DES_LOSS: 0.2092 COM_LOSS: 0.00181606 TOTAL_LOSS: 0.2101\n",
      "  94: TIME: 11.76  DES_LOSS: 0.2716 COM_LOSS: 0.00198257 TOTAL_LOSS: 0.2725\n",
      "  95: TIME: 11.59  DES_LOSS: 0.2048 COM_LOSS: 0.00238796 TOTAL_LOSS: 0.2060\n",
      "  96: TIME: 11.63  DES_LOSS: 0.2128 COM_LOSS: 0.00248136 TOTAL_LOSS: 0.2140\n",
      "  97: TIME: 11.74  DES_LOSS: 0.3276 COM_LOSS: 0.00225272 TOTAL_LOSS: 0.3287\n",
      "  98: TIME: 11.66  DES_LOSS: 0.3076 COM_LOSS: 0.00263745 TOTAL_LOSS: 0.3089\n",
      "  99: TIME: 11.76  DES_LOSS: 0.4129 COM_LOSS: 0.00255739 TOTAL_LOSS: 0.4142\n"
     ]
    }
   ],
   "source": [
    "epochs = 100\n",
    "\n",
    "pre_loss = 0\n",
    "des_loss_list = []\n",
    "com_loss_list = []\n",
    "total_loss_list = []\n",
    "\n",
    "for epoch in range(epochs):\n",
    "    start = time.time()\n",
    "\n",
    "    # Learning rate schedule\n",
    "    if epoch in [60, 100, 300]:\n",
    "        base_lr = base_lr / 2\n",
    "        max_lr = max_lr / 2\n",
    "        step_size = step_size / 2\n",
    "\n",
    "        print('learning rate changed!')\n",
    "\n",
    "    des_loss_avg = tf.metrics.Mean()\n",
    "    com_loss_avg = tf.metrics.Mean()\n",
    "    total_loss_avg = tf.metrics.Mean()\n",
    "\n",
    "    for batch_traget_x ,(batch_related_x,  batch_related_y ) in zip(target_dataset, reference_dataset):\n",
    "\n",
    "        # Calculate cyclic learning rate\n",
    "        global_step = global_step + 1\n",
    "        cycle = np.floor(1 + global_step / (2 * step_size))\n",
    "        x_lr = np.abs(global_step / step_size - 2 * cycle + 1)\n",
    "        clr = base_lr + (max_lr - base_lr) * max(0, 1 - x_lr)\n",
    "\n",
    "        m_optimizer.lr = clr\n",
    "        des_loss, com_loss, total_loss = update_model(batch_traget_x, batch_related_x,  batch_related_y)\n",
    "        des_loss_avg(des_loss)\n",
    "        com_loss_avg(com_loss)\n",
    "        total_loss_avg(total_loss)\n",
    "        epoch_time = time.time() - start\n",
    "\n",
    "     \n",
    "    des_loss_list.append(des_loss)\n",
    "    com_loss_list.append(com_loss)\n",
    "    total_loss_list.append(total_loss)\n",
    "    \n",
    "    if abs(pre_loss - com_loss_avg.result()) < 1e-8:\n",
    "        break\n",
    "    pre_loss  = com_loss_avg.result()\n",
    "\n",
    "    if epoch % 1 == 0:\n",
    "        epoch_time = time.time() - start\n",
    "        print('{:4d}: TIME: {:.2f}  DES_LOSS: {:.4f} COM_LOSS: {:.8f} TOTAL_LOSS: {:.4f}'.format(epoch, \n",
    "                                    epoch_time, des_loss_avg.result(),com_loss_avg.result(),total_loss_avg.result()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(des_loss_list, label = 'descriptiveness')\n",
    "plt.plot(com_loss_list, label = 'compactness')\n",
    "plt.plot(total_loss_list, label = 'total')\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# fine-tuning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model_1\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_1 (InputLayer)         [(None, 224, 224, 3)]     0         \n",
      "_________________________________________________________________\n",
      "block1_conv1 (Conv2D)        (None, 224, 224, 64)      1792      \n",
      "_________________________________________________________________\n",
      "block1_conv2 (Conv2D)        (None, 224, 224, 64)      36928     \n",
      "_________________________________________________________________\n",
      "block1_pool (MaxPooling2D)   (None, 112, 112, 64)      0         \n",
      "_________________________________________________________________\n",
      "block2_conv1 (Conv2D)        (None, 112, 112, 128)     73856     \n",
      "_________________________________________________________________\n",
      "block2_conv2 (Conv2D)        (None, 112, 112, 128)     147584    \n",
      "_________________________________________________________________\n",
      "block2_pool (MaxPooling2D)   (None, 56, 56, 128)       0         \n",
      "_________________________________________________________________\n",
      "block3_conv1 (Conv2D)        (None, 56, 56, 256)       295168    \n",
      "_________________________________________________________________\n",
      "block3_conv2 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_conv3 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_pool (MaxPooling2D)   (None, 28, 28, 256)       0         \n",
      "_________________________________________________________________\n",
      "block4_conv1 (Conv2D)        (None, 28, 28, 512)       1180160   \n",
      "_________________________________________________________________\n",
      "block4_conv2 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_conv3 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_pool (MaxPooling2D)   (None, 14, 14, 512)       0         \n",
      "_________________________________________________________________\n",
      "block5_conv1 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv2 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv3 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_pool (MaxPooling2D)   (None, 7, 7, 512)         0         \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 25088)             0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 1000)              25089000  \n",
      "_________________________________________________________________\n",
      "re_lu (ReLU)                 (None, 1000)              0         \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 1000)              0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 1000)              1001000   \n",
      "_________________________________________________________________\n",
      "re_lu_1 (ReLU)               (None, 1000)              0         \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 1000)              0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 1000)              1001000   \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 101)               101101    \n",
      "=================================================================\n",
      "Total params: 41,906,789\n",
      "Trainable params: 41,906,789\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "related_model.trainable = True \n",
    "target_model.trainable = True \n",
    "related_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_1 (InputLayer)         [(None, 224, 224, 3)]     0         \n",
      "_________________________________________________________________\n",
      "block1_conv1 (Conv2D)        (None, 224, 224, 64)      1792      \n",
      "_________________________________________________________________\n",
      "block1_conv2 (Conv2D)        (None, 224, 224, 64)      36928     \n",
      "_________________________________________________________________\n",
      "block1_pool (MaxPooling2D)   (None, 112, 112, 64)      0         \n",
      "_________________________________________________________________\n",
      "block2_conv1 (Conv2D)        (None, 112, 112, 128)     73856     \n",
      "_________________________________________________________________\n",
      "block2_conv2 (Conv2D)        (None, 112, 112, 128)     147584    \n",
      "_________________________________________________________________\n",
      "block2_pool (MaxPooling2D)   (None, 56, 56, 128)       0         \n",
      "_________________________________________________________________\n",
      "block3_conv1 (Conv2D)        (None, 56, 56, 256)       295168    \n",
      "_________________________________________________________________\n",
      "block3_conv2 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_conv3 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_pool (MaxPooling2D)   (None, 28, 28, 256)       0         \n",
      "_________________________________________________________________\n",
      "block4_conv1 (Conv2D)        (None, 28, 28, 512)       1180160   \n",
      "_________________________________________________________________\n",
      "block4_conv2 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_conv3 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_pool (MaxPooling2D)   (None, 14, 14, 512)       0         \n",
      "_________________________________________________________________\n",
      "block5_conv1 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv2 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv3 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_pool (MaxPooling2D)   (None, 7, 7, 512)         0         \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 25088)             0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 1000)              25089000  \n",
      "_________________________________________________________________\n",
      "re_lu (ReLU)                 (None, 1000)              0         \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 1000)              0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 1000)              1001000   \n",
      "_________________________________________________________________\n",
      "re_lu_1 (ReLU)               (None, 1000)              0         \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 1000)              0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 1000)              1001000   \n",
      "=================================================================\n",
      "Total params: 41,805,688\n",
      "Trainable params: 41,805,688\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "target_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "@tf.function\n",
    "def update_model(batch_traget_x, batch_related_x,  batch_related_y):\n",
    "\n",
    "    with tf.GradientTape() as m_tape:\n",
    "        \n",
    "        output_related = related_model(batch_related_x, training = True)\n",
    "        output_target = target_model(batch_traget_x, training = True)\n",
    "        \n",
    "        des_loss = descriptiveness_loss(batch_related_y, output_related)\n",
    "        com_loss = compactness_loss(num_class, output_target)  \n",
    "        total_loss =  des_loss + com_loss_weight * com_loss\n",
    "\n",
    "    m_grads = m_tape.gradient(total_loss, related_model.trainable_variables)\n",
    "    m_optimizer.apply_gradients(zip(m_grads, related_model.trainable_variables))\n",
    "    \n",
    "    return des_loss, com_loss, total_loss\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "base_lr = 1e-6\n",
    "max_lr = 1e-5\n",
    "layer_num = 0\n",
    "inputs = target_model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   0: TIME: 22.19  DES_LOSS: 32.8479 COM_LOSS: 1.00664580 TOTAL_LOSS: 33.3513\n",
      "   1: TIME: 18.80  DES_LOSS: 4.5094 COM_LOSS: 0.00009839 TOTAL_LOSS: 4.5094\n",
      "   2: TIME: 18.92  DES_LOSS: 4.3814 COM_LOSS: 0.00011309 TOTAL_LOSS: 4.3814\n",
      "   3: TIME: 18.64  DES_LOSS: 4.3311 COM_LOSS: 0.00014941 TOTAL_LOSS: 4.3312\n",
      "   4: TIME: 18.74  DES_LOSS: 4.3218 COM_LOSS: 0.00006518 TOTAL_LOSS: 4.3218\n",
      "   5: TIME: 18.58  DES_LOSS: 4.3113 COM_LOSS: 0.00006343 TOTAL_LOSS: 4.3113\n",
      "   6: TIME: 18.86  DES_LOSS: 4.2936 COM_LOSS: 0.00006008 TOTAL_LOSS: 4.2937\n",
      "   7: TIME: 19.00  DES_LOSS: 4.3387 COM_LOSS: 0.00006372 TOTAL_LOSS: 4.3387\n",
      "   8: TIME: 18.68  DES_LOSS: 4.2530 COM_LOSS: 0.00006423 TOTAL_LOSS: 4.2530\n",
      "   9: TIME: 18.73  DES_LOSS: 4.3222 COM_LOSS: 0.00006274 TOTAL_LOSS: 4.3222\n",
      "  10: TIME: 18.71  DES_LOSS: 4.3016 COM_LOSS: 0.00006436 TOTAL_LOSS: 4.3017\n",
      "  11: TIME: 18.88  DES_LOSS: 4.3004 COM_LOSS: 0.00006538 TOTAL_LOSS: 4.3005\n",
      "  12: TIME: 18.80  DES_LOSS: 4.2401 COM_LOSS: 0.00006454 TOTAL_LOSS: 4.2401\n",
      "  13: TIME: 18.60  DES_LOSS: 4.2049 COM_LOSS: 0.00006827 TOTAL_LOSS: 4.2049\n",
      "  14: TIME: 18.74  DES_LOSS: 4.2631 COM_LOSS: 0.00007145 TOTAL_LOSS: 4.2631\n",
      "  15: TIME: 18.48  DES_LOSS: 4.2671 COM_LOSS: 0.00006866 TOTAL_LOSS: 4.2671\n",
      "  16: TIME: 18.64  DES_LOSS: 4.1932 COM_LOSS: 0.00006855 TOTAL_LOSS: 4.1932\n",
      "  17: TIME: 18.73  DES_LOSS: 4.2015 COM_LOSS: 0.00007091 TOTAL_LOSS: 4.2015\n",
      "  18: TIME: 18.62  DES_LOSS: 4.2041 COM_LOSS: 0.00007418 TOTAL_LOSS: 4.2042\n",
      "  19: TIME: 18.75  DES_LOSS: 4.2092 COM_LOSS: 0.00007289 TOTAL_LOSS: 4.2092\n",
      "  20: TIME: 18.74  DES_LOSS: 4.2613 COM_LOSS: 0.00007165 TOTAL_LOSS: 4.2613\n",
      "  21: TIME: 18.65  DES_LOSS: 4.2659 COM_LOSS: 0.00006946 TOTAL_LOSS: 4.2660\n",
      "  22: TIME: 18.77  DES_LOSS: 4.2327 COM_LOSS: 0.00007118 TOTAL_LOSS: 4.2327\n",
      "  23: TIME: 18.63  DES_LOSS: 4.2053 COM_LOSS: 0.00007841 TOTAL_LOSS: 4.2053\n",
      "  24: TIME: 18.78  DES_LOSS: 4.2787 COM_LOSS: 0.00006858 TOTAL_LOSS: 4.2787\n",
      "  25: TIME: 18.66  DES_LOSS: 4.3093 COM_LOSS: 0.00006715 TOTAL_LOSS: 4.3094\n",
      "  26: TIME: 18.64  DES_LOSS: 4.2114 COM_LOSS: 0.00006672 TOTAL_LOSS: 4.2114\n",
      "  27: TIME: 18.73  DES_LOSS: 4.2159 COM_LOSS: 0.00006603 TOTAL_LOSS: 4.2159\n",
      "  28: TIME: 18.72  DES_LOSS: 4.2831 COM_LOSS: 0.00006607 TOTAL_LOSS: 4.2831\n",
      "  29: TIME: 18.73  DES_LOSS: 4.2908 COM_LOSS: 0.00006554 TOTAL_LOSS: 4.2909\n",
      "  30: TIME: 18.61  DES_LOSS: 4.2720 COM_LOSS: 0.00006410 TOTAL_LOSS: 4.2721\n",
      "  31: TIME: 18.64  DES_LOSS: 4.2985 COM_LOSS: 0.00006484 TOTAL_LOSS: 4.2985\n",
      "  32: TIME: 18.76  DES_LOSS: 4.1790 COM_LOSS: 0.00006366 TOTAL_LOSS: 4.1790\n",
      "  33: TIME: 18.66  DES_LOSS: 4.2261 COM_LOSS: 0.00006546 TOTAL_LOSS: 4.2261\n",
      "  34: TIME: 18.76  DES_LOSS: 4.2526 COM_LOSS: 0.00006486 TOTAL_LOSS: 4.2526\n",
      "  35: TIME: 18.54  DES_LOSS: 4.2640 COM_LOSS: 0.00006434 TOTAL_LOSS: 4.2640\n",
      "  36: TIME: 18.62  DES_LOSS: 4.1649 COM_LOSS: 0.00006484 TOTAL_LOSS: 4.1650\n",
      "  37: TIME: 18.75  DES_LOSS: 4.2788 COM_LOSS: 0.00006588 TOTAL_LOSS: 4.2788\n",
      "  38: TIME: 19.13  DES_LOSS: 4.3393 COM_LOSS: 0.00006339 TOTAL_LOSS: 4.3393\n",
      "  39: TIME: 18.74  DES_LOSS: 4.2622 COM_LOSS: 0.00006143 TOTAL_LOSS: 4.2623\n",
      "  40: TIME: 18.62  DES_LOSS: 4.2302 COM_LOSS: 0.00006182 TOTAL_LOSS: 4.2303\n",
      "  41: TIME: 18.58  DES_LOSS: 4.2440 COM_LOSS: 0.00006341 TOTAL_LOSS: 4.2440\n",
      "  42: TIME: 18.75  DES_LOSS: 4.2168 COM_LOSS: 0.00006457 TOTAL_LOSS: 4.2168\n",
      "  43: TIME: 18.53  DES_LOSS: 4.2543 COM_LOSS: 0.00006261 TOTAL_LOSS: 4.2543\n",
      "  44: TIME: 18.73  DES_LOSS: 4.2118 COM_LOSS: 0.00006322 TOTAL_LOSS: 4.2118\n",
      "  45: TIME: 18.78  DES_LOSS: 4.2175 COM_LOSS: 0.00006344 TOTAL_LOSS: 4.2175\n",
      "  46: TIME: 18.68  DES_LOSS: 4.2069 COM_LOSS: 0.00006332 TOTAL_LOSS: 4.2070\n",
      "  47: TIME: 18.64  DES_LOSS: 4.1651 COM_LOSS: 0.00006723 TOTAL_LOSS: 4.1651\n",
      "  48: TIME: 18.49  DES_LOSS: 4.1948 COM_LOSS: 0.00006212 TOTAL_LOSS: 4.1948\n",
      "  49: TIME: 18.62  DES_LOSS: 4.2334 COM_LOSS: 0.00006273 TOTAL_LOSS: 4.2335\n",
      "  50: TIME: 18.78  DES_LOSS: 4.2444 COM_LOSS: 0.00006172 TOTAL_LOSS: 4.2445\n",
      "  51: TIME: 18.68  DES_LOSS: 4.2779 COM_LOSS: 0.00006091 TOTAL_LOSS: 4.2779\n",
      "  52: TIME: 18.79  DES_LOSS: 4.2695 COM_LOSS: 0.00005912 TOTAL_LOSS: 4.2696\n",
      "  53: TIME: 18.61  DES_LOSS: 4.2844 COM_LOSS: 0.00005766 TOTAL_LOSS: 4.2844\n",
      "  54: TIME: 18.64  DES_LOSS: 4.1886 COM_LOSS: 0.00005747 TOTAL_LOSS: 4.1886\n",
      "  55: TIME: 18.78  DES_LOSS: 4.3590 COM_LOSS: 0.00005655 TOTAL_LOSS: 4.3590\n",
      "  56: TIME: 18.63  DES_LOSS: 4.2668 COM_LOSS: 0.00005521 TOTAL_LOSS: 4.2668\n",
      "  57: TIME: 18.76  DES_LOSS: 4.2891 COM_LOSS: 0.00005481 TOTAL_LOSS: 4.2891\n",
      "  58: TIME: 18.65  DES_LOSS: 4.2227 COM_LOSS: 0.00005568 TOTAL_LOSS: 4.2227\n",
      "  59: TIME: 18.68  DES_LOSS: 4.1981 COM_LOSS: 0.00005689 TOTAL_LOSS: 4.1982\n"
     ]
    }
   ],
   "source": [
    "epochs = 60\n",
    "\n",
    "pre_loss = 0\n",
    "des_loss_list = []\n",
    "com_loss_list = []\n",
    "total_loss_list = []\n",
    "\n",
    "for epoch in range(epochs):\n",
    "    start = time.time()\n",
    "\n",
    "\n",
    "\n",
    "    des_loss_avg = tf.metrics.Mean()\n",
    "    com_loss_avg = tf.metrics.Mean()\n",
    "    total_loss_avg = tf.metrics.Mean()\n",
    "\n",
    "    for batch_traget_x ,(batch_related_x,  batch_related_y ) in zip(target_dataset, reference_dataset):\n",
    "\n",
    "\n",
    "        des_loss, com_loss, total_loss = update_model(batch_traget_x, batch_related_x,  batch_related_y)\n",
    "        des_loss_avg(des_loss)\n",
    "        com_loss_avg(com_loss)\n",
    "        total_loss_avg(total_loss)\n",
    "     \n",
    "    des_loss_list.append(des_loss)\n",
    "    com_loss_list.append(com_loss)\n",
    "    total_loss_list.append(total_loss)\n",
    "    \n",
    "    if abs(pre_loss - total_loss) < 1e-8:\n",
    "        break\n",
    "    pre_loss  = total_loss\n",
    "\n",
    "    if epoch % 1 == 0:\n",
    "        epoch_time = time.time() - start\n",
    "        print('{:4d}: TIME: {:.2f}  DES_LOSS: {:.4f} COM_LOSS: {:.8f} TOTAL_LOSS: {:.4f}'.format(epoch, \n",
    "                                    epoch_time, des_loss_avg.result(),com_loss_avg.result(),total_loss_avg.result()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(des_loss_list, label = 'descriptiveness')\n",
    "plt.plot(com_loss_list, label = 'compactness')\n",
    "plt.plot(total_loss_list, label = 'total')\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD4CAYAAAD8Zh1EAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8li6FKAAAgAElEQVR4nOy9eZQkd3Xn+72R+1pZe1d3Va/qFtq6W6K1jdACCCxAI8/wGCzmYTPPGNljjw0Dzzwzj8eM8TvHx+bZYJ9jGwO2wSODDIyxhWzAArQhUEstqbsltdT7vtSeVblnZObv/RHxi4zMjMiMyIzIrMz6fc7RUVVmdmVkVeaNG9977/cSYwwCgUAg6H+kXh+AQCAQCJxBBHSBQCAYEERAFwgEggFBBHSBQCAYEERAFwgEggHB26snHhsbY1u3bu3V0wsEAkFf8uKLLy4wxsaN7utZQN+6dSsOHDjQq6cXCASCvoSIzprdJyQXgUAgGBBEQBcIBIIBQQR0gUAgGBBEQBcIBIIBQQR0gUAgGBBEQBcIBIIBQQR0gUAgGBBEQBcIbHAxmcOP35jt9WEIBIaIgC4Q2OBvfnIav/53L/X6MAQCQ0RAFwhssJQpIi9XIJcrvT4UgaABSwGdiM4Q0StEdJCIGub1SeFPiegEER0mopucP1SBoPcsZ4sAgGyx3OMjEQgasePl8lbG2ILJfe8CsFP971YAf6H+XyAYKJazMgAgVyxjKOTr8dEIBLU4Jbn8PIC/ZQrPAUgQ0ZRDP1sgWDOs5JSAnimWenwkAkEjVgM6A/CvRPQiET1kcP8mAOd1319Qb6uBiB4iogNEdGB+ft7+0QoEPYZLLjkhuQjWIFYD+h2MsZugSCu/QUR31d1PBv+GNdzA2JcYY/sYY/vGxw3tfAWCNUu5wrQMXWjogrWIpYDOGLuk/n8OwHcA3FL3kAsAZnTfTwO45MQBCgRrhVReBlPTFCG5CNYiLQM6EUWIKMa/BvBOAK/WPexRAL+kdrvcBmCFMXbZ8aMVCHoIL4gCQnIRrE2sdLlMAvgOEfHHf50x9n0i+jUAYIx9EcC/AHg3gBMAsgD+D3cOVyDoHVw/B4BMQWTogrVHy4DOGDsFYI/B7V/Ufc0A/IazhyYQrC1W9Bm63F8ZeqZQwrv+5Bl87n27cev20V4fjsAlxKSoQGARfYbeb0XR+VQB55ayeO3Saq8PReAiIqALBBbRa+jZPpNc8iXlBJTus+MW2MPOpKhAsK5ZyRZBBAS8Ut9l6HlZ8Z4RAX2wEQFdILDIclbGUMgHryQh22caOu/KSeVFQB9kREAXCCyynC1iOOxHucKE5CJYk4iALhBYZCUnIxH2IVcs953kUlCvKNJ5ucUjBf2MCOgCgUWWs0WMRwMg9F+Xi9DQ1weiy0XQktMLGTx9TJipLWdkDIf9CPu9yPbZ6H9eFhr6ekAEdEFLvvjkSXz0kZd7fRg9R5Fc/Aj7PX2YoQsNfT0gArqgJclcEctZGcXS+l27VixVkC6UkAj7+jOgl4Tksh4QAV3QEn6Zvpgp9PhIekcyp0yJDod9CPm9jgb0P/3RcTxxdM6xn2eEXnJhrMHZWjAgiIAuaAkP6AupYotHDi7cxyUR9iPi9yDnoIb+N8+exmOH3DUn5UXRcoVpXwsGDxHQBS1Jqa1uC+n1m6EvawFdlVzksmOZbqFUcd29Ma8bhEoVROvioCICuqAlPEOfX9cBnUsufoT8XjAGxzLdQqni+sIMfUBPi06XgUUEdEFLNMllHQf0FV2GHgl4ADiztUguV1CusK5m6KIwOriIgC5oSl4uo1hWMtH1rKHzDD0R9iPkUwK6E1uLCmr3SabgbteM/mpCZOiDiwjogqboB1HalVwyhf7vrFjOyvB5CBG/B5GAMmDtRKcLH8l3XXIpleGVlF3uKZGhDyyWAzoReYjoZSJ6zOC+zUT0hHr/YSJ6t7OHKegVKZ33x0LKfkBfzhTx5v/3cTzV55OmK7kiEmE/iAghv3OSSzVDd19yGY36AYgMfZCxk6F/FMDrJvd9GsA3GWM3AngQwJ93emCCtQHP0P1eqS0NfS5VQF6u4PxyzulD6yrLGRmJkA8AEHZQcuHadjckl7FoAIDQ0AcZSwGdiKYBvAfAV0wewgDE1a+HAFzq/NAEawEe0LeNRtoK6Gm1Ra7QZ/7h9XDrXADOSi5qhl4sV1ydxM3LZRHQ1wFWM/QvAPgkALN33P8A8EEiugDgXwD8ptGDiOghIjpARAfm5/v7Eny9wCWXbWMRLGdlyGV7QSetZp75Pg/o3DoXgCa5OGHQVdAFcTcNvwqlCuIhH/weSRh0DTAtAzoR3Q9gjjH2YpOHfQDAVxlj0wDeDeB/ElHDz2aMfYkxto8xtm98fLztgxZ0Dy1DH48AAJYy9jpduDac6/OAvpwtagE9rAV054qigLuZc65YRtArIRb0aldNgsHDSoZ+B4AHiOgMgEcAvI2IHq57zIcBfBMAGGM/AxAEMObgcQp6xKqaoW8fUwL6vM3CKA9S/TxuzhjDclbWJJewX5FcnChk6jN0N3X0fKmMkN+DaNAriqIDTMuAzhj7FGNsmjG2FUrB88eMsQ/WPewcgLcDABFdAyWgC01lAOAZ+lYe0G3q6Dx49HOGnpcVfTuhBXTni6KAu62LebmMoM+DaMArNPQBpu0+dCL6LBE9oH77CQAfIaJDAL4B4D+xfm88FgBQAno04MVETCmo2W1dzGgZev8G9OrYvyK5+DwS/B5nFkXXZujuBFrGFEOuoFdCNODFqsjQBxZbK+gYY08CeFL9+jO6249AkWYEA0YqLyMW9GodEgtpexp6Ws06C30suVSnRH3abSG/x5FF0d0I6Pw5Aj4PYkEvLiXzrjyPoPeISVFBU1L5EmJBLyIBL0I+j+3WxUEoiuqtczlOLbkolHSSi0saOj+ZCsll8BEBXdCUVEFGLKhkpmMxfxsBvf/bFrl17rAbAV135eKWhp5XTxpBn6QURUVAH1hEQBc0hWfoADAeDdgO6KkBKIoaSS5OLYrOl9xvW+Qn06DXg2jAJ7pcBhgR0AVNUQK6mqFHA7YdFzMD0La4kqta53JCDmfoEgFZlySXvE5yiQW9KJYrNVKPYHAQAV3QFF4UBYCxWMB222JGK4r2bwBZzhQR9nsQ8Hq02yKOaegV+L0SIi5q2/zqKORXulwAYdA1qIiALmjKqk5yGYsGsJwtomRj/D89AEXR5WzVmIvjlORSKJUR8EqI+L2udbnUSi5qQBc6+kAiArrAlEKpjGKpgrgquYxH/WDM3vg/zwT7uSjKrXP1hPwexxZcBLweRAIe94qi6u8+4FMmRQGsez+XS8kcypXBG5VZdwH92RMLmEuJPlwr8A+9PkMH7E2LDkLb4nJWxnCkNkOP+D3IODQpGlAHftxqW6xq6BJiIkPHfKqAe/6/J/EnPzzW60NxnHUV0OVyBR/66+fxtZ+e6fWh9AUNAT1mb7ioUmFa0MvLlb7dWrScLSIRqs/QvY5l6EGfhLCLkktBa1usZujrWUN//MgsiqUKvvrTMwN3YltXAX0uVUCpwrCUEW5zVuDWubEAl1zsjf/z0Xg+Ml9w0e/bTVayck2HC6Bk6MVyxbadcD0FmUsu7hVFNQ3d59E6lgYtkNnhe69eRjyoWCB8Y/+5rj53tljCb/zdS3jmuDtWV+sqoF9ZUaSW1bwI6FYwz9CtBXSecXKpph91dMYYkjm5ZqgI0Huid/aaCqUyAj4J0YAzXTNGaJKLt9rlsl73iq5kZfzs5CI+cOtm3LZ9BH/1k9OuLhapZyFVxD+/clmLRU6zLgP6eiwInVvM2upOAXQZuprVRfweBH2SZQtd/nuuBvT+y9BX8yWUK6whQ+cWup3KLkqGLiEc6EKXi9qHDqxfyeWHr8+iVGF41/VT+LW7d+DKah7/dPBi155/IaN8dvhnwmnWV0BfVTP03PrK0FfzMu79/FP4zsv23rirdRk6ESnDRXYzdDWz78fCqJGPCwBEAs4silbaFt31WMnpAnrAK8Er0bpdcvH9165gaiiI3ZuGcPeucbxpQwx/+fQpVLrU8bKo1p/4wm6nWV8BfUVZVLzeJJf5VAHFUsW2yx7PsHnbIqBOi1osivKAPhpR3rz9KLnUW+dyQg4tiuZF0Yjfi0KpYvsqygp5uQK/R4JHIhARokHvurxKzRRKePrYPH7uug2Q1N/Fr929Ayfm0vjRG3NdOYbFtMjQazi3mMUfP36srY6JK6vKL3M1t77ezEk1y7R7IuOSC++MAGArQ09rGnr/B3QzyaVzDb3ahw7AkVbIevKyotNzooH1ubXoiaNzKJQquO/6Ddpt9++ewqZECF986mRXjoF/dkYiIkMHAPzLq5fxpz86jq88c9r2v51dp0VRLjGt2JSaUvkSIn4PPBJpt43H2gnofSy55Iwll7BTkovahx4JOLfWruE5Ssq2Ik404F2XRdHvv3oFoxE/bt46ot3m9Uj4yJ3b8OLZZRw4s+T6MSyki4gFvDV/Dyfpu4D+q3dtx7tv2IDf/97reOqYvdafy6uK5FIsVfoyW2yXZE7JMu0H9Kp1Lmc86sdSpmhpyq6+y6Ufl1wsZ7jkUhfQHVpDly9VEPBVA7oTdgINzyErsg4ntg73iublMp54Yw7vvG6yJkEBgPffPIPhsK8rWfpipuiafg7YCOhE5CGil4noMZP7309ER4joNSL6unOH2PA8+Nz79mDXZAy/+fWXcHohY+nfMcYwu1rQ2rbWU5auSS5tZOixYO1Sq7FYABWL4/9pdfKRF0X78STKvdDjdb+HiEOLopUM3YOomvGnXZgWzctlBL21Gfp660P/yfEFZIpl3Hf9VMN9Yb8Xv3T7Vvzw9Tkcm025ehyL6QJGXdLPAXsZ+kcBvG50BxHtBPApAHcwxq4D8DEHjs2USMCLL//SPngkwq987QVN623GclZGsVTBzskogPWlo/OA3o7k0hDQo9Z70TOFEiSCZmzVr5JLPOiF11P7UeF96J2+Jv2kKOCO5MIXRHOiQd+6C+jfe/UK4kEvbt8+anj/h/7NVgR9Eh5+7qyrx7GYLmpNAm5gKaAT0TSA9wD4islDPgLgzxhjywDAGHO9ZDwzEsaf/+9vxpnFLD72yMGWEsBltcNl10QMgPsZ+om51JoZdeeB3H6G3ii5aH4uFnrR0wVlwTQPfv3Yh76cLWLY4AMYdmCwqFSuoFRhWtsi4M4EZ73kEg2sry4XuVzBD1+fxb3XTMLvNQ55IxE/9m0ZwYtnl109lsXM2sjQvwDgkwDMPpG7AOwiomeJ6Dkius/oQUT0EBEdIKID8/Odj77evmMU//3fXosfvTGHP/rXo00fO6v2oFczdPcC+sn5NO7946fxrQMXXHsOOyTVTg27295XDTN0JbhZydB5QOeX+/2YoRtZ5wKKFS0ROloUXVRbFPVFUVc09LqiaCzoXVd96M+dWsRKTq7pbjFi9/QQjl5JuSYNlisMS5mi9hlyg5YBnYjuBzDHGHuxycO8AHYCuAfABwB8hYgS9Q9ijH2JMbaPMbZvfHy8zUOu5Rdv24IP3DKDP3/yJI430b+urCgBaNckz9Ddy1AuLCtXA196pnsDC81IqievdKFkq8/ZMEO3Mf6fKZQQCXgR9Ctvs37U0Feyjda5ACBJhJCvs3F9fsWi+KG7p6Hnio1dLnm5cx+afuH7r15B2O/BXbuax5zd0wmUKgxHLq+6chzJbBEVhp5LLncAeICIzgB4BMDbiOjhusdcAPBPjDGZMXYawFEoAd51iAgf+jdbAQCvX2kW0HOQCNgx4X6GzocHTsylbXfiuAHX0AF7J7LVfKmhGBgLeBHwSpaGi9JqQPd7JBD159ai5azcMFTECXdooctdEAM+j8tti5WGDN2t51prlCsMP3htFm+9eqJlq+Du6SEAwOHzSVeOZTHDp0R7KLkwxj7FGJtmjG0F8CCAHzPGPlj3sH8E8FYAIKIxKBLMKYeP1ZQtIxEAwJkmHS9XVvMYiwa0s6ObGro23hvx48vPdO3XYMpKTgZR9Wsr8OUW9ZKLNv5vQUPPFBTJhogQ9Hr6VHIxztABpTsi14FEUtD5lId8nUs4ZihdLrUaOrA+PI0Onl/GQrqAn2shtwDA1FAQY9EADl9YceVYFlyeEgU66EMnos8S0QPqtz8AsEhERwA8AeC3GWOLThygFUJ+DzYOBZu2MF5eyWNqKIiAV4LfI7na5bKYKcLvkfDQXdvx05OLeO2SO28QqySzRWwcCgGwfmVSdVpszE6t7hZNF0pae1/I7+m7omipXEEqX2qYEuWEO9wryu2EA14PJIkQ9nnca1s0yNDXQ6fLU0fnIRFw987WEi8RYc/0EA5ftP95zRXL+OnJhaaP4YleTzV0PYyxJxlj96tff4Yx9qj6NWOMfZwxdi1j7AbG2CNuHGwzto1HcKpJQJ9dzWMyHgQRIR7yupyhFzAS8ePBWzYj4ve0NdXqFJUKw0pOxsyIEtCtZuj11rl6xqN+S5JLplDWpISgV+q7DJ3/ruqHijidBnRtNZyaPUdcclxs7HJZP57oTx1fwJ6ZBIZMTsr17J5O4OR82vbv5lsvnsd//PJ+rfnCCC7FroUulzXPtrEITs+nTVsFr6zksWEoCEAxm3JVQ1enwYZCPrz/5hl899AlrW2y26SLJVQYsHkkDMBOQK+1ztUzFg3YaFtUMsOg39N3RdFlzWnRLEPvbFG0PkMHFCnE6b2ijLGGLpf1srVoOVPE4QtJ3GUhO+fsnhkCY8ArNmWXU/NKMnlhOWv6mIV0sWYuww0GKKBHsZovGU4wZoslrOZLWkCPhXyudrnop8F++Y5tqDCGr/Zo7R23f90yqtQZrF6ZNMvQx6IBLGUKTXv/GWNalwugtPn1m+SykuPGXMYZeqhjyYXb2rqXoRfLFTCGhi4XwJ060vHZFN7xx0+tib29z55cAGNo2d2iZ/cmtTB6wV5h9PySEsgvNnE0XcwUMBIJQKqzHnCSAQroSgZ6ZrFRduGLLTbEeYbudT1DH1OLrzMjYbzr+il8ff+5nlzi8g6X9jN0o4DuR4VVnQiNKJSUoRmeDQZ9Uv9l6BkuuRhnVJFOA7pcm6GH/R7HF0XrWyM5bmror1xcwfG5NH52smslNFOePjaPeNCLPWr3ihVGowFsSoRsF0bPqQH9UtL8Snwh7W4POjBQAV1pR+SXPnr4YgstoId8rne56O0xf+XObUjlS/jWgfOuPacZ3Jhrw1AQfq9kOaCvGnihc8Zjyu+xWS86DxY8Gwy1KbmcXczgoEttZK2oeqGbZeheZ4qiaobuhscKbxXl07r8eQB3JBf+/urV34zDGMPTxxbwlp1jDbYNrdgzM4TDF60fP2NMC+iXmwR05cpdBHRLTA+H4JXIsNNFy9BrNHR3suVssYScXK4pfNy4eRj7tgzjr589bcml0El4hp4I+Wy9bqPlFhxtWjRlnqFz6YB3ubTbtvjpf3wVH/nbAz2xUeDByaygphRF238fGRVFnZ4U1bYV6cy5wn6lRdKNDH2tBPTjc2lcWc3b0s85u6cTOL+Us2RAByg2GPzk3FxyKWI04l5BFBiggO7zSNg8EjYO6Kt1Ad3FLhezFVO/cuc2nF/K4fEjV1x5XjOSuqA0FLIuNRktt+BYmRblwULT0NvI0PNyGftPL2E+VcDFJpmPWyxni/BKhFig8XcAKJJLTi63fbKpL4pGAs63LWoLonUaOhG55ufCE4bXLq12dflyPU+rA3129HOONmBkUUfn2XnAKzWVXBbT7lrnAgMU0AG108UgoM+u5BEPejVHu3jQ55onOp8Gq9fK7r1mEkTAkcvu2nPWs6LKBkMhH4ZCPltti/XLLThWHBe5Fsz12naKoi+cWdKCwqHz3e/lX87KSIR9IDIuYoX8XjDWvulYQ1HU73xRtLoguvajHnPJQpe/v4qlCt644s4IvRWeOjaPqyai2JgI2f6312uFUWvvufNqZ8tNm4dNu9nychnpQsnVoSJgwAL61rEIzixmGvxTLutaFgFFQwfcqfIvaiumav9wXo+EaMDdYqwRyayMsN+DgNdjq3Zg5OPCiQeVcf5mrYvc/EnL0Nsoij5zfAE+D8HvlXDwvLsueEYspYsYatJi1umi6MYM3YucXHZUlqsG9Nqx96hLSy5WcrLW5tkr2SUvl/H86aW25BZASfi2j0esZ+iLORABN28bwXJWNpTNtLF/F31cgAEL6NvGIsjLFU1i4fChIg73J3FDR9eP/dczFHK3/92IZK7qFmg3QzfqcAH4+L+/6bQolw54H3rIZ19yefrYPPZtGcH1G+Ndz9BXsjKeOT6PPdMNHnManS6K5l0ufk1D55a8zr0v86WqvYAet5ZcrOZlXD0Zw1g0gIPnehPQ959eQqFUwV27xtr+GXumE5Yz9HNLWUzGgtg+prQGGy1jd3s5NGegAjr/hdbLLnzsn+Nmhr6Q4dNgjQE9HnS3u8aIZFbGkNql4VRAB/huUQtFUS1Dt6c3z6XyeONKCm/ZOYY9Mwm8cnHFllNkpzy8/ywyxTI+ctd208d0uig6XyrD5yFN1qoadDknBVYLr/UZus+VvaKrORlDIR/2ziRw0GYvt1M8fWwefq+EW7cZL7Owwu7pIcylClpDRTPOL2WxeSSsxRgj2cWstuY0AxXQt40rAV1vAVAqV7CQLmgti0C1c8ONbHkxXUTY79E+7HriIa/trUGdsqrL0PmErJWg2kxyAdDSoCtj0LZYYYBcthbQnz2h+GLctXMce2cSyMllHJtNW/q3nZKXy/ibZ8/g7l3juGYqbvq4ThdFF+RKTaDlvysnp0XNJBdlr6jz78WVnIx4yIcbNydwaj6jDbZ1k6ePzePWbSM1rZp22a1emR2ycFI6t5TFzEhY0+uNCqPdMOYCBiygT8aCCPoknNb1os+nC6gwYMNQtTgyFOKTcs5nKEuZ2h50PYrk0t3homSuqgMPhXyoMGvtaq0y9LFooGlRlHdQ8LZF3pqXL1nLPp85voDhsA/XbYxj74zy4eqWJvudly9iIV3ArzbJzgEg3KnkUirXSCFO7SmteQ5V1qkPbm4VRXmGvsdGQHSSS8kcjs+l29bPOddOxeGRqKWOnpfLmE3lsXkkjA1DQRAZty5WrXNFhm4ZSSJsHY3UTIte1nrQq2dGNzP0hSZLYHslufAiFQ/sVq4SlG1FTTL0mB+LmaLpAo9MoYSw36ONOfMMMW8h+DHG8JPjC7jjqjFIEmHzSBjDYR8OORDQGWP4w++/gRfOLBneX64wfPnpU7hh0xBu39H8kr3TPaCFUm2GHtYWRTsXaKt96AYausMJjVyuIFMsYyjkw+6ZIRDBkb+ZHZ453n67op6Q34Ndk7GWOvrFZA6MAZtHQ/B5JEzEAobDRQupAkI+4yt3JxmogA4A28drWxdn1YBeUxR1tculOvZfjx0N2wkYY0jmZG0wJm4joKfycsNyCz3j0QDKFWY6/p8pVn1cgGoB0UqL37HZNOZSBS3LIiLsmUk4kqFni2X8+ZMn8ctffQEn5holnMePzOLUQga/evd203ZFDg/A7bpIKgG90ac864KGbtTlkik621HDE6R40It40Icd49Gud7o8fWwBG+JB7FJXTXbCnukhHL6w0lSi5D3oM8OKtcbGRAiXjDT0jPs96MAABvRtYxGcW8pq67V4x8uUTnJx0xNdWQJr/IeLh3zIFstdW/2VlysolipIhPzq81vr7imWKigYLLfQM6GeIOdMdPR0oawFKKAaUKwEP55lvWVntUth70wCx+ZSHWevfNAqlS/hw197Acu6aUDGGL741ElsHgnjvutaL0TodFF0Xi7XLC2OuKKhNw4WAe7o9VzC5AnEnmnlJNytKd9yheEnJxZw586xlidjK+yeTmAlJ2tB2whuysW9kjYmQoZdLs2u3J1kAAN6FOUK037RV1by8HulGoMltzzRGWOqhm4muXR3U0xScwu0J7k0s87lTKjTomb+zxl1QTQnZGOv6DPHF7BjPFIzFLJnJtGWrWk9fGH2Q3dtx+WVPH7t4Re14aUXzizj4PkkPnLnNkv+H45ILrpAyzV0JyWX+k4aTswFC13NKkF9n+3dnMBipqjt2HWbQxeSWMnJHcstHD4xeqjJe+78UhYBr4Rx9fOwcSiIS8lcw0ms2ZW7kwxgQK91XbyymsdkPNBwxnbDE301X4JcZqaOajxz6ZbsovdxAaoftFavu5l1LmeyVYaeL2l91UDVS6RVQFfG/RdxZ11Ra69DRTb+u7/n6nF87n27sf/0Ej79j6+AMYa/fOokRiJ+vO/NM5Z+Fs/Q2+9Dr10Npw0qORnQ5XKNjwuHL7lwMrlY0SQX5Wff2OVi9hvqFPZNW4Yd+XlXb4jB75Wa7hg9p7Ys8viyMRFCoVRp8IFpduXuJJYDOhF5iOhlInqsyWPeR0SMiPY5c3j2qXddvLKSx1S8cfzXDU/06kYSE8nFxWKsETyg12vora5Mmq2f4/CMxGxaNF2XoQcsSi4vnV1GXq7gzp21QyHDET+2jIY7HlZZ0U5yfvz83k34rbfvxDcPXMB/+86r+NEbc/jQ7Vstt7v5PBJ8HkK2Ew1dl6FXM34nNfTa5+BoSy4Kzr0XV+sy9Ks3xBDwSl0L6EsZ3hroTOD0eSRcOxVvmkScW8ppcgsAXeti9cqVMab6uKwtyeWjAF43u5OIYgB+C8D+Tg+qE4bDimcJL4xeWc1jUjdUxHHDE7063msiudgoSjqBtqBB1dCjfi8ksiO5mGfoQZ8H8aDXXHJpsyj69PEFeCXCrdsbO0z2ziQcy9C5DPWxt+/Ee3ZP4RvPn0PQJ+EXb99i6+eF/d62Fzvn5XJNUdQjEUI+j8Nti+WGKVHAnUXR9ZKLzyPh+k1DXet0WUgXEQt4G4aoOuHmrcM4dH7F8MqSMUXandEHdLVWpy+MruZKKFWY6z3ogMWATkTTAN4D4CtNHvZ7AP4QQE9XlRCRZtLFGFMydKOA7oInOp8Ga9aHDrjTXWNEsm6FmiQR4hY6bVYtSC6AIrvMrRpn6PUaOg8qrSSXn5yYx01bhmv+LWfPdAKXV/JN9za2IlkXdCSJ8Ef/YQ/uvWYCH+V7VZUAACAASURBVH37LtO/nRmd7BUt1nW5AOrWIkdH/8vayVSPG0suNMlF53+zV53y7UYjwFKmiBGHZY3bto+iWK7gpXONXkLLWRnpQqk2oCeUWKMfLlpw+MqhGVYz9C8A+CQAw78KEd0IYIYxZirHqI97iIgOENGB+fl5e0dqg+1qQE9mZRRKlZqWRY4bnuiLmebTYFXJpVtF0cadmFZqBzxDN/JC1zMRD5iuGkvl64uirTX0xXQBr15cxZ1XGXtw7N3cuSabzMrweUjTvwHlauMrH7oZ//meHbZ/Xidr6AqlikH3ibNbi5QF0UYauvNF0dW8DL9Xqnm+PTMJFEoVHL3ivsvoUqbouPnVvq0jkAjYf6pxbqG+wwVQkrl6G92qv9MayNCJ6H4Ac4yxF03ulwB8HsAnWv0sxtiXGGP7GGP7xsedqUQbsXUsgssreZxWC6MbjAK6C10uVjP07kkuMvweqSZDs9ILb6UoCgATsSBmDTL0Ullpe9RLLlaKoj9Rx/3vNOlSuHYqDq9EHQX0lZyMoZDfkbY2QOlMaddMq1AqG2foTg4WFU0kFxcydD4lqqebhdGFdMG0w6xdhkI+XLsxjudONa7UO2cQ0IlI7UWvJjqtamtOYiVDvwPAA0R0BsAjAN5GRA/r7o8BuB7Ak+pjbgPwaG8Lo4qnCz+rbjDU0J33RF9MFxRrWa/xrzXoU4po7ZxIfnhkFr/9rUO2enoVY65aP287Ad1I9tAzEQ9gPlVoOCaeYUYM+9DNL71/cnwBQyEfbthkvAMy6PPgmql4R5rsSq6oWT84QUcZep2XC6CcIJxuWzTK0HmLpJMa+mqu1DCMNj0cwkjE35WA7kaGDgC3bRvFy+eTDbGCB/Tp4dqmi42JYJ3k0p2xf8BCQGeMfYoxNs0Y2wrgQQA/Zox9UHf/CmNsjDG2VX3McwAeYIwdcOugW8ED+k9PKhmfYUB3Qc9WpsHMMwQiQjzY3rToPx26hG+9eAEvnrXuC76SK2otixzlyqT5hziVVzzUW/ViT8SCKJYrDa8nrWas+k0/mpdLkxPoKxdX8OYtw4ZLNTh7ZxRb03YnHBUrBOc+WJ1o6PlSWdsnyokEOls83fAcBicNQCnARvwexzX0+gydiBTnRZcDOp8BcSNo3rp9FMVSpeE1nF/KYizqr0lcAKUwWiu5qDsSHHzfmdF2HzoRfZaIHnDyYJyCB/QDZ5ZBVB2C0eOGJ/piunWG0K4n+kl1TP0bz1tfNJ3MNn7ArGboreQWQD9cVCu7cF1W/0aXJELA23zJxWpOblmU3DOTQLpQwqn59pwXV3Tuk07QruRSrjDIZea65GLW5QIobalODxYZLQTZO5PAyfm0q80AvJPEblHbCrdsHQEZ6Ojnl2s7XDhTiRDmUgWtELyQLmA47LO9rLodbD0DY+xJxtj96tefYYw9avCYe3qZnQPKh2IyHkBOLmMsGoDP4BfJM/SUoxl66+GBdvrfKxWGUwtpSAT88yuXLGf4emMujpUul1ShuXUupzpcVFsYre4Trc0MQy32iq7mSy0Lsdx58eU2Mz6jk1wntCu5FEvGI/mOSy6yseQCqFuLnNTQ83JNhwvHqSnfZiw22UPQKUNhH67Z0Kij86GiejYlgmCsupy+Wz3owABOinK2jipZulFBFNBl6A5mKFb+cPGgfU/0y6t55OUKfuHmzcjLFTx68KKlf8cLgHqGQq1rB3Yz9PrWxXovdE6zvaKlcgXpQknzmzFj+1gEsaC3bR19VWdW5gSRNgM63yfqdoaeL1VMM/RowOvokguzDJ0bZeldUJ1mqcUMSKfctn0UL51b1v5ucrmCS8m8YUCv90W3cuXuFAMb0Leryy6M9HPA+alN7jxoRXJJ2XxOLrf8/N6NuG5jHF9//ryl4mgyW2zM0C287lbWuZyJuCq51GXoWkCvOykEfZLppCjPFFtl6JJEmumTXeRyBalCyeEM3dvW6H/9PlFONOBBVi6b2hLbJS8b96EDzi65qFSYYZcLoDhzEjVKc06y0KLDrFNu3T6CQqmirUK8nMyjXGGay6IebgTIrbsXMoWuDBUBAxzQuY5umqE7XBRNZouosNZLYNsZaDqp6sU7xqN48JbNeP3yakuf5mJJ8aau14uttE4q24paZ+hhvxexgLchQ9ckF399QDeXXHgtw+iSvZ69Mwm8ccW+8yI/iTmpoYf9HhTLFduDM9XVcLUfwXDAC8bat+TVwxhrLrk4uOQiUyyhwoxPyF6PhLFoAHMdDIS1YsnlTpJbt3EdXZFdNNtcwwxdiTkXdRl6N4aKgIEO6MplXusM3fob+uiVFL720zOG91U3kjQ/E/OipJ32w5PzacSDXoxF/fj5vRsR8nnwyAvnmv6b+hF3/fPr7zcilW9sPzNjXG1d1JM2k1zUvaJGrGrDTK2f946rxlCuMPxU7Vu3SvV34myXC2DfQlfL0Bu6XJzbWiSXGSqsUafnOLnkon7sv57JeKCjCd9WaJ0kLmXoibAfV0/G8NxpJaCfX1Z70EcbA3rY78Vw2IdLyRyKJaULTGjoHXL1ZAxE1Uy9Hrs94Ss5Gb/81Rfw3x99zXD12oLF4YF40Ae5zCwteuCcms9gx0RUa3u8f/cUHj14qWl2xX1chuqCl5Urk1b7RPVMxBo/qPULojkhn0dbiVaPthzBQvb85i3DiPg9ePKYvWnj+rF/J+CGWnZlF/57qHdCjGp7SjvP0HMmVwHacwW9jvWhG43965k0GUJzisWM8z4u9dy2fRQvnl1GsVTBuaUsfB4yVQCmhkK4vJLXFsB0owcdGOCAvnk0jB987C78nMmiAh4crWjojDH8t394RbuEeuVio9xhtSjDi352CqMn59PYPlbdwPLgLZuRKZbx3UOXTP9NvXUup1WGLpcryMuVmh7yZkzEgg0WuulCGX6P1DBgFfRJpjtFVy3aDQCA3yvhjqvG8NTReVtXOit17pNOwDN0u/4rWlG0PkNv4rF+YTmLf3z5oqVN9IDSsgiYZ+ixgBfpYskRvb4qmRm/bybiQVObCCdww8elntu2jyIvV3D4QhLnlrLYlAiZzkwoiy5y1USvC2P/wAAHdADYNRlrOqQSt9hC+I3nz+OfX7mM//LWq0AEHD7fGNA1v4YWbyq7Bl2pvIzZ1QJ2TFSvNG7anMCuySgeed5cdmkpuZhsY7c69s+ZVP1c9IE1Uyg1FEQBVXIxyTxbBYR67r56HBeTOa2+YIUVlzR0oI0M3aQoyq9qjK6+Pv/4cXzs7w/itt//Ee77wtP4/e+9jp+dXNRaIOsx21bEiQYVvb5d+189ViSXhXTR9Fg7ZTFTcL2T5JZtIwCA504tNrgs1rMpEcTFZE6LC0JD7wJWLHSPzabwu999DXfuHMPH37ELO8ajhpvAF9MFEAHDLfRZu9013Nd9x3g1QyciPHjzZhy6sIIjl1YN/11S5/utJ9aiXdPKtiI9E7Eg8nKl5uelC6WGHnRAkVxaZugWg+3dqt/Lk0etyy58W5EbkotdzdusKMoDutGw0tHZVeyZSeB33vUmJMI+/NUzp/GBLz+Ht/zBjw2Lzfx3bd62qPwenNDR673Q6+HSxLyBXOkEi2nzTWFOMRLx400bYth/egnnTXrQOVOJEFL5Es6qrZpCQ+8CrTpO8nIZ/+XrLyEW9OKP3r8HkkTYvWkIhy82Lo5dyBQxEvY3vSLgzwlYl1z0HS563nvTJvi9kmlxVNOL6zJ0n0dCxO8xfX67GTpvXZzXXU6nC6WGDhdAWXKRK5pr6ESKZ7sVpofDuGoiiqds6Ogr6lWAowFdPXHZzXLNiqJcQ0/XOS5WKgwn5tLYt2UYv3b3Djzy0O14+TPvwK/fswNzqYKhDKMtiDbRlZ1cctHqhMyH0NwqjLrl41LPrdtGsP/0EpazctOAznvRuTwrNPQu0EpD/73HjuDYbBp/9P69mIgpb8jd00OYTxW05dOcpXTRUoXdruRyaj4Dj0QNb55E2I93X78B33npomF2tpItQiIYauHNxv/578NOhg7UDhfVe6FzlKKoWYZeQizghdTihKjnnl3j2H96ybLckcwVEQ14HR3Bbl9yMQ62Zhn/+eUs8nKlZpt9LOjTZAA+KamHSy5mG5j4e8OJwuhKTobU5ITMT/xutC666eNSz22qrwuApgF9k9q6ePjCCvweyXJNqlPWd0BvYlT1/Vev4O/2n8Ov3rVdu7wHgN3q6PmhOh3d6s5A3pZnpmHXc3I+jS0jYUMHx/fs3ohUoWRYpE3mlDFsowAZb+InY3W5BcdouMhcQ29SFM0Zj4034+6rx1EsVQytTY1YcXjsHwDCvvYkF97lYrVt8discqW2czJWczsfWJlP1e6wBHQZupnk4qCF7kqT9xugz9Cdl1zc9HGph59AAeMedA4fLjo+l8Zo1Dm75las74DeJEN/+Lmz2DYWwSfeeXXN7dyTu15Ht+rXUG0btPYhOjmfxvY6uYWzVzu5NGr6yay5CVUzPxeryy04RuP/ioZuXBSVywwlgyGc1bxs+Tk5N28dQcjnwZNH5yw9fiXX6G3TKVxysTsIVNXQ671c+KLo2p93bFZZELFzova9wAO6cYZu/BwcbWuRhffiXzx5Eh/5W3OLptVc87/fSNgPn4dckVxaLZZxktFoQLtKahbQJ2IBeCRCucK6JrcA6z2gh3woGPiaMMZw5PIqbtk6YtB658GuyVhDVryQLmDMQobg80gI+z2WiqLlCsOZhWxNh4ue8VgAmxIhQ6OqZE5u6EHnNJNc7Gro0YAXYb+npnUxXSgZXnpre0UNOh1Wc619XOoJ+jy4fceoZR09aTKa3gkdDxbVvb+8HgkBr9TQBnl8NoWNQ8EGKYxnpQtGGbqJARhH2ytqIUN/8ewynjlu3iZq5uPCkSTCRCzYIFU6AR/q60aGDgBvvXoCmxKhpq/X65G0QnC3WhaB9R7Qg8Ya4uxqAUuZIq6Zihn9M+yZGcLhC9XCaLGkdHlYrbJb9US/sJxFsVxpKIjq2TuTMMzQV7KNXuicoZDPVDfVlltYDOhE1DBclCmUTTJ05e1mpDe3k6EDSrfLmcUsziy0Nn4y8rbpFK6B210UbRbQAeOR/GOz6Qa5BVB68odCPuMMvdhcconZ6HJZyRWRlyvavEXj/a1PlhPxgOkO2k5otSnMaT7xzqvxL791Z8vH8V3G3fJxAdZ7QDcpUL5+WWkFvHaj8eac3dMJrORknF1Uxn/tToMNWfRzqXa4GGfogBLQLyznGqZXk03khWYnlFReRsjnMbQcNkM/XMQYQ6Zo3ocOGC+5aEdDB4B7rlbqG1ay9JVcqcF9slMkidpaclEoleGVyLBAGwl4a04Q5QrDyfl0TUFUz1jUbzi9XG1bNM7QI1pHTeuAvqzWfC7qFjfoWc23vsJSpkWdz9D5SaZbgdPvlSwNp/FOl271oAPrPaCb9IQfUQP6m0wydL4i7bAqu/APk9U/XDzkteQhc3JOyTr1U6L17DHR0Ztp6EMhH9KFkqGWbdU6V8+Ezs8lWyyDsWr7nR4eWAoGhVErXuhGbBmNYOtouGVAZ4yp6+eczdABRXaxO6qvrJ8z/vhFAt6atsVzS1kUShXDDB1QdF3uNqgn32JS1Kvum7Ui//Ee/ksmAd1Khu6Wnwv3cRmOOP+37QQe0IWG3iV4RlFfoDxyeRUzIyHTAHP1hhgCXgmH1SBanRK1liFY2RoEKBn6aMSP4SaXktdvisNTtzi5UmFYzTfT0M3b1ZTlFjYDui7zSpv4uAC6vaJ1vehWvdDNuOfqCfz05EJTj/ecXIZcZo5LLoDSapizOfqvrJ8zyZz9npouF14Q3WUS0E0zdM0vxvxjPhz2aTMLZjDGtEG1i0njgLxi4QprIh7Ear7Ult1wM7rh49IO3HVxTWroROQhopeJ6DGD+z5OREeI6DAR/YiItjh7mO5glqG/fmkV107FTf+dzyPh2o1xzcJ2yWZRJh60Lrk0088BJZjsmozVBPRUvgTGzEfcmw03pSx6oeuZiAeQLZaRLpRMnRYBfVG09gNt1QvdjLt3jSMvV/DCmSXTx5h52zhBW5JLiwxdPyl63KTDhTMWDWhJhZ68XIbPYyzrcBJhP5ZNdHFOuqC0BQLAxeXGDD0vl1EsVVr+/TaYbLjqlG71oNtl49DaztA/CuB1k/teBrCPMbYbwLcB/GGnB9YNjDT0bLGE04sZXNMkoAPAnukEXr2kLCvWJBerRVGLGbrismiun3N4YZSbLCVVp0WzbLSZQdeZxYypg5wZk7qhEc1p0aDLxawoyuUnu1cGnNu2j8LvlZraAPCA7obk0s4aukKpYtmn/NhsGpsSIcOrHkDJAFdycoNPSl6umE6JckYifq0GZEZSNzNhJLm0Gvvn8F50q+ZiVlnMFLpWELXDW3aO4WP37sRt20e79pyWAjoRTQN4D4CvGN3PGHuCMZZVv30OwLQzh+cuRp7ob1xJgTE0zdABRUfPFss4OZ/GYqYIr0SWJYO4qmE3c7lbzhSxmCk21c85e2eGsJovaSu+tGzUZkC/vJLD+aUcbtYNT1iBT4vOrhaqGbqNoqhdH5d6Qn4Pbt020lRH18yjXJBc2lkUXSiVTTP0sN9T04d+bDZlWhAFgLGYEszqO1CayTqc4YhfK3iawd9PEhkXRVsZc3EmtSE0ZztduuHj0g5Bnwcfu3eX6YnbDaxm6F8A8EkAVqzSPgzge0Z3ENFDRHSAiA7Mz9vzsnYDI090bnZ17cYWGfqMUhg9dD6JxXTB1jRYXHW5a9b/e2pB7XCxlKEPA4Amu1R9v42zFrPunudPK5LFrbYDupqhp/JaC5yR5BI06UO3Y51rxj1XT+DEXLpJ0c55Yy5Ouxl6M8mF96GXyhWcms+Y6udAVaOt19GVbUXNP+LDYZ9pKyKHZ/A7xqPGGbrFE/IEl1wcLowuZbq3EWit0zKgE9H9AOYYYy9aeOwHAewD8Dmj+xljX2KM7WOM7RsfHzd6SFcx8kR//fIq4kEvNqkVajO2j0UR8Xtw+MKK4sVsI0PQAmoT2YV3uLTS0AHgqgnlWHinC+9IsJuhv3BmCdGAt6XcVA//oM6nClogataHnjeRXNotigLAdeoJmLtT1lO9anH+g9+Ohp6Xy6ZFvKi6KJoxhrNLyiyCWYcLAIyrGbpxQG+RoYf9WM3Lhh1PHJ4gXLcxjsVMsUEys5qhx4NeBH2So50u3MdlLUouvcBKhn4HgAeI6AyARwC8jYgern8QEd0L4P8G8ABjzL3VJA5T74l+5PIqrpmKt8y2JYlwveq8uGBzZ6CVNXAnF9LweyRMGyyhrccjEW6YHtIy9Fa+32bP//zpJbx5y3BLx8h64kEvAl7lg8rb7czsc4HGoqgTGfr0sHICvrCcNbzfDS90TtjvbS9DN8mewwEPKkzRwI9rHS7mJ/Zqhl4nuciVlhn6SMQPxpq/F3mCcJ06l3FppTZL17YVtaiBEBEm485uLuqmj0s/0DKgM8Y+xRibZoxtBfAggB8zxj6ofwwR3QjgL6EEc2vGGmsEvSd6ucJw9EqqpdzC2TOTwOuXVnFlJW/LulPT7pt0upycy2DrWNhycN0zk8CRy6solMpaNmp2CRz0eeD3SjW1g+VMEcdm0zXmQ1YhImUKMFXQiqJ8CrH+eQGjomhnGjqgdFB4JMIFgy4MQMkyveoQkNMoGbp9c65mGTqgbEHiplxXmXS4AMCYKnktGmXoLYqi/CqumY7O30/8c1Hf6bJqw5Z4Mu7s+P9CF31c+oG2+9CJ6LNE9ID67ecARAF8i4gOEtGjjhxdF9B7op9dzCBbLFuWHHZPD6FYruDKat6Wgf2QBcnllIWWRT03ziQglxmOXFpFMisjGvA2nfasnxblLX/tBHRAmQKcW1UCukTG4+bVomi9hl4CmVj9WsXrkTA1FGyaoSfCPldc7yJ+Zfm1nXV4hVLZNEPXr6E7NpvCzEhIs9U1e/6AV2pLcuGZbbNOl+Ws0ufN7WLrdfRW+0T1TMaDjmrodluGBx1bnyDG2JMAnlS//ozu9nsdPaouEg/6tDfo65eVy9tWHS6cPdMJ7Ws7byhtoMlkWrRYquDsUhbvvmHK8s/UT4wmLUxEDoVqtzU9f3oJfq+E3dPGdgetmIgH8MaVFFJ5xWnRKHB6JILf02ihu5pTTkB2vNCNmB4OmWbobljnckJ+pcidlyum3uP1NC+KVkfyj8+msWvCXD8HlCsko170vFzBSKRVUdS4Q0ZPMisjEfFhw1DQsNNlNScj7LdmFzEZC+CHqwUwxhw5uXbbx2Wts64nRYFaT/Qjl1fglQg7m+iVeqaHQxhWL1ntaOitthadW8qgXGGWOlw4G+JBTMQCOHg+iZVsa5vYej+ZF84sYe9Mou1pu4lYEPNqhm7U4cIJ+KRGyaVNY656pofDTSQXd8b+gfYWReebSC68oLySk3FqwdiUq56xqB8LBm2LLYuiaiBMNsnQk9kiEiE/fB4Jk/FgQ0C3MvbPmYwHkZPLlhwerdBN69x+QAR0XZfL65dTuGoiajmoERFuULN0O+O9Ub8XEplr6CcN9ohaOZa9MwkloFvw/dYPN6ULJbx6adV2u6KeiXgAqUIJC+lC04Ae8nkavFwU61wnAnoIs6m8oVeM8jtxJ4szW0rRjGZ96PznHbm0CrnMmhZEOWPRABbq+rsLsvnwEmdEy9DN5b9lXYKwKRFq0NDtBHSnNxctqRn6WvNx6RUioOs80Y+0GPk3Yo8qUdgZ75UkQqzJcg3usrhtzHqGDiiyy5nFLM4sZhuWQ9ej95N56ewyyhWGm7d2ENDV4aJTCxnTiUZA0dGNM/TOV3RND4fBGHDZwG8k6aLkYmbD3IxWk6IANJ/7Zj3onNGov8FC10ofekjV35tp6PqT4cZEqKHLxc4V1gaHNxetVR+XXiECuvphPLeUxZXVvO0e7HdcO4ldk1HTrUKmzxvymkouJ+cymIwHbHuq3Kjq6AvpQsuJyCHdGroXzizBIxFu2jJs6/n08OGi80vZphl60Cc1FkXbtM6tZ0ZrXTSeZnQtoFsocuthjKHYREPnEs7Bc0kQWbtS4xq6fvrYSpcLoI7/N9HQl7NFTVrcNBzC5WQeZd3zrNi4wnJ6/H9xjfq49AoR0NU34n51L6XVlkXO7ukE/vW/3m07WAzV9b/rOTGXsiW3cG6YHgKvM7Xqt1YMwpThlf2nl3DdxnjTQNwK/kGtMOMedE7I52koiqbatM6tZ1rtwjhf1+lSKleQypdccVoEqm2oVvx5AN1yC7Ndn+rf4WIyh80jYUuF1tFoACXVZRNQTho5C10ugFIYNcvQyxWmZOjq+2ljIoRShWl2yYByIrMrucw6ZNC1tEZ9XHqFCOjqh/G5U0rbnt0MvZPnNQoAlQrD8bm0pcvsemJBH65STwRWiqLlijJld/B8Erd0ILcA1QwdAKIGPeicgJHkkpM7mhLlTMYC8ErU0LrIT5zuZejchtliQOcLok2yZ32L4s4WHS4cXpTnrYtymaHCzLcV6RmOmI//p/Ky4typSi7T6gT1xWT1d2zn7xf2exELem1tLnrx7BJ+eGTW8D6ru3zXC+s+oHOHv/2nF7EhHuza2V4veei5mMwhWyy3FdCBavuiFQ0dAJ45voBiqWLbkKueRNgHv9q2ZrTcgqNk6FXJpVxhSBWcydC9HglTiWCD5KJNibqVoWuSizUNnRdtzSQXv1fSfpdWCqJAtcuDT4u22lakZzjsr3FU1MMHjnjRcaMW0JUMm//97JwslWlR6xn6535wFL/97UOGZnaLmaKtob5BZ90HdP5hXEgXbcstHT2viSf68bnWo97N2KsG9FYaOn/dj7+uZD6dFEQBpctmXM3SmxdFJRR0bovczMsJDR0AphONrYuat43D6+c4rbqW6im0WN4MVGUrqyf20boMnTtatnJbBBQNfclEcqn/3fGlDbzTxap1rh67m4tOzGWwnJW1TWIcxhiWhY9LDSKg6zJDs6XQrjyvSVH06BWlw8VK77ERd+8ax/axSMtuHX6J/PTReeyajDryoeD6aMsuF11Ar/q4dC65AHy4qFZySdqYZGyHVl1L9bTK0IHq79DqTATP0PmgDZd1QhYCeiLsx0pOril0cuqtmGNBH+JBrzaM144Pjx0/l5WsrJ2kfnZyseY+7uMiJJcqIqDrtL9rp9qbkmyHoZAPebnS0DN9fDaFDfFg23rvzEgYP/4/78HMSHNTL/7zU4VSx9k5h+vorfrQ9X7odsbGrTA9HMbsaqHm97rqsuQC1A6otSKvaehNArqa9Vstjg+H/ZCoMUO3oqGPhH2mBl3VZSnVE/6m4bAW0K06LeqZjAcxl8o33QfAOTGf0r5+9uRCzX3cx0VILlXWfUAP+TzwqiPn3c3Q1YBaFwSOzaUsZ2WdoP8AtuvfUg/vdGnetlhbFHXCaVEPd128pOtFd3P9HMesyG2ElqG3kFy2jEYsL0fwSISRiL+qoWv7RC1o6BHz8f9ldeBoWHcy3JSoTou2szhkMhaAXGYtNyUBwIk55Yr1rl3jeP70Us1WJuHj0si6D+hEhHjIh7Bf+QB1C6NWt3KF4USbHS62n9+FgD5hSUOvLYp2un6uHiMbXaevAoyo99VvRsHC8uYHb96Mj9y53dYxjEYC1QzdZlEUMB7/T2aLinFaUB/QQ1pA17zsbUougLXhohNzafi9En5h3wyyxTIOX6juzq0uZxcBnbPuAzqg6Ldv2hCz7QPeCUaOi+eXssjLFVzdhYDOC3kzIyFMDTVf5mEVPi3aarCoWKpol9s8Q3eqpZD3ousLo1bcJzul3hunGdU+dPNg+/6bZ/Afb91s6xjGYn7NQteW5NIkQ0+qPeb6z8bGRAipfAmrebktyYUvRLHSi35yPoPtYxHccdUoiIBnT1R19EVNchEaOkcEdAAP3bUDv3r3jq4+J9fu9Rn6Mb7dvQuS8flfYAAAFmlJREFUiyQpXSm3O7jAdudkFETVTggjgnVLLjQvdIckF6NedDeNuTjxkNextsV2UTJ0JShzWctKhl71RDeQXLKylsFzNmmyVq6tgL5hyPoquhNzaeyYiCIR9uO6jfEaHV34uDTizHVun2M3E3ICLUPXaejH5zrrcLHLwx++VWs1dIIbNw/jxU+/o6mmGdJ5oof91ddvtFS6HYx60e1MMraLWRuqEVaKou2gjP9zyYW3RlrP0I2WXCSzjSdDrRd9OYfVvAyfhyw9D2dc7UppJbnk5TLOL2fx72/cBAC4Y8cY/vrZ08gVywj5PYqPS1D4uOgRGXqP0LYW1WXomxKhjkbw7bBzMua4A2GrAhX/4PPWxdWcjFjA66jcVd+LnrRgJ9wp8ZAP2WIZcpPdnBwrRdF2GI36kSmWkSuWq33oFoJdyKcadBlJLlm5piAKVKdFeYY+FLK3OMTvlTAa8bfcXHRqPgPGqtuabt8xCrnMtGUsYqioERHQe4SRJ/rRK6m2B4r6herWIjWg550x5tJT34ue7EqGzpeWtM7StcEihzP0cW1atKANb1nxgSEiDIf9xl0u2WLDSX8sGoDfI+GCGtDb+ftNWNhcdGK+dv3eLdtG4POQJrsIH5dGLL+jiMhDRC8T0WMG9wWI6O+J6AQR7SeirU4e5CCi7fVUL9NL5QpOzWe60uHSSxoCeq7kWIcLp74X3Yo/fKfwtj0rveial4sLGTqgBHStbdHicwxH/IaSi9GyFEkiTCWCuJTMKz4ubdQ/lGnR5pLLibk0JKraSIf9Xtw4M6wNGAkfl0bspAgfBfC6yX0fBrDMGLsKwOcB/EGnB7YeUFrdlABwdimLYrnSNf28V9QH9JQLGfrMSFXjZYyp6+fczeSMJDQz3CqK6qdFtS4Xi88xEvE1FEXlcgWpQsnQMmHjUAgXl7Nt1yc2WPBzOTmXxsxIuOakdPuOUbxycQUrWVlILgZY+msT0TSA9wD4islDfh7A19Svvw3g7eTGNt4BI67b63l8tjMPl35BXxQFlIzWqQ4XzvRwtXUxL1dQLFe60OXCM/TWAT0vVyARtIE2p6jJ0EtleCWC12KrZiLc6InO5UCjLpJNwyFcSubb9pmfiAexkC6g1KTmcGIurbmHcu64agyMAT87tSB8XAywmiJ8AcAnAZj99jcBOA8AjLESgBUADf1wRPQQER0gogPz8/NtHO5goe9dPnolDaKqXjioaEXRYrUo6oR1rp5p3aKL6ui6+10ugDVPdGX9nMeRJcl6tAw9U0Tewvo5PSMGnuiaMZdB4XxjQln3t5gptvX3m4wHUGFVd8h6SuUKTi9kGj4Pe2cSCPk8+N6rV4SPiwEtAzoR3Q9gjjH2YrOHGdzWYNTAGPsSY2wfY2zf+Pi4jcMcTPTj4sfmUpgZDtd4YQ8iofo+dIcWROuZjAe1XnQ+9t+NPnTAmoWusn7O+X6EoM+DaMCL+VTB0vo5PcMRP5J1Bl3LTSwTphMhMKZYV7Tzu52M8WlRY9nl/HIOxXIFO+oCut8r4eZtI5o/upBcarHyF78DwANEdAbAIwDeRkQP1z3mAoAZACAiL4AhAEsOHudAovdEPz47+B0ugF5DV6ZF0wVnFkTr8UiEjYkQLixXB1/c9HEBdBq6BcmlIFdc650ei/qxmCkiJ5dtPcewatClrwHwk2H9YBFQ7UUH2jtZVsf/jQP6ybnaDhc9d+wYRUa9whOSSy0tAzpj7FOMsWnG2FYADwL4MWPsg3UPexTAh9Sv36c+prWV2jqHO/QVS+ujwwWorl3LyWWkCiUw5px1rh7euqhl6C5LLmG/YvJmpSiaL5VN1891ymg0gIVUAQXZ3lWANv6vk12Ws+ZylX4auN0uFwCYTRl3uvCWRSO3yTuuGtO+Fj4utbT9riKizxLRA+q3fwVglIhOAPg4gN9x4uAGHS65nFnMoFRh6yKgc8mlIJerY/8uZM9KQM9hRdXQ3ZZcuMmb9QzdnYCuZOiK5GKlB53DdXJ9YXSlzgtdT6cZ+mg0gKBPwht1Sys4J+bSGI8FDH/2NVNx7Xbh41KLrXcVY+xJxtj96tefYYw9qn6dZ4z9B8bYVYyxWxhjp9w42EGD7/U8eE5xkOuGh0uv4ZJLrlh23DpXz/RwGHOpgra70umJWCPiQS9WLGno9uQQO4xGFT+XfKlsyTqXMxJuHP9fzhbhlchwcjno82h7TNsJ6B6J8I5rN+CfX7ncsBMAMO5w0f9b7kEkfFxqEZOiPYRnpgfOLtlaZtDP+DwSvBIhXypXrVcd7nIBqp0uRy6vwisRIjay1XaJm+yJrcetoigAjEWUbpVMoWyry4UHRn2GnlQHssy6cTapWXq7V1jvvWkTklkZT7xR2/HGGMPJuXTTjq9fuXMb/vM9O4SPSx0ioPcQntkcOLOMrTaWGfQ7QZ8HebnieoYOAK9dWrXtNdIuVg26CiUXi6KxABgDLq/k7HW5hBs19KTB2L8eLru0K2fdedUYxqIB/MNLF2pun0sVkCqUmgb0fVtH8H/d96a2nneQEQG9h/BAdmohsy7kFk7QJyGn09Dd0Ld5hn5uKet6QZQzZDFDz8tl1zR0rinPrhZsWQuE/YoVhb4XfTkjN+0O6jRD93ok/Lu9G/HE0bkaH5kTTTpcBM0RAb2H6KWG9VAQ5QTVvaLc98SNDJ33ogPutyxyrO4VLZQqrnW5jOm6Puxo6ESkDBc1SC7mGfo7r9uAf3/jJsQ6cAd9703TkMsMjx2+pN0mAnr7iIDeQ/SZ6aB7uOjRArqazTrlha6H96ID7ne4cKzuFS3YLFjaQT85aVenT4R9WMro+9CLTSdsb9k2gs//wl5IHVgYXLsxjmum4vhfL13Ubjsxl0Ys4NVWGgqsIwJ6D9Fnpt1YO7dWCOk09KjDXuh6uOzSjQ4XQJEeiqWKZoxlRkF2L0Mfrwno9k4aIxF/zV5RIy90N/jfbtqEQ+eTWmbOtxQJOyj7iIDeQ7htrFcizSJ0PRD0SWqGXnJlqIjDA3r3MnR1/L9FYdTNomg85IXPowTCkM2APhzxa0XRvFxGTi535WT4wN6NkAhacfTkfPMOF4E5IqD3EK9HQjTgxdaxCPwuFcnWIkGfRymKumCdq2dG7XTpWkDXFn8319HdLIoSkVYYtSu5DId9moaebDJU5DQTsSDu2jWO77x8EStZGXOpggjobbJ+osgaZSzqxzVT8V4fRlfR2hbbXI5glekRLrl0OaA3ydAZY2qG7t5Hj4/D25Zcwn6sqAZd3KXSyMfFDd570zQur+Tx8P6zAGA6VCRozmBb+/UBf/HBN3ftQ7NW4EVRANik8wRxmuluZ+gWllwUy+5sK9LDbXTtPkci7EdFNehaznTH1IzzzmsnEQt48cWnTgJAg8uiwBoiQ+8x10zFsWHIvaC2FglpGrq7GfoNm4bwi7dtwZ07u2PVPKS2oTbrdOH7RLuSodt8Dm7QtZwtah443SooB30evPuGKaTyJfg9EmaGQ63/kaABEdAFXafah+6uhh70efB7/+56jHep/a1qoWuuobu1T1QP73SxK7kM6wL6chc1dM57b9oEQNkhanXTkqAWIbkIuk7Q50GmWIZcrrja5dJtqkVR8wydS01rUUPnLYpLGVmbGO2mHHjz1hFsH49g9/RQ155z0BicT5Ogbwj6PCiq0oObGXq3CfqU8flmRdFuSC5j0Xa7XHSSS1ZGwCvZsuDtFEkifOfX74BfZOdtIwK6oOvoA42bGnoviAd9TdsWuVWsm0Zsm0eUYvCYzX2bmoaeKWK5xZSoW3SrgD2oiIAu6Dr6gRc3rHN7STzkbSq5dCND37d1BD/8+F24asLe9HHY74HfI2EpW1SnRNdX99UgIK5tBF1Hn50OZIbeRHKpaujuShl2gzmgDCUNR3xIZmQks7LIlvuQlgGdiIJE9DwRHSKi14jodw0es5mIniCil4noMBG9253DFQwCNZLLgAWNVksutAzdJS+XThkOK+P/y9miyND7ECvvqgKAtzHG9gDYC+A+Irqt7jGfBvBNxtiNUBZJ/7mzhykYJEIDnKEPhXzW2hbXqNXDsGqhy7cVCfqLlu8qppBWv/Wp/7H6hwHg8+tDAC5BIDAhMMgaerCVhu5+UbQTRlSDrlbbigRrE0tpAhF5iOgggDkAjzPG9tc95H8A+CARXQDwLwB+0+TnPEREB4jowPz8vNFDBOsAfYZutIC4n4mHFE90xupzHoVuFEU7YTjiw6VkDnKZdcU6V+Aslt5VjLEyY2wvgGkAtxDR9XUP+QCArzLGpgG8G8D/JKKGn80Y+xJjbB9jbN/4eHfGsQVrD56dRgPegZsIjAd9KFUYciae6IUuFUXbZTjsR16VhYTk0n/Y+jQxxpIAngRwX91dHwbwTfUxPwMQBDDmwPEJBhBeFB2kKVEOl5DMetH7oSjKEZJL/2Gly2WciBLq1yEA9wJ4o+5h5wC8XX3MNVACutBUBIZwyWXQOlyA6mCMWesiD+huraDrlOFI9W/SLadFgXNYSZGmAHyNiDxQTgDfZIw9RkSfBXCAMfYogE8A+DIR/VcoBdL/xMxERMG6h0susUHM0FtY6BbkMoigbRVaa+gzdG7WJegfWn6iGGOHAdxocPtndF8fAXCHs4cmGFR4QB+0lkWgetVhZqHLl1us1X2ZI7ogLjL0/mNtCnmCgUbT0AcwYLTaK6qsn1ubcgsgNPR+RwR0QdfxeyQQDWpRtPleUbfXz3UKl1kifs+62nM7KIi/mKDrEBEevHkG97xpoteH4jgtNfRSZc0OFQFqIPdIIjvvUwYvRRL0Bb//3t29PgRX8HslhHyeJl0u5TWdoRMREmGf6EHvU9buO0sg6FPiIa95UVSurNkedM5IxC+MufoUkaELBA7TbMlFvrS2i6IA8On3XIuQf22fdATGiIAuEDhMPGTuiV6Q13ZRFADeslMMefcra/udJRD0IUPNAvoaL4oK+hsR0AUCh1EsdM3aFtd2UVTQ34h3lkDgMM0kl3wfSC6C/kW8swQCh1GKojIqlUY7o0IfFEUF/YsI6AKBw8RDXlQYkCk2yi6F0tpvWxT0L+KdJRA4jDYtarBbtCCLoqjAPURAFwgcRvNErxsuYoyJoqjAVcQ7SyBwmLhJQJfLDBW2dveJCvof8c4SCByGSy714/+54treJyrof0RAFwgcRtsrWqehP3tyAQBw7cZ4149JsD6wslM0SETPE9EhInqNiH7X5HHvJ6Ij6mO+7vyhCgT9gZmF7ncPXcJYNIDbto/24rAE6wArXi4FAG9jjKWJyAfgJ0T0PcbYc/wBRLQTwKcA3MEYWyaiwTO6FggsEjPYWpQulPDjN+bw4M0z8Ehrc/2coP+xslOUAUir3/rU/+onJj4C4M8YY8vqv5lz8iAFgn7C65EQDdSO///wyCwKpQr+7Z6NPTwywaBjSUMnIg8RHQQwB+Bxxtj+uofsArCLiJ4loueI6D6Tn/MQER0gogPz8/OdHblAsIaJB2s90b976BKmhoK4afNwD49KMOhYCuiMsTJjbC+AaQC3ENH1dQ/xAtgJ4B4AHwDwFSJKGPycLzHG9jHG9o2Pj3d25ALBGkbv57KSlfH08Xncv3sKkpBbBC5iq8uFMZYE8CSA+gz8AoB/YozJjLHTAI5CCfACwbqE+7kAwA9euwK5zITcInAdK10u4zzbJqIQgHsBvFH3sH8E8Fb1MWNQJJhTzh6qQNA/KBm6oqF/9/AlbBkN44ZNQz0+KsGgYyVDnwLwBBEdBvACFA39MSL6LBE9oD7mBwAWiegIgCcA/DZjbNGdQxYI1j7xkBerORkL6QJ+enIR9++eApGQWwTuYqXL5TCAGw1u/4zuawbg4+p/AsG6h0su33v1CsoVIbcIuoOYFBUIXCAe8iFVKOHRgxexcyKKqydjvT4kwTpABHSBwAXi6nDRC2eWcf/ujUJuEXQFEdAFAhfgjosAcP+eqR4eiWA9IQK6QOAC3BP9uo1x7BiP9vhoBOsFEdAFAhfgBl337xbFUEH3EAFdIHCBGzcn8OG3bMODN8/0+lAE6wgrbosCgcAmQZ8H/8/91/b6MATrDJGhCwQCwYAgArpAIBAMCCKgCwQCwYAgArpAIBAMCCKgCwQCwYAgArpAIBAMCCKgCwQCwYAgArpAIBAMCKRYmffgiYnmAZxt85+PAVhw8HB6zSC9nkF6LYB4PWuZQXotgPXXs4UxZriUuWcBvROI6ABjbF+vj8MpBun1DNJrAcTrWcsM0msBnHk9QnIRCASCAUEEdIFAIBgQ+jWgf6nXB+Awg/R6Bum1AOL1rGUG6bUADryevtTQBQKBQNBIv2boAoFAIKhDBHSBQCAYEPouoBPRfUR0lIhOENHv9Pp47EJEf01Ec0T0qu62ESJ6nIiOq/8f7uUxWoWIZojoCSJ6nYheI6KPqrf36+sJEtHzRHRIfT2/q96+jYj2q6/n74nI3+tjtQoReYjoZSJ6TP2+n1/LGSJ6hYgOEtEB9bZ+fa8liOjbRPSG+vm53YnX0lcBnYg8AP4MwLsAXAvgA0TUb2thvgrgvrrbfgfAjxhjOwH8SP2+HygB+ARj7BoAtwH4DfXv0a+vpwDgbYyxPQD2AriPiG4D8AcAPq++nmUAH+7hMdrlowBe133fz68FAN7KGNur69fu1/fanwD4PmPsTQD2QPkbdf5aGGN98x+A2wH8QPf9pwB8qtfH1cbr2ArgVd33RwFMqV9PATja62Ns83X9E4B3DMLrARAG8BKAW6FM73nV22veg2v5PwDTamB4G4DHAFC/vhb1eM8AGKu7re/eawDiAE5DbUpx8rX0VYYOYBOA87rvL6i39TuTjLHLAKD+f6LHx2MbItoK4EYA+9HHr0eVKA4CmAPwOICTAJKMsZL6kH56z30BwCcBVNTvR9G/rwUAGIB/JaIXiegh9bZ+fK9tBzAP4G9UOewrRBSBA6+l3wI6Gdwm+i57DBFFAfwvAB9jjK32+ng6gTFWZozthZLd3gLgGqOHdfeo7ENE9wOYY4y9qL/Z4KFr/rXouIMxdhMUyfU3iOiuXh9Qm3gB3ATgLxhjNwLIwCGpqN8C+gUAM7rvpwFc6tGxOMksEU0BgPr/uR4fj2WIyAclmP8dY+wf1Jv79vVwGGNJAE9CqQ0kiMir3tUv77k7ADxARGcAPAJFdvkC+vO1AAAYY5fU/88B+A6UE24/vtcuALjAGNuvfv9tKAG+49fSbwH9BQA71Uq9H8CDAB7t8TE5waPA/9++/apEFERxHP+eJCKCCjaDWGziAxgExbDZZtjgU4jgI/gGZoPFYFW7CP5hVVCbQd/B8DPMWbAIuixcZ/h9YNjh3g3nB3PPcmdY+jnvU/ai/72ICOAIeJJ0+O1WrXnmI2Im55PAJuWw6hLYzq9VkUfSnqQFSYuU5+RC0g4VZgGIiKmImB7OgS1gQIVrTdIH8BYRy3lpA3hkHFm6PiAY4UChBzxT9jb3u65nhPqPgXfgk/JLvUvZ2zwHXvJzrus6f5lljfLKfg/c5uhVnGcFuMk8A+Agry8BV8ArcAJMdF3rH3OtA2c1Z8m673I8DJ/9itfaKnCda+0UmB1HFv/138ysEbVtuZiZ2Q/c0M3MGuGGbmbWCDd0M7NGuKGbmTXCDd3MrBFu6GZmjfgCi+9CFXIBFMkAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(des_loss_list)\n",
    "plt.show()\n",
    "plt.plot(com_loss_list)\n",
    "plt.show()\n",
    "plt.plot(total_loss_list)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 利用feature extraction 當input加入SVDD"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 參數設定"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [],
   "source": [
    "img_size = 224\n",
    "flatten_size = 7* 7 * 512\n",
    "n_samples = len(Tdata_train)\n",
    "z_dim = 28\n",
    "h_dim = [8, 4]\n",
    "regular = 1e-5\n",
    "max_lr = 1e-6\n",
    "base_lr = 1e-4\n",
    "\n",
    "n_epochs = 50\n",
    "n_ae_epochs = 100\n",
    "objective = 'soft-boundary'     #'one-class'  'soft-boundary'\n",
    "nu = 0.1\n",
    "warm_up_n_epochs = 10\n",
    "optimizer = tf.keras.optimizers.Adam(lr = base_lr)\n",
    "optimizer = tf.keras.optimizers.Adam(lr = base_lr)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 整理generate data 當作encoder input"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model_14\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_1 (InputLayer)         [(None, 224, 224, 3)]     0         \n",
      "_________________________________________________________________\n",
      "block1_conv1 (Conv2D)        (None, 224, 224, 64)      1792      \n",
      "_________________________________________________________________\n",
      "block1_conv2 (Conv2D)        (None, 224, 224, 64)      36928     \n",
      "_________________________________________________________________\n",
      "block1_pool (MaxPooling2D)   (None, 112, 112, 64)      0         \n",
      "_________________________________________________________________\n",
      "block2_conv1 (Conv2D)        (None, 112, 112, 128)     73856     \n",
      "_________________________________________________________________\n",
      "block2_conv2 (Conv2D)        (None, 112, 112, 128)     147584    \n",
      "_________________________________________________________________\n",
      "block2_pool (MaxPooling2D)   (None, 56, 56, 128)       0         \n",
      "_________________________________________________________________\n",
      "block3_conv1 (Conv2D)        (None, 56, 56, 256)       295168    \n",
      "_________________________________________________________________\n",
      "block3_conv2 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_conv3 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_pool (MaxPooling2D)   (None, 28, 28, 256)       0         \n",
      "_________________________________________________________________\n",
      "block4_conv1 (Conv2D)        (None, 28, 28, 512)       1180160   \n",
      "_________________________________________________________________\n",
      "block4_conv2 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_conv3 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_pool (MaxPooling2D)   (None, 14, 14, 512)       0         \n",
      "_________________________________________________________________\n",
      "block5_conv1 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv2 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv3 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_pool (MaxPooling2D)   (None, 7, 7, 512)         0         \n",
      "_________________________________________________________________\n",
      "tf_op_layer_Reshape_20 (Tens [(None, 25088)]           0         \n",
      "=================================================================\n",
      "Total params: 14,714,688\n",
      "Trainable params: 0\n",
      "Non-trainable params: 14,714,688\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "def make_generate_model(model):\n",
    "    \n",
    "    model.trainable = False\n",
    "    layer_name = 'block5_pool'\n",
    "    x = model.get_layer(layer_name).output\n",
    "    output = tf.reshape(x, shape = (-1, flatten_size))\n",
    "    g_model = tf.keras.Model(inputs=model.input, outputs = output)\n",
    "    \n",
    "    return g_model\n",
    "generate = make_generate_model(target_model)\n",
    "generate.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 建構模型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(None, 14, 14, 8)\n",
      "(None, 7, 7, 4)\n",
      "(None, 28)\n",
      "Model: \"model_15\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "inputs (InputLayer)          [(None, 25088)]           0         \n",
      "_________________________________________________________________\n",
      "tf_op_layer_Reshape_21 (Tens [(None, 28, 28, 32)]      0         \n",
      "_________________________________________________________________\n",
      "conv2d_8 (Conv2D)            (None, 28, 28, 8)         6408      \n",
      "_________________________________________________________________\n",
      "batch_normalization_16 (Batc (None, 28, 28, 8)         32        \n",
      "_________________________________________________________________\n",
      "leaky_re_lu_20 (LeakyReLU)   (None, 28, 28, 8)         0         \n",
      "_________________________________________________________________\n",
      "max_pooling2d_8 (MaxPooling2 (None, 14, 14, 8)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_9 (Conv2D)            (None, 14, 14, 4)         804       \n",
      "_________________________________________________________________\n",
      "batch_normalization_17 (Batc (None, 14, 14, 4)         16        \n",
      "_________________________________________________________________\n",
      "leaky_re_lu_21 (LeakyReLU)   (None, 14, 14, 4)         0         \n",
      "_________________________________________________________________\n",
      "max_pooling2d_9 (MaxPooling2 (None, 7, 7, 4)           0         \n",
      "_________________________________________________________________\n",
      "tf_op_layer_Reshape_22 (Tens [(None, 196)]             0         \n",
      "_________________________________________________________________\n",
      "dense_12 (Dense)             (None, 28)                5516      \n",
      "=================================================================\n",
      "Total params: 12,776\n",
      "Trainable params: 12,752\n",
      "Non-trainable params: 24\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "\n",
    "\n",
    "def make_encoder():\n",
    "\n",
    "    inputs = tf.keras.Input(shape=(flatten_size,), name='inputs')\n",
    "    x = tf.reshape(inputs, shape = (-1, 28, 28, 32))\n",
    "    x = tf.keras.layers.Conv2D(h_dim[0], (5, 5), activation='linear', padding='same', kernel_initializer='random_uniform', kernel_regularizer=tf.keras.regularizers.l2(regular))(x)\n",
    "    x = tf.keras.layers.BatchNormalization()(x)\n",
    "    x = tf.keras.layers.LeakyReLU()(x)\n",
    "    x = tf.keras.layers.MaxPool2D((2, 2), padding='same')(x)\n",
    "    print(x.shape)\n",
    "    x = tf.keras.layers.Conv2D(h_dim[1], (5, 5), activation='linear', padding='same', kernel_initializer='random_uniform', kernel_regularizer=tf.keras.regularizers.l2(regular))(x)\n",
    "    x = tf.keras.layers.BatchNormalization()(x)\n",
    "    x = tf.keras.layers.LeakyReLU()(x)\n",
    "    x = tf.keras.layers.MaxPool2D((2, 2), padding='same')(x)\n",
    "    print(x.shape)\n",
    "    _, h, _, _ = x.shape\n",
    "    x = tf.reshape(x, shape = (-1, h * h * h_dim[1]))\n",
    "    x = tf.keras.layers.Dense(z_dim)(x)\n",
    "    print(x.shape)\n",
    "    model = tf.keras.Model(inputs=inputs, outputs=x)\n",
    "\n",
    "    return model    \n",
    "   \n",
    "def make_decoder():\n",
    "\n",
    "    inputs = tf.keras.Input(shape=(z_dim,), name='inputs')\n",
    "    x = tf.keras.layers.Dense(h_dim[1] * 7 * 7)(inputs)\n",
    "    x = tf.reshape(x, shape = (-1, 7, 7, h_dim[1]))\n",
    "    x = tf.keras.layers.LeakyReLU()(x)\n",
    "    #x = tf.keras.layers.UpSampling2D((2, 2))(x)    \n",
    "    \n",
    "    x = tf.keras.layers.Conv2DTranspose(h_dim[1], (5, 5), padding='same', kernel_initializer='random_uniform', kernel_regularizer=tf.keras.regularizers.l2(regular))(x)\n",
    "    x = tf.keras.layers.BatchNormalization()(x)\n",
    "    x = tf.keras.layers.LeakyReLU()(x)\n",
    "    x = tf.keras.layers.UpSampling2D((2, 2))(x)   \n",
    "    print(x.shape)\n",
    "        \n",
    "    x = tf.keras.layers.Conv2DTranspose(h_dim[0], (5, 5), padding='same', kernel_initializer='random_uniform', kernel_regularizer=tf.keras.regularizers.l2(regular))(x)\n",
    "    x = tf.keras.layers.BatchNormalization()(x)\n",
    "    x = tf.keras.layers.LeakyReLU()(x)\n",
    "    x = tf.keras.layers.UpSampling2D((2, 2))(x) \n",
    "    print(x.shape)\n",
    "    \n",
    "    x = tf.keras.layers.Conv2DTranspose(1, (5, 5), activation='sigmoid', padding='same', kernel_initializer='random_uniform', kernel_regularizer=tf.keras.regularizers.l2(regular))(x)\n",
    "    print(x.shape)\n",
    "    \n",
    "    x = tf.reshape(x, shape = (-1, flatten_size))\n",
    "    model = tf.keras.Model(inputs=inputs, outputs=x)\n",
    "\n",
    "    return model    \n",
    "\n",
    "encoder = make_encoder()\n",
    "encoder.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(None, 14, 14, 4)\n",
      "(None, 28, 28, 8)\n",
      "(None, 28, 28, 32)\n",
      "Model: \"model_16\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "inputs (InputLayer)          [(None, 28)]              0         \n",
      "_________________________________________________________________\n",
      "dense_13 (Dense)             (None, 196)               5684      \n",
      "_________________________________________________________________\n",
      "tf_op_layer_Reshape_23 (Tens [(None, 7, 7, 4)]         0         \n",
      "_________________________________________________________________\n",
      "leaky_re_lu_22 (LeakyReLU)   (None, 7, 7, 4)           0         \n",
      "_________________________________________________________________\n",
      "conv2d_transpose_12 (Conv2DT (None, 7, 7, 4)           404       \n",
      "_________________________________________________________________\n",
      "batch_normalization_18 (Batc (None, 7, 7, 4)           16        \n",
      "_________________________________________________________________\n",
      "leaky_re_lu_23 (LeakyReLU)   (None, 7, 7, 4)           0         \n",
      "_________________________________________________________________\n",
      "up_sampling2d_8 (UpSampling2 (None, 14, 14, 4)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_transpose_13 (Conv2DT (None, 14, 14, 8)         808       \n",
      "_________________________________________________________________\n",
      "batch_normalization_19 (Batc (None, 14, 14, 8)         32        \n",
      "_________________________________________________________________\n",
      "leaky_re_lu_24 (LeakyReLU)   (None, 14, 14, 8)         0         \n",
      "_________________________________________________________________\n",
      "up_sampling2d_9 (UpSampling2 (None, 28, 28, 8)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_transpose_14 (Conv2DT (None, 28, 28, 32)        6432      \n",
      "_________________________________________________________________\n",
      "tf_op_layer_Reshape_24 (Tens [(None, 25088)]           0         \n",
      "=================================================================\n",
      "Total params: 13,376\n",
      "Trainable params: 13,352\n",
      "Non-trainable params: 24\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "   def make_decoder():\n",
    "\n",
    "    inputs = tf.keras.Input(shape=(z_dim,), name='inputs')\n",
    "    x = tf.keras.layers.Dense(h_dim[1] * 7 * 7)(inputs)\n",
    "    x = tf.reshape(x, shape = (-1, 7, 7, h_dim[1]))\n",
    "    x = tf.keras.layers.LeakyReLU()(x)\n",
    "    #x = tf.keras.layers.UpSampling2D((2, 2))(x)    \n",
    "    \n",
    "    x = tf.keras.layers.Conv2DTranspose(h_dim[1], (5, 5), padding='same', kernel_initializer='random_uniform', kernel_regularizer=tf.keras.regularizers.l2(regular))(x)\n",
    "    x = tf.keras.layers.BatchNormalization()(x)\n",
    "    x = tf.keras.layers.LeakyReLU()(x)\n",
    "    x = tf.keras.layers.UpSampling2D((2, 2))(x)   \n",
    "    print(x.shape)\n",
    "        \n",
    "    x = tf.keras.layers.Conv2DTranspose(h_dim[0], (5, 5), padding='same', kernel_initializer='random_uniform', kernel_regularizer=tf.keras.regularizers.l2(regular))(x)\n",
    "    x = tf.keras.layers.BatchNormalization()(x)\n",
    "    x = tf.keras.layers.LeakyReLU()(x)\n",
    "    x = tf.keras.layers.UpSampling2D((2, 2))(x) \n",
    "    print(x.shape)\n",
    "    \n",
    "    x = tf.keras.layers.Conv2DTranspose(32, (5, 5), activation='sigmoid', padding='same', kernel_initializer='random_uniform', kernel_regularizer=tf.keras.regularizers.l2(regular))(x)\n",
    "    print(x.shape)\n",
    "    \n",
    "    x = tf.reshape(x, shape = (-1, flatten_size))\n",
    "    model = tf.keras.Model(inputs=inputs, outputs=x)\n",
    "\n",
    "    return model    \n",
    "decoder = make_decoder()\n",
    "decoder.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "def get_radius(dist, nu):\n",
    "    \"\"\"Optimally solve for radius R via the (1-nu)-quantile of distances.\"\"\"\n",
    "    return tfp.stats.percentile(tf.sqrt(dist), 1 - nu)\n",
    "\n",
    "def init_center_c(eps = 0.1):\n",
    "    \n",
    "    positive = tf.fill([z_dim], eps)\n",
    "    negative = tf.fill([z_dim], -eps)\n",
    "    c = tf.zeros(shape = [z_dim], dtype = tf.float32, name = 'c')\n",
    "    n_sample = 0\n",
    "    for batch, (batch_x) in enumerate(target_dataset):\n",
    "        center= tf.reduce_sum(encoder(generate(batch_x, training = False), training = False), axis = 0, name = 'outputs')\n",
    "        c = c +center\n",
    "        n_sample = n_sample + batch_x.shape[0]\n",
    "    c = c/n_sample\n",
    "    # If c_i is too close to 0, set to +-eps. Reason: a zero unit can be trivially matched with zero weights.\n",
    "\n",
    "    c = tf.where(tf.logical_and(tf.less(c, 0), tf.greater(c, -eps)), negative, c)\n",
    "    c = tf.where(tf.logical_and(tf.greater(c, 0), tf.less(c, eps)), positive, c)\n",
    "    return c\n",
    "\n",
    "\n",
    "@tf.function\n",
    "def train_ae_model(batch_x):\n",
    "    with tf.GradientTape() as ae_tape:\n",
    "        generate_output = generate(batch_x, training =False)\n",
    "        ae_output = decoder(encoder(generate_output, training=True), training=True)\n",
    "        loss = tf.reduce_mean((ae_output - generate_output) ** 2)\n",
    "\n",
    "    grads = ae_tape.gradient(loss, encoder.trainable_variables + decoder.trainable_variables)\n",
    "    optimizer.apply_gradients(zip(grads, encoder.trainable_variables + decoder.trainable_variables))\n",
    "\n",
    "    return loss\n",
    "\n",
    "@tf.function\n",
    "def train_model(batch_x, c, R, objective):\n",
    "    with tf.GradientTape() as tape:\n",
    "        generate_output = generate(batch_x, training =False)\n",
    "        output = encoder(generate_output, training=True)\n",
    "        dist = tf.reduce_sum((output - c) ** 2, 1)\n",
    "      \n",
    "        if objective == 'soft-boundary':\n",
    "            scores = dist - R ** 2\n",
    "            output_loss = R ** 2 + (1 / nu) * tf.reduce_mean(tf.maximum(tf.zeros_like(scores), scores))\n",
    "        else:\n",
    "            output_loss = tf.reduce_mean(dist)\n",
    "\n",
    "    grads = tape.gradient(output_loss, encoder.trainable_variables)\n",
    "    optimizer.apply_gradients(zip(grads, encoder.trainable_variables))\n",
    "\n",
    "    return output_loss, dist"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Prtrain"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   0: TIME: 4.10  LOSS: 0.3587 \n",
      "  10: TIME: 2.87  LOSS: 0.1912 \n",
      "  20: TIME: 2.78  LOSS: 0.1583 \n",
      "  30: TIME: 2.80  LOSS: 0.1489 \n",
      "  40: TIME: 2.77  LOSS: 0.1447 \n",
      "  50: TIME: 2.81  LOSS: 0.1406 \n",
      "learning rate changed!\n",
      "  60: TIME: 2.79  LOSS: 0.1358 \n",
      "  70: TIME: 2.81  LOSS: 0.1339 \n",
      "  80: TIME: 2.76  LOSS: 0.1323 \n",
      "  90: TIME: 2.77  LOSS: 0.1311 \n"
     ]
    }
   ],
   "source": [
    "global_step = 0  \n",
    "step_size = 2 * np.ceil(n_samples / batch_size)\n",
    "\n",
    "for epoch in range(n_ae_epochs):\n",
    "    start = time.time()\n",
    "\n",
    "    # Learning rate schedule\n",
    "    if epoch in [60, 100, 300]:\n",
    "        base_lr = base_lr / 2\n",
    "        max_lr = max_lr / 2\n",
    "        step_size = step_size / 2\n",
    "\n",
    "        print('learning rate changed!')\n",
    "\n",
    "    epoch_ae_loss_avg = tf.metrics.Mean()\n",
    "    \n",
    "    for batch, (batch_x) in enumerate(target_dataset):\n",
    "         # Calculate cyclic learning rate\n",
    "        global_step = global_step + 1\n",
    "        cycle = np.floor(1 + global_step / (2 * step_size))\n",
    "        x_lr = np.abs(global_step / step_size - 2 * cycle + 1)\n",
    "        clr = base_lr + (max_lr - base_lr) * max(0, 1 - x_lr)\n",
    "\n",
    "        optimizer.lr = clr\n",
    "        ae_loss = train_ae_model(batch_x)\n",
    "        \n",
    "        epoch_ae_loss_avg(ae_loss)\n",
    "        \n",
    "    if epoch % 10 == 0:\n",
    "        epoch_time = time.time() - start\n",
    "        print('{:4d}: TIME: {:.2f}  LOSS: {:.4f} '.format(epoch, epoch_time, epoch_ae_loss_avg.result()))\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   0: TIME: 3.36  LOSS: 0.3296 \n",
      "  10: TIME: 2.73  LOSS: 0.0222 \n",
      "  20: TIME: 2.85  LOSS: 0.0118 \n",
      "  30: TIME: 2.78  LOSS: 0.0074 \n",
      "  40: TIME: 2.73  LOSS: 0.0075 \n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "R = tf.zeros(shape = [1], dtype = tf.float32, name = 'R')\n",
    "c = init_center_c(eps = 0.1)\n",
    "\n",
    "pre_loss = 0\n",
    "loss_list = []\n",
    "\n",
    "for epoch in range(n_epochs):\n",
    "    start = time.time()\n",
    "\n",
    "    # Learning rate schedule\n",
    "    if epoch in [60, 100, 300]:\n",
    "        base_lr = base_lr / 2\n",
    "        max_lr = max_lr / 2\n",
    "        step_size = step_size / 2\n",
    "\n",
    "        print('learning rate changed!')\n",
    "\n",
    "    epoch_loss_avg = tf.metrics.Mean()    \n",
    "   \n",
    "    for batch, (batch_x) in enumerate(target_dataset):\n",
    "\n",
    "        # Calculate cyclic learning rate\n",
    "        global_step = global_step + 1\n",
    "        cycle = np.floor(1 + global_step / (2 * step_size))\n",
    "        x_lr = np.abs(global_step / step_size - 2 * cycle + 1)\n",
    "        clr = base_lr + (max_lr - base_lr) * max(0, 1 - x_lr)\n",
    "\n",
    "        optimizer.lr = clr\n",
    "        loss_, dist = train_model(batch_x, c, R, objective)\n",
    "        \n",
    "        epoch_loss_avg(loss_)\n",
    "\n",
    "    if (objective == 'soft-boundary') and (epoch >= warm_up_n_epochs):\n",
    "        R = get_radius(dist, nu)\n",
    "        \n",
    "    loss_list.append(loss_)\n",
    "    if abs(pre_loss - loss_) < 1e-8:\n",
    "        break\n",
    "    pre_loss = loss_\n",
    "\n",
    "    if epoch % 10 == 0:\n",
    "        epoch_time = time.time() - start\n",
    "        print('{:4d}: TIME: {:.2f}  LOSS: {:.4f} '.format(epoch, epoch_time, epoch_loss_avg.result()))\n",
    "\n",
    "plt.plot(loss_list)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {},
   "outputs": [],
   "source": [
    "def dist(target_dataset, center):\n",
    "    i = 0\n",
    "    for batch, (batch_x) in enumerate(target_dataset):\n",
    "        \n",
    "        generate_output = generate(batch_x, training =False)\n",
    "        output = encoder(generate_output, training=True)\n",
    "        distance = tf.reduce_sum(tf.square(output - center), 1) \n",
    "        if i==0:\n",
    "            distances=distance\n",
    "        else:\n",
    "            distances = tf.concat([distances, distance], 0) \n",
    "        i = i + 1\n",
    "    dis = distances.numpy() \n",
    "    plt.hist(dis, rwidth=0.3, color='b')\n",
    "    plt.show()\n",
    "\n",
    "    return distances"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {},
   "outputs": [],
   "source": [
    "def threshold(data, dists):\n",
    "    dict_score = {}\n",
    "    for i in range(len(dists)):\n",
    "        dict_score[float(dists[i])] = i\n",
    "    sort_score = [(k, dict_score[k]) for k in sorted(dict_score.keys())]\n",
    "    \n",
    "    n_digits = 15  # how many digits we will display\n",
    "    fig = plt.figure(figsize=(20, 4))\n",
    "    for i in range(1, n_digits):\n",
    "        # display normal\n",
    "        ax = plt.subplot(2, n_digits, i + 1)\n",
    "        plt.imshow(data[sort_score[i][1]].reshape(img_size, img_size,3), cmap = plt.cm.gray)\n",
    "        ax.get_xaxis().set_visible(False)\n",
    "        ax.get_yaxis().set_visible(False)\n",
    "\n",
    "        # display anomaly\n",
    "        ax = plt.subplot(2, n_digits, i + 1 + n_digits)\n",
    "        plt.imshow(data[sort_score[-i][1]].reshape(img_size, img_size,3), cmap = plt.cm.gray)\n",
    "        ax.get_xaxis().set_visible(False)\n",
    "        ax.get_yaxis().set_visible(False)\n",
    "\n",
    "    plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "dists = dist(target_dataset, c)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x288 with 14 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "img_size = 224\n",
    "threshold(Tdata_train, dists)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 測試用的資料"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "metadata": {},
   "outputs": [],
   "source": [
    "target_test_x = np.vstack((Rdata_test,Tdata_test))\n",
    "target_test_y = [0] * len(Rdata_test) + [1] * len(Tdata_test)\n",
    "target_test_x = np.array(target_test_x)\n",
    "target_test_y = np.array(target_test_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {},
   "outputs": [],
   "source": [
    "idx = np.random.permutation(len(target_test_x))\n",
    "target_test_x, target_test_y = target_test_x[idx], target_test_y[idx]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = []\n",
    "y_scores = []\n",
    "result_test = np.zeros([2, len(target_test_x)]) \n",
    "for i in range(len(target_test_x)):\n",
    "    \n",
    "    genrate_output = generate(tf.expand_dims(target_test_x[i],0), training =False)\n",
    "    dist = tf.reduce_sum(tf.square(encoder(genrate_output, training=False) - c), 1)\n",
    "    \n",
    "    if objective == 'soft-boundary':\n",
    "        scores = dist - R ** 2\n",
    "    else:\n",
    "        scores = dist\n",
    "        \n",
    "    result_test[0, i] = target_test_y[i]\n",
    "    result_test[1, i] = 1/scores\n",
    "    y_true.append(result_test[0, i])\n",
    "    y_scores.append(result_test[1, i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "AUC: 0.8923755324813631\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import roc_auc_score\n",
    "test_auc = roc_auc_score(y_true, y_scores)\n",
    "print('AUC:', test_auc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x288 with 28 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "threshold(target_test_x, y_scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = np.array(y_true)\n",
    "y_scores = np.array(y_scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(y_scores[y_true==1.], rwidth = 0.3, color = 'r',label = 'normal')\n",
    "plt.hist(y_scores[y_true==0.], rwidth = 0.3, color = 'b',label = 'anomaly')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 173,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([50., 70., 57., 33., 10.,  9.,  6.,  4.,  0.,  1.]),\n",
       " array([-4.74951230e-06,  4.26634611e-04,  8.58018734e-04,  1.28940286e-03,\n",
       "         1.72078698e-03,  2.15217110e-03,  2.58355523e-03,  3.01493935e-03,\n",
       "         3.44632347e-03,  3.87770759e-03,  4.30909172e-03]),\n",
       " <a list of 10 Patch objects>)"
      ]
     },
     "execution_count": 173,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(y_scores[y_true==1.], rwidth = 0.3, color = 'r',label = 'normal')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([2.481e+03, 1.400e+01, 4.000e+00, 4.000e+00, 0.000e+00, 0.000e+00,\n",
       "        0.000e+00, 0.000e+00, 0.000e+00, 1.000e+00]),\n",
       " array([7.68842292e-05, 5.74099906e-02, 1.14743097e-01, 1.72076203e-01,\n",
       "        2.29409310e-01, 2.86742416e-01, 3.44075523e-01, 4.01408629e-01,\n",
       "        4.58741735e-01, 5.16074842e-01, 5.73407948e-01]),\n",
       " <a list of 10 Patch objects>)"
      ]
     },
     "execution_count": 174,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(y_scores[y_true==0.], rwidth = 0.3, color = 'b',label = 'anomaly')\n"
   ]
  }
 ],
 "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": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}