snapshot..

modules/deeplearning/srcnn_l1b_l2.py

@@ -62,30 +62,30 @@ label_idx = params.index(label_param)
 print('data_params: ', data_params)
 print('label_param: ', label_param)
 
-# Kernel size: 3, target size: (128, 128)
-slc_x = slice(2, 132)
-slc_y = slice(2, 132)
-slc_x_2 = slice(1, 134, 2)
-slc_y_2 = slice(1, 134, 2)
-x_128 = slice(3, 131)
-y_128 = slice(3, 131)
-t = np.arange(1, 66, 0.5)
-s = np.arange(1, 66, 0.5)
-x_2 = np.arange(67)
-y_2 = np.arange(67)
-# ----------------------------------------
-
-# Kernel size: 5, target_size: (128, 128)
-# slc_x = slice(3, 135)
-# slc_y = slice(3, 135)
-# slc_x_2 = slice(2, 137, 2)
-# slc_y_2 = slice(2, 137, 2)
-# x_128 = slice(5, 133)
-# y_128 = slice(5, 133)
-# t = np.arange(1, 67, 0.5)
-# s = np.arange(1, 67, 0.5)
-# x_2 = np.arange(68)
-# y_2 = np.arange(68)
+KERNEL_SIZE = 3  # target size: (128, 128)
+
+if KERNEL_SIZE == 3:
+    slc_x = slice(2, 132)
+    slc_y = slice(2, 132)
+    slc_x_2 = slice(1, 134, 2)
+    slc_y_2 = slice(1, 134, 2)
+    x_128 = slice(3, 131)
+    y_128 = slice(3, 131)
+    t = np.arange(1, 66, 0.5)
+    s = np.arange(1, 66, 0.5)
+    x_2 = np.arange(67)
+    y_2 = np.arange(67)
+elif KERNEL_SIZE == 5:
+    slc_x = slice(3, 135)
+    slc_y = slice(3, 135)
+    slc_x_2 = slice(2, 137, 2)
+    slc_y_2 = slice(2, 137, 2)
+    x_128 = slice(5, 133)
+    y_128 = slice(5, 133)
+    t = np.arange(1, 67, 0.5)
+    s = np.arange(1, 67, 0.5)
+    x_2 = np.arange(68)
+    y_2 = np.arange(68)
 # ----------------------------------------
 
 
@@ -412,7 +412,7 @@ class SRCNN:
         input_2d = self.inputs[0]
         print('input: ', input_2d.shape)
 
-        conv = conv_b = tf.keras.layers.Conv2D(num_filters, kernel_size=3, kernel_initializer='he_uniform', activation=activation, padding='VALID')(input_2d)
+        conv = conv_b = tf.keras.layers.Conv2D(num_filters, kernel_size=KERNEL_SIZE, kernel_initializer='he_uniform', activation=activation, padding='VALID')(input_2d)
         print(conv.shape)
 
         # if NOISE_TRAINING:
@@ -420,15 +420,15 @@ class SRCNN:
 
         scale = 0.2
 
-        conv_b = build_residual_conv2d_block(conv_b, num_filters, 'Residual_Block_1', kernel_size=3, scale=scale)
+        conv_b = build_residual_conv2d_block(conv_b, num_filters, 'Residual_Block_1', kernel_size=KERNEL_SIZE, scale=scale)
 
-        conv_b = build_residual_conv2d_block(conv_b, num_filters, 'Residual_Block_2', kernel_size=3, scale=scale)
+        conv_b = build_residual_conv2d_block(conv_b, num_filters, 'Residual_Block_2', kernel_size=KERNEL_SIZE, scale=scale)
 
-        conv_b = build_residual_conv2d_block(conv_b, num_filters, 'Residual_Block_3', kernel_size=3, scale=scale)
+        conv_b = build_residual_conv2d_block(conv_b, num_filters, 'Residual_Block_3', kernel_size=KERNEL_SIZE, scale=scale)
 
-        #conv_b = build_residual_conv2d_block(conv_b, num_filters, 'Residual_Block_4', kernel_size=3, scale=scale)
+        #conv_b = build_residual_conv2d_block(conv_b, num_filters, 'Residual_Block_4', kernel_size=KERNEL_SIZE, scale=scale)
 
-        #conv_b = build_residual_conv2d_block(conv_b, num_filters, 'Residual_Block_5', kernel_size=3, scale=scale)
+        #conv_b = build_residual_conv2d_block(conv_b, num_filters, 'Residual_Block_5', kernel_size=KERNEL_SIZE, scale=scale)
 
         conv_b = tf.keras.layers.Conv2D(num_filters, kernel_size=3, strides=1, activation=activation, kernel_initializer='he_uniform', padding=padding)(conv_b)