"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"fig, ax = plt.subplots(nrows=2, ncols=5, figsize=(10,6))\n",
"\n",
"offset = 10\n",
"for i in range(5):\n",
" # train\n",
" ax[0,i].imshow(train_images[i+offset].reshape(img_rows,img_cols), cmap=plt.cm.Blues)\n",
" ax[0,i].set_xlabel(ytrain[i], fontsize=10)\n",
" ax[0,i].set_xticks([]); ax[0,i].set_yticks([]); ax[0,i].grid(False)\n",
" # val\n",
" ax[1,i].imshow(test_images[i+offset].reshape(img_rows,img_cols), cmap=plt.cm.Blues)\n",
" ax[1,i].set_xlabel(ytest[i], fontsize=10)\n",
" ax[1,i].set_xticks([]); ax[1,i].set_yticks([]); ax[1,i].grid(False)\n",
" # test\n",
" # ax[2,i].imshow(test_images[i].reshape(28,28), cmap=plt.cm.binary)\n",
" # ax[2,i].set_xlabel(np.argmax(test_labels[i]), fontsize=18)\n",
" # ax[2,i].set_xticks([]); ax[2,i].set_yticks([]); ax[2,i].grid(False)\n",
"\n",
"plt.tight_layout()\n",
"plt.show()"
]
},
{
"cell_type": "code",
"execution_count": 43,
"id": "dabf0efc",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"Model: \"functional\"\n",
"\n"
],
"text/plain": [
"\u001b[1mModel: \"functional\"\u001b[0m\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Layer (type) ┃ Output Shape ┃ Param # ┃ Connected to ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━┩\n",
"│ image_input │ (None, 27, 27, 1) │ 0 │ - │\n",
"│ (InputLayer) │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ conv2d (Conv2D) │ (None, 25, 25, │ 320 │ image_input[0][0] │\n",
"│ │ 32) │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ max_pooling2d │ (None, 12, 12, │ 0 │ conv2d[0][0] │\n",
"│ (MaxPooling2D) │ 32) │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ conv2d_1 (Conv2D) │ (None, 10, 10, │ 18,496 │ max_pooling2d[0]… │\n",
"│ │ 64) │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ max_pooling2d_1 │ (None, 5, 5, 64) │ 0 │ conv2d_1[0][0] │\n",
"│ (MaxPooling2D) │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ conv2d_2 (Conv2D) │ (None, 3, 3, 128) │ 73,856 │ max_pooling2d_1[… │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ max_pooling2d_2 │ (None, 1, 1, 128) │ 0 │ conv2d_2[0][0] │\n",
"│ (MaxPooling2D) │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ flatten (Flatten) │ (None, 128) │ 0 │ max_pooling2d_2[… │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ scalar_input │ (None, 7) │ 0 │ - │\n",
"│ (InputLayer) │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ concatenate │ (None, 135) │ 0 │ flatten[0][0], │\n",
"│ (Concatenate) │ │ │ scalar_input[0][… │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dense (Dense) │ (None, 256) │ 34,816 │ concatenate[0][0] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dropout (Dropout) │ (None, 256) │ 0 │ dense[0][0] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dense_1 (Dense) │ (None, 128) │ 32,896 │ dropout[0][0] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dropout_1 (Dropout) │ (None, 128) │ 0 │ dense_1[0][0] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ output (Dense) │ (None, 1) │ 129 │ dropout_1[0][0] │\n",
"└─────────────────────┴───────────────────┴────────────┴───────────────────┘\n",
"\n"
],
"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",
"│ image_input │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m27\u001b[0m, \u001b[38;5;34m27\u001b[0m, \u001b[38;5;34m1\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ - │\n",
"│ (\u001b[38;5;33mInputLayer\u001b[0m) │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ conv2d (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m25\u001b[0m, \u001b[38;5;34m25\u001b[0m, │ \u001b[38;5;34m320\u001b[0m │ image_input[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"│ │ \u001b[38;5;34m32\u001b[0m) │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ max_pooling2d │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m12\u001b[0m, \u001b[38;5;34m12\u001b[0m, │ \u001b[38;5;34m0\u001b[0m │ conv2d[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"│ (\u001b[38;5;33mMaxPooling2D\u001b[0m) │ \u001b[38;5;34m32\u001b[0m) │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ conv2d_1 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m10\u001b[0m, \u001b[38;5;34m10\u001b[0m, │ \u001b[38;5;34m18,496\u001b[0m │ max_pooling2d[\u001b[38;5;34m0\u001b[0m]… │\n",
"│ │ \u001b[38;5;34m64\u001b[0m) │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ max_pooling2d_1 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m5\u001b[0m, \u001b[38;5;34m5\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ conv2d_1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"│ (\u001b[38;5;33mMaxPooling2D\u001b[0m) │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ conv2d_2 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m3\u001b[0m, \u001b[38;5;34m3\u001b[0m, \u001b[38;5;34m128\u001b[0m) │ \u001b[38;5;34m73,856\u001b[0m │ max_pooling2d_1[\u001b[38;5;34m…\u001b[0m │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ max_pooling2d_2 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m1\u001b[0m, \u001b[38;5;34m128\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ conv2d_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"│ (\u001b[38;5;33mMaxPooling2D\u001b[0m) │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ flatten (\u001b[38;5;33mFlatten\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ max_pooling2d_2[\u001b[38;5;34m…\u001b[0m │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ scalar_input │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ - │\n",
"│ (\u001b[38;5;33mInputLayer\u001b[0m) │ │ │ │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ concatenate │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m135\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ flatten[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m], │\n",
"│ (\u001b[38;5;33mConcatenate\u001b[0m) │ │ │ scalar_input[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m…\u001b[0m │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dense (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m) │ \u001b[38;5;34m34,816\u001b[0m │ concatenate[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dropout (\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[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dense_1 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m) │ \u001b[38;5;34m32,896\u001b[0m │ dropout[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ dropout_1 (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ dense_1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
"│ output (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m) │ \u001b[38;5;34m129\u001b[0m │ dropout_1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n",
"└─────────────────────┴───────────────────┴────────────┴───────────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
" Total params: 160,513 (627.00 KB)\n",
"\n"
],
"text/plain": [
"\u001b[1m Total params: \u001b[0m\u001b[38;5;34m160,513\u001b[0m (627.00 KB)\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
" Trainable params: 160,513 (627.00 KB)\n",
"\n"
],
"text/plain": [
"\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m160,513\u001b[0m (627.00 KB)\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
" Non-trainable params: 0 (0.00 B)\n",
"\n"
],
"text/plain": [
"\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# 1. Define input shapes\n",
"IMAGE_HEIGHT = 27\n",
"IMAGE_WIDTH = 27\n",
"IMAGE_CHANNELS = 1\n",
"NUM_SCALAR_FEATURES = 7 # Example: 10 scalar features\n",
"\n",
"# 2. Define the CNN Branch\n",
"input_image = layers.Input(shape=(IMAGE_HEIGHT, IMAGE_WIDTH, IMAGE_CHANNELS), name='image_input')\n",
"\n",
"x = layers.Conv2D(32, (3, 3), activation='relu')(input_image)\n",
"x = layers.MaxPooling2D((2, 2))(x)\n",
"x = layers.Conv2D(64, (3, 3), activation='relu')(x)\n",
"x = layers.MaxPooling2D((2, 2))(x)\n",
"x = layers.Conv2D(128, (3, 3), activation='relu')(x)\n",
"x = layers.MaxPooling2D((2, 2))(x)\n",
"x = layers.Flatten()(x) # Flatten the output of the CNN for dense layers\n",
"\n",
"# 3. Define the Scalar Feature Branch\n",
"input_scalar = layers.Input(shape=(NUM_SCALAR_FEATURES,), name='scalar_input')\n",
"\n",
"# You can optionally add dense layers here for scalar features if needed\n",
"# y = layers.Dense(32, activation='relu')(input_scalar)\n",
"y = input_scalar # For simplicity, we'll use the scalar features directly\n",
"\n",
"# 4. Concatenate the Branches\n",
"concatenated = layers.concatenate([x, y])\n",
"\n",
"# 5. Define the Perceptron (Dense) Layers\n",
"z = layers.Dense(256, activation='relu')(concatenated)\n",
"z = layers.Dropout(0.5)(z) # Optional: Add dropout for regularization\n",
"z = layers.Dense(128, activation='relu')(z)\n",
"z = layers.Dropout(0.5)(z)\n",
"\n",
"# Output layer (e.g., for binary classification)\n",
"output = layers.Dense(1, name='output')(z)\n",
"\n",
"# 6. Create the Model\n",
"model = keras.Model(inputs=[input_image, input_scalar], outputs=output)\n",
"\n",
"# Compile the model\n",
"model.compile(optimizer='adam',\n",
" loss='mse',\n",
" metrics=['mape'])\n",
"\n",
"# Print model summary to see the architecture\n",
"model.summary()"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "dde2eab6",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 1/3\n",
"\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m19s\u001b[0m 58ms/step - loss: 79.4667 - mape: 79.4667 - val_loss: 70.0264 - val_mape: 70.0264\n",
"Epoch 2/3\n",
"\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m13s\u001b[0m 50ms/step - loss: 79.2389 - mape: 79.2389 - val_loss: 66.0984 - val_mape: 66.0984\n",
"Epoch 3/3\n",
"\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m13s\u001b[0m 50ms/step - loss: 70.0583 - mape: 70.0583 - val_loss: 61.4946 - val_mape: 61.4946\n"
]
}
],
"source": [
"# specify optimization strategy and metric used for monitoring during training\n",
"model.compile(loss='mse',\n",
" optimizer='adam',\n",
" metrics=['mape'])\n",
"\n",
"# the history object will contain a record of loss and metric values during training\n",
"history = model.fit({'image_input': train_images, 'scalar_input': train_feats}, ytrain,\n",
" batch_size=256,\n",
" epochs=3,\n",
" validation_split = 0.2)"
]
},
{
"cell_type": "code",
"execution_count": 25,
"id": "99879fe3",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"1st convolution layer:\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"2nd convolution layer:\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# model.layers will print a list of layer parameters/values\n",
"filters1, biases1 = model.layers[0].get_weights()\n",
"filters2, biases2 = model.layers[2].get_weights()\n",
"\n",
"# normalize filter values to range 0-1 for better colormapping during plotting\n",
"def norm_filter(kernel):\n",
" return (kernel - np.min(kernel)) / (np.max(kernel) - np.min(kernel))\n",
"\n",
"print('1st convolution layer:')\n",
"fig, axs = plt.subplots(2,5, figsize=(10, 6))\n",
"axs = axs.ravel()\n",
"for i in range(10):\n",
" axs[i].imshow(norm_filter(filters1[:,:,0,i]), cmap=plt.cm.binary)\n",
" axs[i].set_xticks([]); axs[i].set_yticks([]); axs[i].grid(False)\n",
"plt.show()\n",
"\n",
"print('2nd convolution layer:')\n",
"fig, axs = plt.subplots(2,5, figsize=(10, 6))\n",
"axs = axs.ravel()\n",
"for i in range(10):\n",
" axs[i].imshow(norm_filter(filters2[:,:,0,i]), cmap=plt.cm.binary)\n",
" axs[i].set_xticks([]); axs[i].set_yticks([]); axs[i].grid(False)"
]
},
{
"cell_type": "code",
"execution_count": 48,
"id": "2db64cc2",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"624/624 - 6s - 9ms/step - loss: 64.5244 - mape: 64.5244\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# evaluating model using all data (not in batches)\n",
"val_loss, val_acc = model.evaluate({'image_input': test_images, 'scalar_input': test_feats}, ytest, verbose=2)\n",
"\n",
"fig,ax = plt.subplots(nrows=2,ncols=1,figsize=(12,12))\n",
"fs_L, fs_M, fs_S = 18, 16, 14\n",
"ax[0].plot(history.history['mape'], label='train')\n",
"ax[0].plot(history.history['val_mape'], label='validation')\n",
"ax[0].set_xlabel('Epoch', fontsize=fs_M)\n",
"ax[0].set_ylabel('Mean Absolute Percentage Error', fontsize=fs_M)\n",
"ax[0].tick_params(axis='both', which='major', labelsize=fs_S)\n",
"ax[0].set_title('Final mean validation MAPE: {}'.format(val_acc), fontsize=fs_M)\n",
"ax[0].set_xticks(range(0,5))\n",
"ax[0].legend(loc='lower right', fontsize=fs_M)\n",
"\n",
"ax[1].plot(history.history['loss'], label='train')\n",
"ax[1].plot(history.history['val_loss'], label='validation')\n",
"ax[1].set_xlabel('Epoch', fontsize=fs_M)\n",
"ax[1].set_ylabel('Loss', fontsize=fs_M)\n",
"ax[1].tick_params(axis='both', which='major', labelsize=fs_S)\n",
"ax[1].set_xticks(range(0,5))\n",
"ax[1].legend(loc='upper right', fontsize=fs_M)\n",
"\n",
"plt.show()"
]
}
],
"metadata": {
"kernelspec": {
"display_name": ".appMLvenv",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.11.3"
}
},
"nbformat": 4,
"nbformat_minor": 5
}