snapshot...

modules/deeplearning/espcn_l1b_l2.py

@@ -72,6 +72,8 @@ y_134_2 = y_134[2:133:2]
 
 slc_x = slice(3, 131)
 slc_y = slice(3, 131)
+slc_x_2 = slice(3, 131, 2)
+slc_y_2 = slice(3, 131, 2)
 
 
 def build_residual_conv2d_block(conv, num_filters, block_name, activation=tf.nn.leaky_relu, padding='SAME', scale=None):
@@ -214,17 +216,19 @@ class ESPCN:
 
         data_norm = []
         for k, param in enumerate(data_params):
-            tmp = input_data[:, k, :, :]
+            # tmp = input_data[:, k, :, :]
+            tmp = input_data[:, k, slc_y_2, slc_x_2]
             tmp = normalize(tmp, param, mean_std_dct, add_noise=add_noise, noise_scale=noise_scale)
-            tmp = resample_2d_linear(x_134, y_134, tmp, x_134_2, y_134_2)
+            # tmp = resample_2d_linear(x_134, y_134, tmp, x_134_2, y_134_2)
             data_norm.append(tmp)
 
-        tmp = input_data[:, label_idx, :, ]
+        # tmp = input_data[:, label_idx, :, ]
+        tmp = input_data[:, label_idx, slc_y_2, slc_x_2]
         if label_param != 'cloud_fraction':
             tmp = normalize(tmp, label_param, mean_std_dct, add_noise=add_noise, noise_scale=noise_scale)
         else:
             tmp = np.where(np.isnan(tmp), 0, tmp)
-        tmp = resample_2d_linear(x_134, y_134, tmp, x_134_2, y_134_2)
+        # tmp = resample_2d_linear(x_134, y_134, tmp, x_134_2, y_134_2)
         data_norm.append(tmp)
 
         data = np.stack(data_norm, axis=3)
@@ -336,7 +340,6 @@ class ESPCN:
         print('num test samples: ', tst_idxs.shape[0])
         print('setup_pipeline: Done')
 
-
     def setup_test_pipeline(self, test_data_files):
         self.test_data_files = test_data_files
         tst_idxs = np.arange(len(test_data_files))
@@ -389,7 +392,7 @@ class ESPCN:
         conv = conv_b
         print(conv.shape)
 
-        conv = tf.keras.layers.Conv2D(IMG_DEPTH * (factor ** 2), 3, padding='same')(conv)
+        conv = tf.keras.layers.Conv2D(IMG_DEPTH * (factor ** 2), 3, padding='same', activation=activation)(conv)
         print(conv.shape)
 
         conv = tf.nn.depth_to_space(conv, factor)