Spaces:
Sleeping
Sleeping
| #https://youtu.be/csFGTLT6_WQ | |
| # u-net model | |
| import os | |
| import segmentation_models as sm | |
| from keras.layers import ( | |
| BatchNormalization, | |
| Conv2D, | |
| Conv2DTranspose, | |
| Dropout, | |
| Input, | |
| MaxPooling2D, | |
| concatenate, | |
| ) | |
| from keras.models import Model | |
| os.environ["SM_FRAMEWORK"] = "tf.keras" | |
| os.environ["_TF_KERAS_FRAMEWORK_NAME"] = "tf.keras" | |
| sm.set_framework('tf.keras') | |
| ################################################################ | |
| def SpecSeg(IMG_HEIGHT, IMG_WIDTH, IMG_CHANNELS): | |
| #Build the model | |
| inputs = Input((IMG_HEIGHT, IMG_WIDTH, IMG_CHANNELS)) | |
| s = inputs | |
| #Contraction path | |
| c1 = Conv2D(16, (3, 3), activation='relu', kernel_initializer='RandomNormal', padding='same')(s) | |
| c1 = Dropout(0.1)(c1) | |
| c1 = Conv2D(16, (3, 3), activation='relu', kernel_initializer='RandomNormal', padding='same')(c1) | |
| c1 = BatchNormalization(axis=-1)(c1) | |
| p1 = MaxPooling2D((2, 2))(c1) | |
| c2 = Conv2D(32, (3, 3), activation='relu', kernel_initializer='RandomNormal', padding='same')(p1) | |
| c2 = Dropout(0.1)(c2) | |
| c2 = Conv2D(32, (3, 3), activation='relu', kernel_initializer='RandomNormal', padding='same')(c2) | |
| c2 = BatchNormalization(axis=-1)(c2) | |
| p2 = MaxPooling2D((2, 2))(c2) | |
| c3 = Conv2D(64, (3, 3), activation='relu', kernel_initializer='RandomNormal', padding='same')(p2) | |
| c3 = Dropout(0.2)(c3) | |
| c3 = Conv2D(64, (3, 3), activation='relu', kernel_initializer='RandomNormal', padding='same')(c3) | |
| c3 = BatchNormalization(axis=-1)(c3) | |
| p3 = MaxPooling2D((2, 2))(c3) | |
| c4 = Conv2D(128, (3, 3), activation='relu', kernel_initializer='RandomNormal', padding='same')(p3) | |
| c4 = Dropout(0.2)(c4) | |
| c4 = Conv2D(128, (3, 3), activation='relu', kernel_initializer='RandomNormal', padding='same')(c4) | |
| c4 = BatchNormalization(axis=-1)(c4) | |
| p4 = MaxPooling2D((2, 2))(c4) | |
| c5 = Conv2D(256, (3, 3), activation='relu', kernel_initializer='RandomNormal', padding='same')(p4) | |
| c5 = Dropout(0.3)(c5) | |
| c5 = Conv2D(256, (3, 3), activation='relu', kernel_initializer='RandomNormal', padding='same')(c5) | |
| c5 = BatchNormalization(axis=-1)(c5) | |
| #Expansive path | |
| u6 = Conv2DTranspose(128, (2, 2), strides=(2, 2), padding='same')(c5) | |
| u6 = concatenate([u6, c4]) | |
| c6 = Conv2D(128, (3, 3), activation='relu', kernel_initializer='RandomNormal', padding='same')(u6) | |
| c6 = Dropout(0.2)(c6) | |
| c6 = Conv2D(128, (3, 3), activation='relu', kernel_initializer='RandomNormal', padding='same')(c6) | |
| u7 = Conv2DTranspose(64, (2, 2), strides=(2, 2), padding='same')(c6) | |
| u7 = concatenate([u7, c3]) | |
| c7 = Conv2D(64, (3, 3), activation='relu', kernel_initializer='RandomNormal', padding='same')(u7) | |
| c7 = Dropout(0.2)(c7) | |
| c7 = Conv2D(64, (3, 3), activation='relu', kernel_initializer='RandomNormal', padding='same')(c7) | |
| u8 = Conv2DTranspose(32, (2, 2), strides=(2, 2), padding='same')(c7) | |
| u8 = concatenate([u8, c2]) | |
| c8 = Conv2D(32, (3, 3), activation='relu', kernel_initializer='RandomNormal', padding='same')(u8) | |
| c8 = Dropout(0.1)(c8) | |
| c8 = Conv2D(32, (3, 3), activation='relu', kernel_initializer='RandomNormal', padding='same')(c8) | |
| u9 = Conv2DTranspose(16, (2, 2), strides=(2, 2), padding='same')(c8) | |
| u9 = concatenate([u9, c1], axis=3) | |
| c9 = Conv2D(16, (3, 3), activation='relu', kernel_initializer='RandomNormal', padding='same')(u9) | |
| c9 = Dropout(0.1)(c9) | |
| c9 = Conv2D(16, (3, 3), activation='relu', kernel_initializer='RandomNormal', padding='same')(c9) | |
| outputs = Conv2D(1, (1, 1), activation='sigmoid')(c9) | |
| model = Model(inputs=[inputs], outputs=[outputs], name = 'SpecSeg') | |
| # model.summary() | |
| return model | |