Upload input_preprocess.py

input_preprocess.py

@@ -49,10 +49,6 @@ def tokenizer(selected_scenario_names=None, manual_data=None, gen_raw=True):
         
         cleaned_deepmimo_data = [deepmimo_data_cleaning(deepmimo_data[scenario_idx]) for scenario_idx in range(n_scenarios)]
         
-        print(len(cleaned_deepmimo_data))
-        print(len(cleaned_deepmimo_data[0]))
-        print(len(cleaned_deepmimo_data[0][0]))
-        
         patches = [patch_maker(cleaned_deepmimo_data[scenario_idx]) for scenario_idx in range(n_scenarios)]
         patches = np.vstack(patches)
 
@@ -77,7 +73,7 @@ def tokenizer(selected_scenario_names=None, manual_data=None, gen_raw=True):
 def deepmimo_data_cleaning(deepmimo_data):
     idxs = np.where(deepmimo_data['user']['LoS'] != -1)[0]
     cleaned_deepmimo_data = deepmimo_data['user']['channel'][idxs]
-    return cleaned_deepmimo_data
+    return np.array(cleaned_deepmimo_data) * 1e6
 
 #%% Patch Creation
 def patch_maker(original_ch, patch_size=16, norm_factor=1e6):
@@ -98,7 +94,7 @@ def patch_maker(original_ch, patch_size=16, norm_factor=1e6):
     # # Reshaping and normalizing channels
     # original_ch = data['user']['channel'][idxs]
     flat_channels = original_ch.reshape((original_ch.shape[0], -1)).astype(np.csingle)
-    flat_channels_complex = np.hstack((flat_channels.real, flat_channels.imag)) * norm_factor
+    flat_channels_complex = np.hstack((flat_channels.real, flat_channels.imag))
     
     # Create patches
     n_patches = flat_channels_complex.shape[1] // patch_size