From d4259a07d52d7aeb3d22fb7a3462962a2a3bbe84 Mon Sep 17 00:00:00 2001 From: Guna Venkat Date: Sat, 3 Jan 2026 17:50:11 +0530 Subject: [PATCH] adding assginment-3 file DoddiGunaVenkat --- .../DoddiGunaVenkat-Assignment3.ipynb | 1 + 1 file changed, 1 insertion(+) create mode 100644 Assignment 3/Assignment3_DoddiGunaVenkat/DoddiGunaVenkat-Assignment3.ipynb diff --git a/Assignment 3/Assignment3_DoddiGunaVenkat/DoddiGunaVenkat-Assignment3.ipynb b/Assignment 3/Assignment3_DoddiGunaVenkat/DoddiGunaVenkat-Assignment3.ipynb new file mode 100644 index 00000000..175524f4 --- /dev/null +++ b/Assignment 3/Assignment3_DoddiGunaVenkat/DoddiGunaVenkat-Assignment3.ipynb @@ -0,0 +1 @@ +{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.12.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"gpu","dataSources":[{"sourceId":8077192,"sourceType":"datasetVersion","datasetId":4766889}],"dockerImageVersionId":31236,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"## Import Necessary Libraries","metadata":{}},{"cell_type":"code","source":"import os\nimport numpy as np\nimport pandas as pd\nimport matplotlib.pyplot as plt\nimport seaborn as sns\n\nimport tensorflow as tf\nfrom tensorflow.keras.preprocessing.image import ImageDataGenerator\nfrom tensorflow.keras import layers, models, optimizers\nfrom tensorflow.keras.applications import ResNet50\n\nfrom sklearn.metrics import (\n classification_report,\n confusion_matrix,\n roc_auc_score,\n roc_curve,\n f1_score\n)\n\n# set seeds for reproducibility\n\nSEED = 42\ntf.random.set_seed(SEED)\nnp.random.seed(SEED)\n","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true,"execution":{"iopub.status.busy":"2026-01-02T17:44:07.760746Z","iopub.execute_input":"2026-01-02T17:44:07.761295Z","iopub.status.idle":"2026-01-02T17:44:22.444362Z","shell.execute_reply.started":"2026-01-02T17:44:07.761265Z","shell.execute_reply":"2026-01-02T17:44:22.443768Z"}},"outputs":[{"name":"stderr","text":"2026-01-02 17:44:10.458331: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:467] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\nWARNING: All log messages before absl::InitializeLog() is called are written to STDERR\nE0000 00:00:1767375850.640924 55 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\nE0000 00:00:1767375850.698029 55 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\nW0000 00:00:1767375851.141277 55 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.\nW0000 00:00:1767375851.141316 55 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.\nW0000 00:00:1767375851.141319 55 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.\nW0000 00:00:1767375851.141321 55 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.\n","output_type":"stream"}],"execution_count":1},{"cell_type":"code","source":"# Define Dataset Paths\n\ntrain_ds_path = '/kaggle/input/candle-image-data/Train'\ntest_ds_path = '/kaggle/input/candle-image-data/Test'","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-02T17:44:22.445476Z","iopub.execute_input":"2026-01-02T17:44:22.445992Z","iopub.status.idle":"2026-01-02T17:44:22.450195Z","shell.execute_reply.started":"2026-01-02T17:44:22.445966Z","shell.execute_reply":"2026-01-02T17:44:22.449427Z"}},"outputs":[],"execution_count":2},{"cell_type":"markdown","source":"## Data Exploration & Visualization","metadata":{}},{"cell_type":"code","source":"# count sample and class distributions\n\ndef count_images(path):\n counts = {}\n total = 0\n for cls in os.listdir(path):\n cls_path = os.path.join(path, cls)\n if os.path.isdir(cls_path):\n n = len(os.listdir(cls_path))\n counts[cls] = n\n total += n\n return counts, total\n\ntrain_counts, train_total = count_images(train_ds_path)\ntest_counts, test_total = count_images(test_ds_path)\n\nprint(\"Train Distribution:\", train_counts, \"Total:\", train_total)\nprint(\"Test Distribution:\", test_counts, \"Total:\", test_total)\n\n# visualize class Imbalance\n\nplt.figure(figsize=(6,4))\nsns.barplot(x=list(train_counts.keys()), y=list(train_counts.values()))\nplt.title(\"Train Class Distribution\")\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-02T17:44:22.451113Z","iopub.execute_input":"2026-01-02T17:44:22.451396Z","iopub.status.idle":"2026-01-02T17:44:22.712390Z","shell.execute_reply.started":"2026-01-02T17:44:22.451366Z","shell.execute_reply":"2026-01-02T17:44:22.711813Z"}},"outputs":[{"name":"stdout","text":"Train Distribution: {'Down': 624, 'Up': 809} Total: 1433\nTest Distribution: {'Down': 157, 'Up': 194} Total: 351\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"
","image/png":"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\n"},"metadata":{}}],"execution_count":3},{"cell_type":"code","source":"# display sample images\n# Top row → UP, Bottom row → DOWN\ndef show_samples(base_path):\n classes = ['Up', 'Down']\n fig, axes = plt.subplots(2, 4, figsize=(14, 6))\n\n for row, cls in enumerate(classes):\n cls_path = os.path.join(base_path, cls)\n images = os.listdir(cls_path)[:4]\n\n for col, img_name in enumerate(images):\n img_path = os.path.join(cls_path, img_name)\n img = plt.imread(img_path)\n axes[row, col].imshow(img)\n axes[row, col].axis('on')\n axes[row, col].set_title(cls+\" \"+img_name)\n\n plt.tight_layout()\n plt.show()\n\nshow_samples(train_ds_path)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-02T17:44:22.714247Z","iopub.execute_input":"2026-01-02T17:44:22.714519Z","iopub.status.idle":"2026-01-02T17:44:23.836255Z","shell.execute_reply.started":"2026-01-02T17:44:22.714498Z","shell.execute_reply":"2026-01-02T17:44:23.835495Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"
","image/png":"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\n"},"metadata":{}}],"execution_count":4},{"cell_type":"markdown","source":"### Data Augmentation\n\nIn **candlestick chart images**, the goal of data augmentation is to **add small, realistic variations without changing the market meaning**.\n\n**Why zooming, shifting, rotation are OK**\n\n* **Small rotation (±10°)**: simulates slightly different chart orientations or capture noise without changing price relationships.\n* **Zooming (±10%)**: mimics different time-window scales while preserving the trend structure.\n* **Width/height shifts**: represent minor cropping or alignment differences; the relative candle order stays the same.\n* **Brightness change**: simulates different rendering styles or lighting without affecting price patterns.\n\nThese augmentations **preserve the semantic meaning** of the candlestick patterns.\n\n**Why horizontal or vertical flip is NOT OK**\n\n* **Horizontal flip** reverses time → future appears as past, which is invalid in financial data.\n* **Vertical flip** inverts prices → bullish patterns become bearish (high ↔ low), completely changing the market signal.\n\nSo in short:\n\n> **Allowed augmentations add noise without changing financial meaning; flips distort time or price direction, breaking the logic of candlestick patterns.**\n","metadata":{}},{"cell_type":"code","source":"# Data Augmentation\n\nIMG_SIZE = (224, 224)\nBATCH_SIZE = 32\n\ntrain_datagen = ImageDataGenerator(\n rescale=1./255, # scale image\n rotation_range=10, # rotate\n zoom_range=0.1, # zoom\n width_shift_range=0.05, # width shift\n height_shift_range=0.05, # height shift\n brightness_range=[0.9,1.1],\n horizontal_flip=False\n)\n\ntest_datagen = ImageDataGenerator(rescale=1./255)\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-02T17:44:23.837253Z","iopub.execute_input":"2026-01-02T17:44:23.837554Z","iopub.status.idle":"2026-01-02T17:44:23.842415Z","shell.execute_reply.started":"2026-01-02T17:44:23.837523Z","shell.execute_reply":"2026-01-02T17:44:23.841657Z"}},"outputs":[],"execution_count":5},{"cell_type":"code","source":"# Image dataLoaders\n\ntrain_gen = train_datagen.flow_from_directory(\n train_ds_path,\n target_size=IMG_SIZE,\n batch_size=BATCH_SIZE,\n class_mode='binary',\n shuffle=True\n)\n\ntest_gen = test_datagen.flow_from_directory(\n test_ds_path,\n target_size=IMG_SIZE,\n batch_size=BATCH_SIZE,\n class_mode='binary',\n shuffle=False\n)\n\nprint(train_gen.class_indices)\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-02T17:44:23.843300Z","iopub.execute_input":"2026-01-02T17:44:23.843594Z","iopub.status.idle":"2026-01-02T17:44:25.074982Z","shell.execute_reply.started":"2026-01-02T17:44:23.843574Z","shell.execute_reply":"2026-01-02T17:44:25.074275Z"}},"outputs":[{"name":"stdout","text":"Found 1433 images belonging to 2 classes.\nFound 351 images belonging to 2 classes.\n{'Down': 0, 'Up': 1}\n","output_type":"stream"}],"execution_count":6},{"cell_type":"code","source":"# Visualize Augmentations\n\ndef show_augmentations(class_name):\n img_path = os.path.join(train_ds_path, class_name)\n img_name = os.listdir(img_path)[0]\n img = tf.keras.preprocessing.image.load_img(\n os.path.join(img_path, img_name),\n target_size=IMG_SIZE\n ) # load image\n img = tf.keras.preprocessing.image.img_to_array(img) # convert to array\n img = img.reshape((1,) + img.shape) # reshape\n\n aug_iter = train_datagen.flow(img, batch_size=1) # data augmentation\n\n fig, axes = plt.subplots(1, 4, figsize=(14,4))\n titles = [\"Rotation\", \"Zoom\", \"Shift\", \"Brightness\"]\n\n for i in range(4):\n aug_img = next(aug_iter)[0]\n axes[i].imshow(aug_img)\n axes[i].set_title(titles[i])\n axes[i].axis('off')\n\n plt.show()\n\nprint(\"UP Augmentations\")\nshow_augmentations('Up')\n\nprint(\"DOWN Augmentations\")\nshow_augmentations('Down')","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-02T17:44:25.075980Z","iopub.execute_input":"2026-01-02T17:44:25.076356Z","iopub.status.idle":"2026-01-02T17:44:25.490077Z","shell.execute_reply.started":"2026-01-02T17:44:25.076321Z","shell.execute_reply":"2026-01-02T17:44:25.489500Z"}},"outputs":[{"name":"stdout","text":"UP Augmentations\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"
","image/png":"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\n"},"metadata":{}},{"name":"stdout","text":"DOWN Augmentations\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"
","image/png":"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\n"},"metadata":{}}],"execution_count":7},{"cell_type":"markdown","source":"## Model Building & Training\n\n**We tried three types of models:**\n\n1. **Basic CNN (from scratch)**\n\n * Simple convolution + pooling layers.\n * Learns low-level candlestick features (edges, shapes).\n * Used as a **baseline** to see if patterns can be learned at all.\n\n2. **Regularized CNN (improved from scratch)**\n\n * Adds **Batch Normalization** and **Dropout**.\n * Helps reduce overfitting and stabilizes training.\n * Learns more robust candlestick features than the basic CNN.\n\n3. **Transfer Learning (ResNet50)**\n\n * Uses a **pretrained ResNet50** (ImageNet) as a feature extractor.\n * CNN backbone is frozen; only the custom classification head is trained.\n * Leverages **strong visual features** (edges, textures, shapes) to improve performance with limited data.\n\n> We progressed from **simple → regularized → pretrained deep model** to compare learning from scratch vs. using powerful pretrained representations for candlestick pattern classification.\n","metadata":{}},{"cell_type":"code","source":"# models and helper functions\n\ndef build_basic_cnn():\n model = models.Sequential([\n layers.Conv2D(32, (3,3), activation='relu', input_shape=(*IMG_SIZE,3)),\n layers.MaxPooling2D(2,2),\n\n layers.Conv2D(64, (3,3), activation='relu'),\n layers.MaxPooling2D(2,2),\n\n layers.Flatten(),\n layers.Dense(128, activation='relu'),\n layers.Dense(1, activation='sigmoid')\n ])\n return model\n\ndef build_regularized_cnn():\n model = models.Sequential([\n layers.Conv2D(32, (3,3), padding='same', input_shape=(*IMG_SIZE,3)),\n layers.BatchNormalization(),\n layers.Activation('relu'),\n layers.MaxPooling2D(),\n\n layers.Conv2D(64, (3,3), padding='same'),\n layers.BatchNormalization(),\n layers.Activation('relu'),\n layers.MaxPooling2D(),\n\n layers.Flatten(),\n layers.Dense(256, activation='relu'),\n layers.Dropout(0.5),\n layers.Dense(1, activation='sigmoid')\n ])\n return model\n\n\ndef plot_history(history):\n fig, axes = plt.subplots(1,2, figsize=(14,5))\n\n axes[0].plot(history.history['accuracy'], label='Train')\n axes[0].plot(history.history['val_accuracy'], label='Test')\n axes[0].set_title('Accuracy')\n axes[0].legend()\n\n axes[1].plot(history.history['loss'], label='Train')\n axes[1].plot(history.history['val_loss'], label='Test')\n axes[1].set_title('Loss')\n axes[1].legend()\n\n plt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-02T17:52:09.689809Z","iopub.execute_input":"2026-01-02T17:52:09.690526Z","iopub.status.idle":"2026-01-02T17:52:09.698238Z","shell.execute_reply.started":"2026-01-02T17:52:09.690489Z","shell.execute_reply":"2026-01-02T17:52:09.697566Z"}},"outputs":[],"execution_count":12},{"cell_type":"code","source":"# check for GPU Compatibility\n\nimport tensorflow as tf\n\nprint(\"TensorFlow version:\", tf.__version__)\n\ngpus = tf.config.list_physical_devices('GPU')\n\nif gpus:\n print(\"GPUs detected:\", gpus)\nelse:\n print(\"❌ No GPU detected\")\n \nwith tf.device('/GPU:0'):\n a = tf.random.normal([1000,1000])\n b = tf.random.normal([1000,1000])\n c = tf.matmul(a, b)\n\nprint(\"Matrix multiplication completed\")\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-02T17:44:37.917471Z","iopub.execute_input":"2026-01-02T17:44:37.918110Z","iopub.status.idle":"2026-01-02T17:44:38.674163Z","shell.execute_reply.started":"2026-01-02T17:44:37.918080Z","shell.execute_reply":"2026-01-02T17:44:38.673445Z"}},"outputs":[{"name":"stdout","text":"TensorFlow version: 2.19.0\nGPUs detected: [PhysicalDevice(name='/physical_device:GPU:0', device_type='GPU')]\nMatrix multiplication completed\n","output_type":"stream"},{"name":"stderr","text":"I0000 00:00:1767375878.565106 55 gpu_device.cc:2019] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 15513 MB memory: -> device: 0, name: Tesla P100-PCIE-16GB, pci bus id: 0000:00:04.0, compute capability: 6.0\n","output_type":"stream"}],"execution_count":9},{"cell_type":"code","source":"# Base Model\n\nbasic_model = build_basic_cnn()\nbasic_model.compile(\n optimizer='adam',\n loss='binary_crossentropy',\n metrics=['accuracy']\n)\n\nbasic_model.summary()\n\nEPOCHS = 15\n\nbase_history = basic_model.fit(\n train_gen,\n validation_data=test_gen,\n epochs=EPOCHS\n)\n\nplot_history(base_history)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-02T17:44:50.622871Z","iopub.execute_input":"2026-01-02T17:44:50.623478Z","iopub.status.idle":"2026-01-02T17:50:27.928667Z","shell.execute_reply.started":"2026-01-02T17:44:50.623449Z","shell.execute_reply":"2026-01-02T17:50:27.927930Z"}},"outputs":[{"name":"stderr","text":"/usr/local/lib/python3.12/dist-packages/keras/src/layers/convolutional/base_conv.py:113: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"\u001b[1mModel: \"sequential\"\u001b[0m\n","text/html":"
Model: \"sequential\"\n
\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n┃\u001b[1m \u001b[0m\u001b[1mLayer (type) \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m Param #\u001b[0m\u001b[1m \u001b[0m┃\n┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n│ conv2d (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m222\u001b[0m, \u001b[38;5;34m222\u001b[0m, \u001b[38;5;34m32\u001b[0m) │ \u001b[38;5;34m896\u001b[0m │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ max_pooling2d (\u001b[38;5;33mMaxPooling2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m111\u001b[0m, \u001b[38;5;34m111\u001b[0m, \u001b[38;5;34m32\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ conv2d_1 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m109\u001b[0m, \u001b[38;5;34m109\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m18,496\u001b[0m │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ max_pooling2d_1 (\u001b[38;5;33mMaxPooling2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m54\u001b[0m, \u001b[38;5;34m54\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ flatten (\u001b[38;5;33mFlatten\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m186624\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ dense (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m) │ \u001b[38;5;34m23,888,000\u001b[0m │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ dense_1 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m) │ \u001b[38;5;34m129\u001b[0m │\n└─────────────────────────────────┴────────────────────────┴───────────────┘\n","text/html":"
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n┃ Layer (type)                    ┃ Output Shape           ┃       Param # ┃\n┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n│ conv2d (Conv2D)                 │ (None, 222, 222, 32)   │           896 │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ max_pooling2d (MaxPooling2D)    │ (None, 111, 111, 32)   │             0 │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ conv2d_1 (Conv2D)               │ (None, 109, 109, 64)   │        18,496 │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ max_pooling2d_1 (MaxPooling2D)  │ (None, 54, 54, 64)     │             0 │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ flatten (Flatten)               │ (None, 186624)         │             0 │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ dense (Dense)                   │ (None, 128)            │    23,888,000 │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ dense_1 (Dense)                 │ (None, 1)              │           129 │\n└─────────────────────────────────┴────────────────────────┴───────────────┘\n
\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"\u001b[1m Total params: \u001b[0m\u001b[38;5;34m23,907,521\u001b[0m (91.20 MB)\n","text/html":"
 Total params: 23,907,521 (91.20 MB)\n
\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m23,907,521\u001b[0m (91.20 MB)\n","text/html":"
 Trainable params: 23,907,521 (91.20 MB)\n
\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n","text/html":"
 Non-trainable params: 0 (0.00 B)\n
\n"},"metadata":{}},{"name":"stderr","text":"/usr/local/lib/python3.12/dist-packages/keras/src/trainers/data_adapters/py_dataset_adapter.py:121: UserWarning: Your `PyDataset` class should call `super().__init__(**kwargs)` in its constructor. `**kwargs` can include `workers`, `use_multiprocessing`, `max_queue_size`. Do not pass these arguments to `fit()`, as they will be ignored.\n self._warn_if_super_not_called()\n","output_type":"stream"},{"name":"stdout","text":"Epoch 1/15\n","output_type":"stream"},{"name":"stderr","text":"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\nI0000 00:00:1767375894.798313 154 service.cc:152] XLA service 0x7db6dc004700 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\nI0000 00:00:1767375894.798349 154 service.cc:160] StreamExecutor device (0): Tesla P100-PCIE-16GB, Compute Capability 6.0\nI0000 00:00:1767375895.188856 154 cuda_dnn.cc:529] Loaded cuDNN version 91002\n","output_type":"stream"},{"name":"stdout","text":"\u001b[1m 1/45\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m4:02\u001b[0m 6s/step - accuracy: 0.5000 - loss: 0.6944","output_type":"stream"},{"name":"stderr","text":"I0000 00:00:1767375898.566734 154 device_compiler.h:188] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n","output_type":"stream"},{"name":"stdout","text":"\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m34s\u001b[0m 637ms/step - accuracy: 0.5297 - loss: 0.7444 - val_accuracy: 0.5527 - val_loss: 0.6882\nEpoch 2/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 474ms/step - accuracy: 0.5600 - loss: 0.6862 - val_accuracy: 0.5527 - val_loss: 0.6880\nEpoch 3/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 477ms/step - accuracy: 0.5631 - loss: 0.6846 - val_accuracy: 0.5527 - val_loss: 0.6894\nEpoch 4/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 485ms/step - accuracy: 0.5492 - loss: 0.6843 - val_accuracy: 0.5527 - val_loss: 0.6875\nEpoch 5/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 473ms/step - accuracy: 0.5520 - loss: 0.6877 - val_accuracy: 0.5527 - val_loss: 0.6918\nEpoch 6/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 479ms/step - accuracy: 0.5528 - loss: 0.6871 - val_accuracy: 0.5499 - val_loss: 0.6911\nEpoch 7/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 475ms/step - accuracy: 0.5655 - loss: 0.6795 - val_accuracy: 0.5527 - val_loss: 0.7107\nEpoch 8/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 478ms/step - accuracy: 0.5684 - loss: 0.6801 - val_accuracy: 0.5556 - val_loss: 0.6994\nEpoch 9/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 479ms/step - accuracy: 0.5788 - loss: 0.6734 - val_accuracy: 0.5413 - val_loss: 0.6922\nEpoch 10/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 482ms/step - accuracy: 0.5781 - loss: 0.6803 - val_accuracy: 0.5470 - val_loss: 0.7199\nEpoch 11/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 478ms/step - accuracy: 0.5835 - loss: 0.6726 - val_accuracy: 0.5356 - val_loss: 0.6904\nEpoch 12/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 477ms/step - accuracy: 0.5706 - loss: 0.6672 - val_accuracy: 0.5385 - val_loss: 0.6984\nEpoch 13/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 474ms/step - accuracy: 0.5817 - loss: 0.6639 - val_accuracy: 0.5271 - val_loss: 0.7148\nEpoch 14/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 477ms/step - accuracy: 0.6071 - loss: 0.6639 - val_accuracy: 0.4929 - val_loss: 0.7243\nEpoch 15/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 475ms/step - accuracy: 0.6022 - loss: 0.6662 - val_accuracy: 0.5128 - val_loss: 0.7407\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"
","image/png":"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\n"},"metadata":{}}],"execution_count":10},{"cell_type":"code","source":"\nreg_model = build_regularized_cnn()\nreg_model.compile(\n optimizer='adam',\n loss='binary_crossentropy',\n metrics=['accuracy']\n)\n\nreg_model.summary()\n\nEPOCHS = 15\n\nhistory = reg_model.fit(\n train_gen,\n validation_data=test_gen,\n epochs=EPOCHS\n)\n\nplot_history(history)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-02T17:52:14.372090Z","iopub.execute_input":"2026-01-02T17:52:14.372670Z","iopub.status.idle":"2026-01-02T17:57:46.613990Z","shell.execute_reply.started":"2026-01-02T17:52:14.372643Z","shell.execute_reply":"2026-01-02T17:57:46.613362Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"\u001b[1mModel: \"sequential_2\"\u001b[0m\n","text/html":"
Model: \"sequential_2\"\n
\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n┃\u001b[1m \u001b[0m\u001b[1mLayer (type) \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m Param #\u001b[0m\u001b[1m \u001b[0m┃\n┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n│ conv2d_4 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m32\u001b[0m) │ \u001b[38;5;34m896\u001b[0m │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ batch_normalization_2 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m32\u001b[0m) │ \u001b[38;5;34m128\u001b[0m │\n│ (\u001b[38;5;33mBatchNormalization\u001b[0m) │ │ │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ activation_2 (\u001b[38;5;33mActivation\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m32\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ max_pooling2d_4 (\u001b[38;5;33mMaxPooling2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m32\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ conv2d_5 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m18,496\u001b[0m │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ batch_normalization_3 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m256\u001b[0m │\n│ (\u001b[38;5;33mBatchNormalization\u001b[0m) │ │ │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ activation_3 (\u001b[38;5;33mActivation\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ max_pooling2d_5 (\u001b[38;5;33mMaxPooling2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ flatten_2 (\u001b[38;5;33mFlatten\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m200704\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ dense_4 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m) │ \u001b[38;5;34m51,380,480\u001b[0m │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ dropout_1 (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ dense_5 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m) │ \u001b[38;5;34m257\u001b[0m │\n└─────────────────────────────────┴────────────────────────┴───────────────┘\n","text/html":"
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n┃ Layer (type)                    ┃ Output Shape           ┃       Param # ┃\n┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n│ conv2d_4 (Conv2D)               │ (None, 224, 224, 32)   │           896 │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ batch_normalization_2           │ (None, 224, 224, 32)   │           128 │\n│ (BatchNormalization)            │                        │               │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ activation_2 (Activation)       │ (None, 224, 224, 32)   │             0 │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ max_pooling2d_4 (MaxPooling2D)  │ (None, 112, 112, 32)   │             0 │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ conv2d_5 (Conv2D)               │ (None, 112, 112, 64)   │        18,496 │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ batch_normalization_3           │ (None, 112, 112, 64)   │           256 │\n│ (BatchNormalization)            │                        │               │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ activation_3 (Activation)       │ (None, 112, 112, 64)   │             0 │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ max_pooling2d_5 (MaxPooling2D)  │ (None, 56, 56, 64)     │             0 │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ flatten_2 (Flatten)             │ (None, 200704)         │             0 │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ dense_4 (Dense)                 │ (None, 256)            │    51,380,480 │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ dropout_1 (Dropout)             │ (None, 256)            │             0 │\n├─────────────────────────────────┼────────────────────────┼───────────────┤\n│ dense_5 (Dense)                 │ (None, 1)              │           257 │\n└─────────────────────────────────┴────────────────────────┴───────────────┘\n
\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"\u001b[1m Total params: \u001b[0m\u001b[38;5;34m51,400,513\u001b[0m (196.08 MB)\n","text/html":"
 Total params: 51,400,513 (196.08 MB)\n
\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m51,400,321\u001b[0m (196.08 MB)\n","text/html":"
 Trainable params: 51,400,321 (196.08 MB)\n
\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m192\u001b[0m (768.00 B)\n","text/html":"
 Non-trainable params: 192 (768.00 B)\n
\n"},"metadata":{}},{"name":"stdout","text":"Epoch 1/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m27s\u001b[0m 513ms/step - accuracy: 0.5367 - loss: 25.5425 - val_accuracy: 0.5356 - val_loss: 0.6930\nEpoch 2/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 478ms/step - accuracy: 0.5143 - loss: 0.7568 - val_accuracy: 0.5527 - val_loss: 0.6915\nEpoch 3/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 476ms/step - accuracy: 0.5684 - loss: 0.6894 - val_accuracy: 0.5527 - val_loss: 0.6895\nEpoch 4/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 488ms/step - accuracy: 0.5451 - loss: 0.6907 - val_accuracy: 0.5527 - val_loss: 0.6904\nEpoch 5/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m23s\u001b[0m 514ms/step - accuracy: 0.5606 - loss: 0.6901 - val_accuracy: 0.5527 - val_loss: 0.6895\nEpoch 6/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 493ms/step - accuracy: 0.5726 - loss: 0.6806 - val_accuracy: 0.5527 - val_loss: 0.6914\nEpoch 7/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 478ms/step - accuracy: 0.5647 - loss: 0.6836 - val_accuracy: 0.5527 - val_loss: 0.6893\nEpoch 8/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 480ms/step - accuracy: 0.5789 - loss: 0.6817 - val_accuracy: 0.5527 - val_loss: 0.6907\nEpoch 9/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 475ms/step - accuracy: 0.5744 - loss: 0.6750 - val_accuracy: 0.5527 - val_loss: 0.6903\nEpoch 10/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 475ms/step - accuracy: 0.5669 - loss: 0.6822 - val_accuracy: 0.5527 - val_loss: 0.6929\nEpoch 11/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 484ms/step - accuracy: 0.5677 - loss: 0.6838 - val_accuracy: 0.5527 - val_loss: 0.7002\nEpoch 12/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 473ms/step - accuracy: 0.5662 - loss: 0.6846 - val_accuracy: 0.5527 - val_loss: 0.6921\nEpoch 13/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 479ms/step - accuracy: 0.5423 - loss: 0.6872 - val_accuracy: 0.5527 - val_loss: 0.6982\nEpoch 14/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 477ms/step - accuracy: 0.5571 - loss: 0.6762 - val_accuracy: 0.5527 - val_loss: 0.6913\nEpoch 15/15\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 482ms/step - accuracy: 0.5715 - loss: 0.6806 - val_accuracy: 0.5527 - val_loss: 0.6914\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"
","image/png":"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\n"},"metadata":{}}],"execution_count":13},{"cell_type":"code","source":"# Transfer Learning\n\nresnet = ResNet50(\n weights='imagenet',\n include_top=False,\n input_shape=(*IMG_SIZE,3)\n)\nresnet.trainable = False\n\nx = resnet.output\nx = layers.GlobalAveragePooling2D()(x)\nx = layers.BatchNormalization()(x)\nx = layers.Dense(256, activation='relu')(x)\nx = layers.Dropout(0.5)(x)\noutput = layers.Dense(1, activation='sigmoid')(x)\n\nresnet_model = models.Model(resnet.input, output)\n\nresnet_model.compile(\n optimizer=optimizers.Adam(1e-4),\n loss='binary_crossentropy',\n metrics=['accuracy']\n)\n\nresnet_model.summary()\n\nresnet_history = resnet_model.fit(\n train_gen,\n validation_data=test_gen,\n epochs=10\n)\n\nplot_history(resnet_history)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-02T17:59:14.639563Z","iopub.execute_input":"2026-01-02T17:59:14.640132Z","iopub.status.idle":"2026-01-02T18:03:46.716877Z","shell.execute_reply.started":"2026-01-02T17:59:14.640103Z","shell.execute_reply":"2026-01-02T18:03:46.716220Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"\u001b[1mModel: \"functional_3\"\u001b[0m\n","text/html":"
Model: \"functional_3\"\n
\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"┏━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┓\n┃\u001b[1m \u001b[0m\u001b[1mLayer (type) \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m Param #\u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mConnected to \u001b[0m\u001b[1m \u001b[0m┃\n┡━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━┩\n│ input_layer_4 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m224\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ - │\n│ (\u001b[38;5;33mInputLayer\u001b[0m) │ \u001b[38;5;34m3\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv1_pad │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m230\u001b[0m, \u001b[38;5;34m230\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ input_layer_4[\u001b[38;5;34m0\u001b[0m]… │\n│ (\u001b[38;5;33mZeroPadding2D\u001b[0m) │ \u001b[38;5;34m3\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv1_conv (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m112\u001b[0m, │ \u001b[38;5;34m9,472\u001b[0m │ conv1_pad[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n│ │ \u001b[38;5;34m64\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv1_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m112\u001b[0m, │ \u001b[38;5;34m256\u001b[0m │ conv1_conv[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv1_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m112\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv1_bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ pool1_pad │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m114\u001b[0m, \u001b[38;5;34m114\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv1_relu[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n│ (\u001b[38;5;33mZeroPadding2D\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ pool1_pool │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ pool1_pad[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n│ (\u001b[38;5;33mMaxPooling2D\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_1_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m4,160\u001b[0m │ pool1_pool[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_1_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m256\u001b[0m │ conv2_block1_1_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_1_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv2_block1_1_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_2_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m36,928\u001b[0m │ conv2_block1_1_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_2_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m256\u001b[0m │ conv2_block1_2_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_2_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv2_block1_2_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_0_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m16,640\u001b[0m │ pool1_pool[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_3_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m16,640\u001b[0m │ conv2_block1_2_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_0_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m1,024\u001b[0m │ conv2_block1_0_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_3_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m1,024\u001b[0m │ conv2_block1_3_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_add │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv2_block1_0_b… │\n│ (\u001b[38;5;33mAdd\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ conv2_block1_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_out │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv2_block1_add… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block2_1_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m16,448\u001b[0m │ conv2_block1_out… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block2_1_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m256\u001b[0m │ conv2_block2_1_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block2_1_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv2_block2_1_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block2_2_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m36,928\u001b[0m │ conv2_block2_1_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block2_2_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m256\u001b[0m │ conv2_block2_2_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block2_2_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv2_block2_2_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block2_3_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m16,640\u001b[0m │ conv2_block2_2_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block2_3_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m1,024\u001b[0m │ conv2_block2_3_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block2_add │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv2_block1_out… │\n│ (\u001b[38;5;33mAdd\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ conv2_block2_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block2_out │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv2_block2_add… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block3_1_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m16,448\u001b[0m │ conv2_block2_out… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block3_1_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m256\u001b[0m │ conv2_block3_1_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block3_1_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv2_block3_1_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block3_2_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m36,928\u001b[0m │ conv2_block3_1_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block3_2_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m256\u001b[0m │ conv2_block3_2_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m64\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block3_2_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv2_block3_2_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m64\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block3_3_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m16,640\u001b[0m │ conv2_block3_2_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block3_3_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m1,024\u001b[0m │ conv2_block3_3_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block3_add │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv2_block2_out… │\n│ (\u001b[38;5;33mAdd\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ conv2_block3_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block3_out │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv2_block3_add… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_1_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m32,896\u001b[0m │ conv2_block3_out… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_1_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m512\u001b[0m │ conv3_block1_1_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_1_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv3_block1_1_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_2_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m147,584\u001b[0m │ conv3_block1_1_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_2_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m512\u001b[0m │ conv3_block1_2_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_2_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv3_block1_2_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_0_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m131,584\u001b[0m │ conv2_block3_out… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m512\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_3_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m66,048\u001b[0m │ conv3_block1_2_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m512\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_0_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m2,048\u001b[0m │ conv3_block1_0_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m512\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_3_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m2,048\u001b[0m │ conv3_block1_3_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m512\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_add │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv3_block1_0_b… │\n│ (\u001b[38;5;33mAdd\u001b[0m) │ \u001b[38;5;34m512\u001b[0m) │ │ conv3_block1_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_out │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv3_block1_add… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m512\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block2_1_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m65,664\u001b[0m │ conv3_block1_out… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block2_1_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m512\u001b[0m │ conv3_block2_1_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block2_1_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv3_block2_1_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block2_2_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m147,584\u001b[0m │ conv3_block2_1_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block2_2_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m512\u001b[0m │ conv3_block2_2_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block2_2_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv3_block2_2_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block2_3_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m66,048\u001b[0m │ conv3_block2_2_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m512\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block2_3_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m2,048\u001b[0m │ conv3_block2_3_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m512\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block2_add │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv3_block1_out… │\n│ (\u001b[38;5;33mAdd\u001b[0m) │ \u001b[38;5;34m512\u001b[0m) │ │ conv3_block2_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block2_out │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv3_block2_add… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m512\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block3_1_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m65,664\u001b[0m │ conv3_block2_out… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block3_1_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m512\u001b[0m │ conv3_block3_1_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block3_1_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv3_block3_1_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block3_2_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m147,584\u001b[0m │ conv3_block3_1_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block3_2_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m512\u001b[0m │ conv3_block3_2_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block3_2_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv3_block3_2_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block3_3_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m66,048\u001b[0m │ conv3_block3_2_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m512\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block3_3_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m2,048\u001b[0m │ conv3_block3_3_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m512\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block3_add │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv3_block2_out… │\n│ (\u001b[38;5;33mAdd\u001b[0m) │ \u001b[38;5;34m512\u001b[0m) │ │ conv3_block3_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block3_out │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv3_block3_add… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m512\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block4_1_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m65,664\u001b[0m │ conv3_block3_out… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block4_1_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m512\u001b[0m │ conv3_block4_1_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block4_1_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv3_block4_1_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block4_2_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m147,584\u001b[0m │ conv3_block4_1_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block4_2_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m512\u001b[0m │ conv3_block4_2_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block4_2_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv3_block4_2_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m128\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block4_3_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m66,048\u001b[0m │ conv3_block4_2_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m512\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block4_3_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m2,048\u001b[0m │ conv3_block4_3_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m512\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block4_add │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv3_block3_out… │\n│ (\u001b[38;5;33mAdd\u001b[0m) │ \u001b[38;5;34m512\u001b[0m) │ │ conv3_block4_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block4_out │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv3_block4_add… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m512\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_1_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m131,328\u001b[0m │ conv3_block4_out… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_1_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m1,024\u001b[0m │ conv4_block1_1_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_1_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block1_1_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_2_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m590,080\u001b[0m │ conv4_block1_1_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_2_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m1,024\u001b[0m │ conv4_block1_2_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_2_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block1_2_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_0_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m525,312\u001b[0m │ conv3_block4_out… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m1024\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_3_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m263,168\u001b[0m │ conv4_block1_2_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m1024\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_0_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m4,096\u001b[0m │ conv4_block1_0_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m1024\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_3_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m4,096\u001b[0m │ conv4_block1_3_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m1024\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_add │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block1_0_b… │\n│ (\u001b[38;5;33mAdd\u001b[0m) │ \u001b[38;5;34m1024\u001b[0m) │ │ conv4_block1_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_out │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block1_add… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m1024\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block2_1_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m262,400\u001b[0m │ conv4_block1_out… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block2_1_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m1,024\u001b[0m │ conv4_block2_1_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block2_1_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block2_1_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block2_2_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m590,080\u001b[0m │ conv4_block2_1_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block2_2_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m1,024\u001b[0m │ conv4_block2_2_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block2_2_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block2_2_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block2_3_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m263,168\u001b[0m │ conv4_block2_2_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m1024\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block2_3_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m4,096\u001b[0m │ conv4_block2_3_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m1024\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block2_add │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block1_out… │\n│ (\u001b[38;5;33mAdd\u001b[0m) │ \u001b[38;5;34m1024\u001b[0m) │ │ conv4_block2_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block2_out │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block2_add… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m1024\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block3_1_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m262,400\u001b[0m │ conv4_block2_out… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block3_1_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m1,024\u001b[0m │ conv4_block3_1_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block3_1_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block3_1_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block3_2_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m590,080\u001b[0m │ conv4_block3_1_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block3_2_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m1,024\u001b[0m │ conv4_block3_2_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block3_2_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block3_2_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block3_3_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m263,168\u001b[0m │ conv4_block3_2_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m1024\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block3_3_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m4,096\u001b[0m │ conv4_block3_3_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m1024\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block3_add │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block2_out… │\n│ (\u001b[38;5;33mAdd\u001b[0m) │ \u001b[38;5;34m1024\u001b[0m) │ │ conv4_block3_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block3_out │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block3_add… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m1024\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block4_1_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m262,400\u001b[0m │ conv4_block3_out… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block4_1_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m1,024\u001b[0m │ conv4_block4_1_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block4_1_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block4_1_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block4_2_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m590,080\u001b[0m │ conv4_block4_1_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block4_2_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m1,024\u001b[0m │ conv4_block4_2_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block4_2_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block4_2_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block4_3_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m263,168\u001b[0m │ conv4_block4_2_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m1024\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block4_3_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m4,096\u001b[0m │ conv4_block4_3_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m1024\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block4_add │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block3_out… │\n│ (\u001b[38;5;33mAdd\u001b[0m) │ \u001b[38;5;34m1024\u001b[0m) │ │ conv4_block4_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block4_out │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block4_add… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m1024\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block5_1_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m262,400\u001b[0m │ conv4_block4_out… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block5_1_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m1,024\u001b[0m │ conv4_block5_1_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block5_1_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block5_1_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block5_2_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m590,080\u001b[0m │ conv4_block5_1_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block5_2_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m1,024\u001b[0m │ conv4_block5_2_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block5_2_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block5_2_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block5_3_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m263,168\u001b[0m │ conv4_block5_2_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m1024\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block5_3_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m4,096\u001b[0m │ conv4_block5_3_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m1024\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block5_add │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block4_out… │\n│ (\u001b[38;5;33mAdd\u001b[0m) │ \u001b[38;5;34m1024\u001b[0m) │ │ conv4_block5_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block5_out │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block5_add… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m1024\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block6_1_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m262,400\u001b[0m │ conv4_block5_out… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block6_1_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m1,024\u001b[0m │ conv4_block6_1_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block6_1_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block6_1_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block6_2_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m590,080\u001b[0m │ conv4_block6_1_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block6_2_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m1,024\u001b[0m │ conv4_block6_2_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block6_2_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block6_2_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m256\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block6_3_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m263,168\u001b[0m │ conv4_block6_2_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m1024\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block6_3_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m4,096\u001b[0m │ conv4_block6_3_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m1024\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block6_add │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block5_out… │\n│ (\u001b[38;5;33mAdd\u001b[0m) │ \u001b[38;5;34m1024\u001b[0m) │ │ conv4_block6_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block6_out │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv4_block6_add… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m1024\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_1_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m524,800\u001b[0m │ conv4_block6_out… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_1_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m2,048\u001b[0m │ conv5_block1_1_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_1_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ conv5_block1_1_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_2_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m2,359,808\u001b[0m │ conv5_block1_1_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_2_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m2,048\u001b[0m │ conv5_block1_2_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_2_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ conv5_block1_2_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_0_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, │ \u001b[38;5;34m2,099,200\u001b[0m │ conv4_block6_out… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m2048\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_3_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, │ \u001b[38;5;34m1,050,624\u001b[0m │ conv5_block1_2_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m2048\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_0_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, │ \u001b[38;5;34m8,192\u001b[0m │ conv5_block1_0_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m2048\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_3_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, │ \u001b[38;5;34m8,192\u001b[0m │ conv5_block1_3_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m2048\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_add │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv5_block1_0_b… │\n│ (\u001b[38;5;33mAdd\u001b[0m) │ \u001b[38;5;34m2048\u001b[0m) │ │ conv5_block1_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_out │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv5_block1_add… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m2048\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block2_1_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m1,049,088\u001b[0m │ conv5_block1_out… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block2_1_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m2,048\u001b[0m │ conv5_block2_1_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block2_1_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ conv5_block2_1_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block2_2_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m2,359,808\u001b[0m │ conv5_block2_1_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block2_2_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m2,048\u001b[0m │ conv5_block2_2_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block2_2_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ conv5_block2_2_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block2_3_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, │ \u001b[38;5;34m1,050,624\u001b[0m │ conv5_block2_2_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m2048\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block2_3_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, │ \u001b[38;5;34m8,192\u001b[0m │ conv5_block2_3_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m2048\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block2_add │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv5_block1_out… │\n│ (\u001b[38;5;33mAdd\u001b[0m) │ \u001b[38;5;34m2048\u001b[0m) │ │ conv5_block2_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block2_out │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv5_block2_add… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m2048\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block3_1_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m1,049,088\u001b[0m │ conv5_block2_out… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block3_1_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m2,048\u001b[0m │ conv5_block3_1_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block3_1_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ conv5_block3_1_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block3_2_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m2,359,808\u001b[0m │ conv5_block3_1_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block3_2_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m2,048\u001b[0m │ conv5_block3_2_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block3_2_relu │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ conv5_block3_2_b… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block3_3_conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, │ \u001b[38;5;34m1,050,624\u001b[0m │ conv5_block3_2_r… │\n│ (\u001b[38;5;33mConv2D\u001b[0m) │ \u001b[38;5;34m2048\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block3_3_bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, │ \u001b[38;5;34m8,192\u001b[0m │ conv5_block3_3_c… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ \u001b[38;5;34m2048\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block3_add │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv5_block2_out… │\n│ (\u001b[38;5;33mAdd\u001b[0m) │ \u001b[38;5;34m2048\u001b[0m) │ │ conv5_block3_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block3_out │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv5_block3_add… │\n│ (\u001b[38;5;33mActivation\u001b[0m) │ \u001b[38;5;34m2048\u001b[0m) │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ global_average_poo… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2048\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ conv5_block3_out… │\n│ (\u001b[38;5;33mGlobalAveragePool…\u001b[0m │ │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ batch_normalizatio… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2048\u001b[0m) │ \u001b[38;5;34m8,192\u001b[0m │ global_average_p… │\n│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ dense_8 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m) │ \u001b[38;5;34m524,544\u001b[0m │ batch_normalizat… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ dropout_3 (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ dense_8[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ dense_9 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m) │ \u001b[38;5;34m257\u001b[0m │ dropout_3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n└─────────────────────┴───────────────────┴────────────┴───────────────────┘\n","text/html":"
┏━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┓\n┃ Layer (type)        ┃ Output Shape      ┃    Param # ┃ Connected to      ┃\n┡━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━┩\n│ input_layer_4       │ (None, 224, 224,  │          0 │ -                 │\n│ (InputLayer)        │ 3)                │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv1_pad           │ (None, 230, 230,  │          0 │ input_layer_4[0]… │\n│ (ZeroPadding2D)     │ 3)                │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv1_conv (Conv2D) │ (None, 112, 112,  │      9,472 │ conv1_pad[0][0]   │\n│                     │ 64)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv1_bn            │ (None, 112, 112,  │        256 │ conv1_conv[0][0]  │\n│ (BatchNormalizatio… │ 64)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv1_relu          │ (None, 112, 112,  │          0 │ conv1_bn[0][0]    │\n│ (Activation)        │ 64)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ pool1_pad           │ (None, 114, 114,  │          0 │ conv1_relu[0][0]  │\n│ (ZeroPadding2D)     │ 64)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ pool1_pool          │ (None, 56, 56,    │          0 │ pool1_pad[0][0]   │\n│ (MaxPooling2D)      │ 64)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_1_conv │ (None, 56, 56,    │      4,160 │ pool1_pool[0][0]  │\n│ (Conv2D)            │ 64)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_1_bn   │ (None, 56, 56,    │        256 │ conv2_block1_1_c… │\n│ (BatchNormalizatio… │ 64)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_1_relu │ (None, 56, 56,    │          0 │ conv2_block1_1_b… │\n│ (Activation)        │ 64)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_2_conv │ (None, 56, 56,    │     36,928 │ conv2_block1_1_r… │\n│ (Conv2D)            │ 64)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_2_bn   │ (None, 56, 56,    │        256 │ conv2_block1_2_c… │\n│ (BatchNormalizatio… │ 64)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_2_relu │ (None, 56, 56,    │          0 │ conv2_block1_2_b… │\n│ (Activation)        │ 64)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_0_conv │ (None, 56, 56,    │     16,640 │ pool1_pool[0][0]  │\n│ (Conv2D)            │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_3_conv │ (None, 56, 56,    │     16,640 │ conv2_block1_2_r… │\n│ (Conv2D)            │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_0_bn   │ (None, 56, 56,    │      1,024 │ conv2_block1_0_c… │\n│ (BatchNormalizatio… │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_3_bn   │ (None, 56, 56,    │      1,024 │ conv2_block1_3_c… │\n│ (BatchNormalizatio… │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_add    │ (None, 56, 56,    │          0 │ conv2_block1_0_b… │\n│ (Add)               │ 256)              │            │ conv2_block1_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block1_out    │ (None, 56, 56,    │          0 │ conv2_block1_add… │\n│ (Activation)        │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block2_1_conv │ (None, 56, 56,    │     16,448 │ conv2_block1_out… │\n│ (Conv2D)            │ 64)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block2_1_bn   │ (None, 56, 56,    │        256 │ conv2_block2_1_c… │\n│ (BatchNormalizatio… │ 64)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block2_1_relu │ (None, 56, 56,    │          0 │ conv2_block2_1_b… │\n│ (Activation)        │ 64)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block2_2_conv │ (None, 56, 56,    │     36,928 │ conv2_block2_1_r… │\n│ (Conv2D)            │ 64)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block2_2_bn   │ (None, 56, 56,    │        256 │ conv2_block2_2_c… │\n│ (BatchNormalizatio… │ 64)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block2_2_relu │ (None, 56, 56,    │          0 │ conv2_block2_2_b… │\n│ (Activation)        │ 64)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block2_3_conv │ (None, 56, 56,    │     16,640 │ conv2_block2_2_r… │\n│ (Conv2D)            │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block2_3_bn   │ (None, 56, 56,    │      1,024 │ conv2_block2_3_c… │\n│ (BatchNormalizatio… │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block2_add    │ (None, 56, 56,    │          0 │ conv2_block1_out… │\n│ (Add)               │ 256)              │            │ conv2_block2_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block2_out    │ (None, 56, 56,    │          0 │ conv2_block2_add… │\n│ (Activation)        │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block3_1_conv │ (None, 56, 56,    │     16,448 │ conv2_block2_out… │\n│ (Conv2D)            │ 64)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block3_1_bn   │ (None, 56, 56,    │        256 │ conv2_block3_1_c… │\n│ (BatchNormalizatio… │ 64)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block3_1_relu │ (None, 56, 56,    │          0 │ conv2_block3_1_b… │\n│ (Activation)        │ 64)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block3_2_conv │ (None, 56, 56,    │     36,928 │ conv2_block3_1_r… │\n│ (Conv2D)            │ 64)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block3_2_bn   │ (None, 56, 56,    │        256 │ conv2_block3_2_c… │\n│ (BatchNormalizatio… │ 64)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block3_2_relu │ (None, 56, 56,    │          0 │ conv2_block3_2_b… │\n│ (Activation)        │ 64)               │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block3_3_conv │ (None, 56, 56,    │     16,640 │ conv2_block3_2_r… │\n│ (Conv2D)            │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block3_3_bn   │ (None, 56, 56,    │      1,024 │ conv2_block3_3_c… │\n│ (BatchNormalizatio… │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block3_add    │ (None, 56, 56,    │          0 │ conv2_block2_out… │\n│ (Add)               │ 256)              │            │ conv2_block3_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv2_block3_out    │ (None, 56, 56,    │          0 │ conv2_block3_add… │\n│ (Activation)        │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_1_conv │ (None, 28, 28,    │     32,896 │ conv2_block3_out… │\n│ (Conv2D)            │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_1_bn   │ (None, 28, 28,    │        512 │ conv3_block1_1_c… │\n│ (BatchNormalizatio… │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_1_relu │ (None, 28, 28,    │          0 │ conv3_block1_1_b… │\n│ (Activation)        │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_2_conv │ (None, 28, 28,    │    147,584 │ conv3_block1_1_r… │\n│ (Conv2D)            │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_2_bn   │ (None, 28, 28,    │        512 │ conv3_block1_2_c… │\n│ (BatchNormalizatio… │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_2_relu │ (None, 28, 28,    │          0 │ conv3_block1_2_b… │\n│ (Activation)        │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_0_conv │ (None, 28, 28,    │    131,584 │ conv2_block3_out… │\n│ (Conv2D)            │ 512)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_3_conv │ (None, 28, 28,    │     66,048 │ conv3_block1_2_r… │\n│ (Conv2D)            │ 512)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_0_bn   │ (None, 28, 28,    │      2,048 │ conv3_block1_0_c… │\n│ (BatchNormalizatio… │ 512)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_3_bn   │ (None, 28, 28,    │      2,048 │ conv3_block1_3_c… │\n│ (BatchNormalizatio… │ 512)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_add    │ (None, 28, 28,    │          0 │ conv3_block1_0_b… │\n│ (Add)               │ 512)              │            │ conv3_block1_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block1_out    │ (None, 28, 28,    │          0 │ conv3_block1_add… │\n│ (Activation)        │ 512)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block2_1_conv │ (None, 28, 28,    │     65,664 │ conv3_block1_out… │\n│ (Conv2D)            │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block2_1_bn   │ (None, 28, 28,    │        512 │ conv3_block2_1_c… │\n│ (BatchNormalizatio… │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block2_1_relu │ (None, 28, 28,    │          0 │ conv3_block2_1_b… │\n│ (Activation)        │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block2_2_conv │ (None, 28, 28,    │    147,584 │ conv3_block2_1_r… │\n│ (Conv2D)            │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block2_2_bn   │ (None, 28, 28,    │        512 │ conv3_block2_2_c… │\n│ (BatchNormalizatio… │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block2_2_relu │ (None, 28, 28,    │          0 │ conv3_block2_2_b… │\n│ (Activation)        │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block2_3_conv │ (None, 28, 28,    │     66,048 │ conv3_block2_2_r… │\n│ (Conv2D)            │ 512)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block2_3_bn   │ (None, 28, 28,    │      2,048 │ conv3_block2_3_c… │\n│ (BatchNormalizatio… │ 512)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block2_add    │ (None, 28, 28,    │          0 │ conv3_block1_out… │\n│ (Add)               │ 512)              │            │ conv3_block2_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block2_out    │ (None, 28, 28,    │          0 │ conv3_block2_add… │\n│ (Activation)        │ 512)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block3_1_conv │ (None, 28, 28,    │     65,664 │ conv3_block2_out… │\n│ (Conv2D)            │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block3_1_bn   │ (None, 28, 28,    │        512 │ conv3_block3_1_c… │\n│ (BatchNormalizatio… │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block3_1_relu │ (None, 28, 28,    │          0 │ conv3_block3_1_b… │\n│ (Activation)        │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block3_2_conv │ (None, 28, 28,    │    147,584 │ conv3_block3_1_r… │\n│ (Conv2D)            │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block3_2_bn   │ (None, 28, 28,    │        512 │ conv3_block3_2_c… │\n│ (BatchNormalizatio… │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block3_2_relu │ (None, 28, 28,    │          0 │ conv3_block3_2_b… │\n│ (Activation)        │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block3_3_conv │ (None, 28, 28,    │     66,048 │ conv3_block3_2_r… │\n│ (Conv2D)            │ 512)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block3_3_bn   │ (None, 28, 28,    │      2,048 │ conv3_block3_3_c… │\n│ (BatchNormalizatio… │ 512)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block3_add    │ (None, 28, 28,    │          0 │ conv3_block2_out… │\n│ (Add)               │ 512)              │            │ conv3_block3_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block3_out    │ (None, 28, 28,    │          0 │ conv3_block3_add… │\n│ (Activation)        │ 512)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block4_1_conv │ (None, 28, 28,    │     65,664 │ conv3_block3_out… │\n│ (Conv2D)            │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block4_1_bn   │ (None, 28, 28,    │        512 │ conv3_block4_1_c… │\n│ (BatchNormalizatio… │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block4_1_relu │ (None, 28, 28,    │          0 │ conv3_block4_1_b… │\n│ (Activation)        │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block4_2_conv │ (None, 28, 28,    │    147,584 │ conv3_block4_1_r… │\n│ (Conv2D)            │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block4_2_bn   │ (None, 28, 28,    │        512 │ conv3_block4_2_c… │\n│ (BatchNormalizatio… │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block4_2_relu │ (None, 28, 28,    │          0 │ conv3_block4_2_b… │\n│ (Activation)        │ 128)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block4_3_conv │ (None, 28, 28,    │     66,048 │ conv3_block4_2_r… │\n│ (Conv2D)            │ 512)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block4_3_bn   │ (None, 28, 28,    │      2,048 │ conv3_block4_3_c… │\n│ (BatchNormalizatio… │ 512)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block4_add    │ (None, 28, 28,    │          0 │ conv3_block3_out… │\n│ (Add)               │ 512)              │            │ conv3_block4_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv3_block4_out    │ (None, 28, 28,    │          0 │ conv3_block4_add… │\n│ (Activation)        │ 512)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_1_conv │ (None, 14, 14,    │    131,328 │ conv3_block4_out… │\n│ (Conv2D)            │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_1_bn   │ (None, 14, 14,    │      1,024 │ conv4_block1_1_c… │\n│ (BatchNormalizatio… │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_1_relu │ (None, 14, 14,    │          0 │ conv4_block1_1_b… │\n│ (Activation)        │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_2_conv │ (None, 14, 14,    │    590,080 │ conv4_block1_1_r… │\n│ (Conv2D)            │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_2_bn   │ (None, 14, 14,    │      1,024 │ conv4_block1_2_c… │\n│ (BatchNormalizatio… │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_2_relu │ (None, 14, 14,    │          0 │ conv4_block1_2_b… │\n│ (Activation)        │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_0_conv │ (None, 14, 14,    │    525,312 │ conv3_block4_out… │\n│ (Conv2D)            │ 1024)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_3_conv │ (None, 14, 14,    │    263,168 │ conv4_block1_2_r… │\n│ (Conv2D)            │ 1024)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_0_bn   │ (None, 14, 14,    │      4,096 │ conv4_block1_0_c… │\n│ (BatchNormalizatio… │ 1024)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_3_bn   │ (None, 14, 14,    │      4,096 │ conv4_block1_3_c… │\n│ (BatchNormalizatio… │ 1024)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_add    │ (None, 14, 14,    │          0 │ conv4_block1_0_b… │\n│ (Add)               │ 1024)             │            │ conv4_block1_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block1_out    │ (None, 14, 14,    │          0 │ conv4_block1_add… │\n│ (Activation)        │ 1024)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block2_1_conv │ (None, 14, 14,    │    262,400 │ conv4_block1_out… │\n│ (Conv2D)            │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block2_1_bn   │ (None, 14, 14,    │      1,024 │ conv4_block2_1_c… │\n│ (BatchNormalizatio… │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block2_1_relu │ (None, 14, 14,    │          0 │ conv4_block2_1_b… │\n│ (Activation)        │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block2_2_conv │ (None, 14, 14,    │    590,080 │ conv4_block2_1_r… │\n│ (Conv2D)            │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block2_2_bn   │ (None, 14, 14,    │      1,024 │ conv4_block2_2_c… │\n│ (BatchNormalizatio… │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block2_2_relu │ (None, 14, 14,    │          0 │ conv4_block2_2_b… │\n│ (Activation)        │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block2_3_conv │ (None, 14, 14,    │    263,168 │ conv4_block2_2_r… │\n│ (Conv2D)            │ 1024)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block2_3_bn   │ (None, 14, 14,    │      4,096 │ conv4_block2_3_c… │\n│ (BatchNormalizatio… │ 1024)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block2_add    │ (None, 14, 14,    │          0 │ conv4_block1_out… │\n│ (Add)               │ 1024)             │            │ conv4_block2_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block2_out    │ (None, 14, 14,    │          0 │ conv4_block2_add… │\n│ (Activation)        │ 1024)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block3_1_conv │ (None, 14, 14,    │    262,400 │ conv4_block2_out… │\n│ (Conv2D)            │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block3_1_bn   │ (None, 14, 14,    │      1,024 │ conv4_block3_1_c… │\n│ (BatchNormalizatio… │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block3_1_relu │ (None, 14, 14,    │          0 │ conv4_block3_1_b… │\n│ (Activation)        │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block3_2_conv │ (None, 14, 14,    │    590,080 │ conv4_block3_1_r… │\n│ (Conv2D)            │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block3_2_bn   │ (None, 14, 14,    │      1,024 │ conv4_block3_2_c… │\n│ (BatchNormalizatio… │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block3_2_relu │ (None, 14, 14,    │          0 │ conv4_block3_2_b… │\n│ (Activation)        │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block3_3_conv │ (None, 14, 14,    │    263,168 │ conv4_block3_2_r… │\n│ (Conv2D)            │ 1024)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block3_3_bn   │ (None, 14, 14,    │      4,096 │ conv4_block3_3_c… │\n│ (BatchNormalizatio… │ 1024)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block3_add    │ (None, 14, 14,    │          0 │ conv4_block2_out… │\n│ (Add)               │ 1024)             │            │ conv4_block3_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block3_out    │ (None, 14, 14,    │          0 │ conv4_block3_add… │\n│ (Activation)        │ 1024)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block4_1_conv │ (None, 14, 14,    │    262,400 │ conv4_block3_out… │\n│ (Conv2D)            │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block4_1_bn   │ (None, 14, 14,    │      1,024 │ conv4_block4_1_c… │\n│ (BatchNormalizatio… │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block4_1_relu │ (None, 14, 14,    │          0 │ conv4_block4_1_b… │\n│ (Activation)        │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block4_2_conv │ (None, 14, 14,    │    590,080 │ conv4_block4_1_r… │\n│ (Conv2D)            │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block4_2_bn   │ (None, 14, 14,    │      1,024 │ conv4_block4_2_c… │\n│ (BatchNormalizatio… │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block4_2_relu │ (None, 14, 14,    │          0 │ conv4_block4_2_b… │\n│ (Activation)        │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block4_3_conv │ (None, 14, 14,    │    263,168 │ conv4_block4_2_r… │\n│ (Conv2D)            │ 1024)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block4_3_bn   │ (None, 14, 14,    │      4,096 │ conv4_block4_3_c… │\n│ (BatchNormalizatio… │ 1024)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block4_add    │ (None, 14, 14,    │          0 │ conv4_block3_out… │\n│ (Add)               │ 1024)             │            │ conv4_block4_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block4_out    │ (None, 14, 14,    │          0 │ conv4_block4_add… │\n│ (Activation)        │ 1024)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block5_1_conv │ (None, 14, 14,    │    262,400 │ conv4_block4_out… │\n│ (Conv2D)            │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block5_1_bn   │ (None, 14, 14,    │      1,024 │ conv4_block5_1_c… │\n│ (BatchNormalizatio… │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block5_1_relu │ (None, 14, 14,    │          0 │ conv4_block5_1_b… │\n│ (Activation)        │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block5_2_conv │ (None, 14, 14,    │    590,080 │ conv4_block5_1_r… │\n│ (Conv2D)            │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block5_2_bn   │ (None, 14, 14,    │      1,024 │ conv4_block5_2_c… │\n│ (BatchNormalizatio… │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block5_2_relu │ (None, 14, 14,    │          0 │ conv4_block5_2_b… │\n│ (Activation)        │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block5_3_conv │ (None, 14, 14,    │    263,168 │ conv4_block5_2_r… │\n│ (Conv2D)            │ 1024)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block5_3_bn   │ (None, 14, 14,    │      4,096 │ conv4_block5_3_c… │\n│ (BatchNormalizatio… │ 1024)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block5_add    │ (None, 14, 14,    │          0 │ conv4_block4_out… │\n│ (Add)               │ 1024)             │            │ conv4_block5_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block5_out    │ (None, 14, 14,    │          0 │ conv4_block5_add… │\n│ (Activation)        │ 1024)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block6_1_conv │ (None, 14, 14,    │    262,400 │ conv4_block5_out… │\n│ (Conv2D)            │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block6_1_bn   │ (None, 14, 14,    │      1,024 │ conv4_block6_1_c… │\n│ (BatchNormalizatio… │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block6_1_relu │ (None, 14, 14,    │          0 │ conv4_block6_1_b… │\n│ (Activation)        │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block6_2_conv │ (None, 14, 14,    │    590,080 │ conv4_block6_1_r… │\n│ (Conv2D)            │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block6_2_bn   │ (None, 14, 14,    │      1,024 │ conv4_block6_2_c… │\n│ (BatchNormalizatio… │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block6_2_relu │ (None, 14, 14,    │          0 │ conv4_block6_2_b… │\n│ (Activation)        │ 256)              │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block6_3_conv │ (None, 14, 14,    │    263,168 │ conv4_block6_2_r… │\n│ (Conv2D)            │ 1024)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block6_3_bn   │ (None, 14, 14,    │      4,096 │ conv4_block6_3_c… │\n│ (BatchNormalizatio… │ 1024)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block6_add    │ (None, 14, 14,    │          0 │ conv4_block5_out… │\n│ (Add)               │ 1024)             │            │ conv4_block6_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv4_block6_out    │ (None, 14, 14,    │          0 │ conv4_block6_add… │\n│ (Activation)        │ 1024)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_1_conv │ (None, 7, 7, 512) │    524,800 │ conv4_block6_out… │\n│ (Conv2D)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_1_bn   │ (None, 7, 7, 512) │      2,048 │ conv5_block1_1_c… │\n│ (BatchNormalizatio… │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_1_relu │ (None, 7, 7, 512) │          0 │ conv5_block1_1_b… │\n│ (Activation)        │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_2_conv │ (None, 7, 7, 512) │  2,359,808 │ conv5_block1_1_r… │\n│ (Conv2D)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_2_bn   │ (None, 7, 7, 512) │      2,048 │ conv5_block1_2_c… │\n│ (BatchNormalizatio… │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_2_relu │ (None, 7, 7, 512) │          0 │ conv5_block1_2_b… │\n│ (Activation)        │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_0_conv │ (None, 7, 7,      │  2,099,200 │ conv4_block6_out… │\n│ (Conv2D)            │ 2048)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_3_conv │ (None, 7, 7,      │  1,050,624 │ conv5_block1_2_r… │\n│ (Conv2D)            │ 2048)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_0_bn   │ (None, 7, 7,      │      8,192 │ conv5_block1_0_c… │\n│ (BatchNormalizatio… │ 2048)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_3_bn   │ (None, 7, 7,      │      8,192 │ conv5_block1_3_c… │\n│ (BatchNormalizatio… │ 2048)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_add    │ (None, 7, 7,      │          0 │ conv5_block1_0_b… │\n│ (Add)               │ 2048)             │            │ conv5_block1_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block1_out    │ (None, 7, 7,      │          0 │ conv5_block1_add… │\n│ (Activation)        │ 2048)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block2_1_conv │ (None, 7, 7, 512) │  1,049,088 │ conv5_block1_out… │\n│ (Conv2D)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block2_1_bn   │ (None, 7, 7, 512) │      2,048 │ conv5_block2_1_c… │\n│ (BatchNormalizatio… │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block2_1_relu │ (None, 7, 7, 512) │          0 │ conv5_block2_1_b… │\n│ (Activation)        │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block2_2_conv │ (None, 7, 7, 512) │  2,359,808 │ conv5_block2_1_r… │\n│ (Conv2D)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block2_2_bn   │ (None, 7, 7, 512) │      2,048 │ conv5_block2_2_c… │\n│ (BatchNormalizatio… │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block2_2_relu │ (None, 7, 7, 512) │          0 │ conv5_block2_2_b… │\n│ (Activation)        │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block2_3_conv │ (None, 7, 7,      │  1,050,624 │ conv5_block2_2_r… │\n│ (Conv2D)            │ 2048)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block2_3_bn   │ (None, 7, 7,      │      8,192 │ conv5_block2_3_c… │\n│ (BatchNormalizatio… │ 2048)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block2_add    │ (None, 7, 7,      │          0 │ conv5_block1_out… │\n│ (Add)               │ 2048)             │            │ conv5_block2_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block2_out    │ (None, 7, 7,      │          0 │ conv5_block2_add… │\n│ (Activation)        │ 2048)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block3_1_conv │ (None, 7, 7, 512) │  1,049,088 │ conv5_block2_out… │\n│ (Conv2D)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block3_1_bn   │ (None, 7, 7, 512) │      2,048 │ conv5_block3_1_c… │\n│ (BatchNormalizatio… │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block3_1_relu │ (None, 7, 7, 512) │          0 │ conv5_block3_1_b… │\n│ (Activation)        │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block3_2_conv │ (None, 7, 7, 512) │  2,359,808 │ conv5_block3_1_r… │\n│ (Conv2D)            │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block3_2_bn   │ (None, 7, 7, 512) │      2,048 │ conv5_block3_2_c… │\n│ (BatchNormalizatio… │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block3_2_relu │ (None, 7, 7, 512) │          0 │ conv5_block3_2_b… │\n│ (Activation)        │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block3_3_conv │ (None, 7, 7,      │  1,050,624 │ conv5_block3_2_r… │\n│ (Conv2D)            │ 2048)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block3_3_bn   │ (None, 7, 7,      │      8,192 │ conv5_block3_3_c… │\n│ (BatchNormalizatio… │ 2048)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block3_add    │ (None, 7, 7,      │          0 │ conv5_block2_out… │\n│ (Add)               │ 2048)             │            │ conv5_block3_3_b… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ conv5_block3_out    │ (None, 7, 7,      │          0 │ conv5_block3_add… │\n│ (Activation)        │ 2048)             │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ global_average_poo… │ (None, 2048)      │          0 │ conv5_block3_out… │\n│ (GlobalAveragePool… │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ batch_normalizatio… │ (None, 2048)      │      8,192 │ global_average_p… │\n│ (BatchNormalizatio… │                   │            │                   │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ dense_8 (Dense)     │ (None, 256)       │    524,544 │ batch_normalizat… │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ dropout_3 (Dropout) │ (None, 256)       │          0 │ dense_8[0][0]     │\n├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n│ dense_9 (Dense)     │ (None, 1)         │        257 │ dropout_3[0][0]   │\n└─────────────────────┴───────────────────┴────────────┴───────────────────┘\n
\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"\u001b[1m Total params: \u001b[0m\u001b[38;5;34m24,120,705\u001b[0m (92.01 MB)\n","text/html":"
 Total params: 24,120,705 (92.01 MB)\n
\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m528,897\u001b[0m (2.02 MB)\n","text/html":"
 Trainable params: 528,897 (2.02 MB)\n
\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m23,591,808\u001b[0m (90.00 MB)\n","text/html":"
 Non-trainable params: 23,591,808 (90.00 MB)\n
\n"},"metadata":{}},{"name":"stdout","text":"Epoch 1/10\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m44s\u001b[0m 735ms/step - accuracy: 0.5072 - loss: 0.7229 - val_accuracy: 0.4473 - val_loss: 0.7059\nEpoch 2/10\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 494ms/step - accuracy: 0.5598 - loss: 0.6889 - val_accuracy: 0.4701 - val_loss: 0.6954\nEpoch 3/10\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 483ms/step - accuracy: 0.5376 - loss: 0.7013 - val_accuracy: 0.5043 - val_loss: 0.6941\nEpoch 4/10\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m46s\u001b[0m 1s/step - accuracy: 0.5763 - loss: 0.6894 - val_accuracy: 0.5527 - val_loss: 0.6887\nEpoch 5/10\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m26s\u001b[0m 583ms/step - accuracy: 0.5558 - loss: 0.7014 - val_accuracy: 0.5527 - val_loss: 0.6886\nEpoch 6/10\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 489ms/step - accuracy: 0.5772 - loss: 0.6743 - val_accuracy: 0.5271 - val_loss: 0.6899\nEpoch 7/10\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 483ms/step - accuracy: 0.5839 - loss: 0.6791 - val_accuracy: 0.5499 - val_loss: 0.6887\nEpoch 8/10\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 484ms/step - accuracy: 0.5837 - loss: 0.6850 - val_accuracy: 0.5328 - val_loss: 0.6906\nEpoch 9/10\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 476ms/step - accuracy: 0.5633 - loss: 0.6865 - val_accuracy: 0.5442 - val_loss: 0.6917\nEpoch 10/10\n\u001b[1m45/45\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m22s\u001b[0m 492ms/step - accuracy: 0.5768 - loss: 0.6850 - val_accuracy: 0.5071 - val_loss: 0.6976\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"
","image/png":"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\n"},"metadata":{}}],"execution_count":15},{"cell_type":"code","source":"y_true = test_gen.classes\ny_prob = resnet_model.predict(test_gen).ravel()\ny_pred = (y_prob > 0.5).astype(int)\n\nprint(classification_report(y_true, y_pred))\nprint(\"F1 Score:\", f1_score(y_true, y_pred))\nprint(\"ROC AUC:\", roc_auc_score(y_true, y_prob))\n\nfpr, tpr, _ = roc_curve(y_true, y_prob)\n\nplt.figure(figsize=(6,5))\nplt.plot(fpr, tpr)\nplt.plot([0,1], [0,1], linestyle='--')\nplt.title(\"ROC Curve\")\nplt.xlabel(\"False Positive Rate\")\nplt.ylabel(\"True Positive Rate\")\nplt.show()\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-02T18:04:36.267192Z","iopub.execute_input":"2026-01-02T18:04:36.267830Z","iopub.status.idle":"2026-01-02T18:04:45.336687Z","shell.execute_reply.started":"2026-01-02T18:04:36.267802Z","shell.execute_reply":"2026-01-02T18:04:45.336109Z"}},"outputs":[{"name":"stdout","text":"\u001b[1m11/11\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 490ms/step\n precision recall f1-score support\n\n 0 0.38 0.15 0.22 157\n 1 0.54 0.79 0.64 194\n\n accuracy 0.51 351\n macro avg 0.46 0.47 0.43 351\nweighted avg 0.46 0.51 0.45 351\n\nF1 Score: 0.6403326403326404\nROC AUC: 0.48483157134414606\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"
","image/png":"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\n"},"metadata":{}}],"execution_count":16},{"cell_type":"markdown","source":"## **Observations**\n\nThe **ResNet model shows weak overall performance**. Although class 1 has **high recall (0.79)** and a decent **F1 score (~0.64)**, this comes at the cost of **poor class 0 detection**, indicating a strong **class bias**. The **overall accuracy (~51%)** and **ROC-AUC (<0.5)** suggest the model is **close to random guessing** and fails to separate classes reliably.\n\nCompared to this, the **baseline/regularized CNN achieves slightly better and more balanced validation accuracy (~55%)**, indicating that **transfer learning did not provide a clear advantage** for this candlestick dataset, possibly due to domain mismatch or limited fine-tuning.\n","metadata":{}},{"cell_type":"markdown","source":"## Error Analysis","metadata":{}},{"cell_type":"code","source":"# Error Analysis\n\nerrors = np.where(y_pred != y_true)[0]\n\nplt.figure(figsize=(12,6))\nfor i, idx in enumerate(errors[:8]):\n plt.subplot(2,4,i+1)\n img_path = test_gen.filepaths[idx]\n img = plt.imread(img_path)\n plt.imshow(img)\n plt.axis('on')\n plt.title(f\"True:{y_true[idx]} Pred:{y_pred[idx]}\")\n\nplt.tight_layout()\nplt.show()\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-01-02T18:05:10.274153Z","iopub.execute_input":"2026-01-02T18:05:10.274664Z","iopub.status.idle":"2026-01-02T18:05:11.281202Z","shell.execute_reply.started":"2026-01-02T18:05:10.274638Z","shell.execute_reply":"2026-01-02T18:05:11.280562Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"
","image/png":"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\n"},"metadata":{}}],"execution_count":18},{"cell_type":"markdown","source":"**Short summary of work done:**\n\n* **Data transformations:**\n Candlestick images were resized and normalized, with light augmentation (small rotations, zoom, shifts, brightness changes) to improve generalization while preserving financial meaning; flips were avoided to prevent distorting time or price direction.\n\n* **Models used:**\n\n 1. A **basic CNN** trained from scratch as a baseline,\n 2. A **regularized CNN** with Batch Normalization and Dropout to reduce overfitting,\n 3. A **transfer-learning model (ResNet50)** used as a frozen feature extractor with a custom classification head.\n\n* **Results:**\n The baseline and regularized CNNs achieved around **55% validation accuracy**.\n The ResNet50 model showed **high recall for one class but poor overall discrimination**, with ~**51% accuracy** and **ROC-AUC < 0.5**, indicating performance close to random and no clear advantage from transfer learning.\n","metadata":{}},{"cell_type":"code","source":"","metadata":{"trusted":true},"outputs":[],"execution_count":null}]} \ No newline at end of file