Pseudocode ResNet50 |
Input: Chest Radiographs Output: classification results: Covid Or Normal Start lr ← 1 × 10−4 ▷ lr is Initial_Learning_rate Batch_Size ← 32 Number_of_Epochs ← 28 Base_Model ← ResNet50(weights ← “imagenet”, include_top ← False, input_tensor ← Input (shape ← (224, 224, 3))) ▷ ResNet50 is the base Model headModel ← baseModel.output headModel ← AveragePooling2D(pool_size ← (7, 7))(headModel) headModel ← Flatten(name ← “flatten”)(headModel) headModel ← Dense(256, activation ← “relu”)(headModel) headModel ← Dropout(0.5)(headModel) headModel ← Dense(len(CLASSES), activation ← “softmax”)(headModel) model ← Model(inputs ← baseModel.input, outputs ← headModel) for layer in baseModel.layers: layer.trainable ← True end for opt ← optimizers.Adam (lr ← INIT_LR, decay ← INIT_LR/Number_of_Epochs) model.compile (loss ← “binary_crossentropy”, optimizer ← opt, metrics ← [“Accuracy”]) H ← model.fit_generator (trainGen, steps_per_epoch ← totalTrain, validation_data ← valGen, validation_steps ← totalVal, epochs ← Number_of_Epochs) End |