snapshot...

modules/deeplearning/srcnn_l1b_l2.py

@@ -198,15 +198,12 @@ class SRCNN:
             files = self.test_data_files
 
         data_s = []
-        label_s = []
         for k in idxs:
             f = files[k]
             nda = np.load(f)
-            data_s.append(nda[0:len(data_params), :, :])
-            label_s.append(nda[3, :, :])
+            data_s.append(nda)
 
         data = np.concatenate(data_s)
-        label = np.concatenate(label_s)
 
         add_noise = None
         noise_scale = None
@@ -215,7 +212,7 @@ class SRCNN:
             noise_scale = 0.005
 
         data_norm = []
-        for k, param in enumerate(data_params):
+        for k, param in enumerate(params):
             tmp = data[:, k, 3:131:2, 3:131:2]
             tmp = resample(y_64, x_64, tmp, s, t)
             tmp = normalize(tmp, param, mean_std_dct, add_noise=add_noise, noise_scale=noise_scale)
@@ -223,8 +220,8 @@ class SRCNN:
         data = np.stack(data_norm, axis=3)
         data = data.astype(np.float32)
 
-        # label = label[:, 3:131:2, 3:131:2]
-        label = label[:, 3:131, 3:131]
+        # label = data[:, label_idx, 3:131:2, 3:131:2]
+        label = data[:, label_idx, 3:131, 3:131]
         label = np.expand_dims(label, axis=3)
         if label_param != 'cloud_fraction':
             label = normalize(label, label_param, mean_std_dct)