use meshgrid

modules/deeplearning/icing_cnn.py

@@ -988,12 +988,13 @@ def run_evaluate_static(h5f, ckpt_dir_s_path, prob_thresh=0.5, satellite='GOES16
         if not os.path.isdir(ckpt_dir):
             continue
         nn = IcingIntensityNN()
-        nn.setup_eval_pipeline(data_dct, len(ll))
+        nn.setup_eval_pipeline(data_dct, len(ll) * len(cc))
         nn.build_model()
         nn.build_training()
         nn.build_evaluation()
         nn.do_evaluate(ckpt_dir)
         prob_s.append(nn.test_probs)
+
     num = len(prob_s)
     prob_avg = prob_s[0]
     for k in range(num-1):
@@ -1006,12 +1007,12 @@ def run_evaluate_static(h5f, ckpt_dir_s_path, prob_thresh=0.5, satellite='GOES16
     else:
         preds = np.argmax(probs, axis=1)
 
-    ll_grd, cc_grd = np.meshgrid(ll, cc, indexing='ij')
-    cc_grd = cc_grd.flatten()
-    ll_grd = ll_grd.flatten()
+    ll, cc = np.meshgrid(ll, cc, indexing='ij')
+    cc = cc.flatten()
+    ll = ll.flatten()
     ice_mask = preds == 1
-    ice_cc = cc_grd[ice_mask]
-    ice_ll = ll_grd[ice_mask]
+    ice_cc = cc[ice_mask]
+    ice_ll = ll[ice_mask]
 
     nav = get_navigation(satellite, domain)
     ice_lons = []