datasets

common_dataset.py

@@ -165,18 +165,17 @@ def load_data(path_train, path_val, path_test, batch_size, length, pad=False, co
     return train_loader, val_loader, test_loader
 
 if __name__ =='__main__' :
-    pass
-    # irt_train = np.load('data/intensity/irt_train.npy')
-    # seq_train = np.load('data/intensity/sequence_train.npy')
-    # charge_train = np.load('data/intensity/precursor_charge_train.npy')
-    # spectra_train = np.load('data/intensity/intensity_train.npy')
-    #
-    # irt_holdout = np.load('data/intensity/irt_holdout.npy')
-    # seq_holdout = np.load('data/intensity/sequence_holdout.npy')
-    # charge_holdout = np.load('data/intensity/precursor_charge_holdout.npy')
-    # spectra_holdout = np.load('data/intensity/intensity_holdout.npy')
-    #
-    # dataset_train = pd.DataFrame({'Sequence':list(seq_train), 'Retention time':list(irt_train), 'Charge':list(charge_train), 'Spectra' : list(spectra_train)},index=list(range(6787933)))
-    # dataset_train.to_pickle('database/data_prosit_merged_train.pkl')
-    # dataset_test = pd.DataFrame({'Sequence':list(seq_holdout), 'Retention time':list(irt_holdout), 'Charge':list(charge_holdout), 'Spectra' : list(spectra_holdout)},index=list(range(754215)))
-    # dataset_test.to_pickle('database/data_prosit_merged_holdout.pkl')
+    irt_train = np.load('data/intensity/irt_train.npy')
+    seq_train = np.load('data/intensity/sequence_train.npy')
+    charge_train = np.load('data/intensity/precursor_charge_train.npy')
+    spectra_train = np.load('data/intensity/intensity_train.npy')
+
+    irt_holdout = np.load('data/intensity/irt_holdout.npy')
+    seq_holdout = np.load('data/intensity/sequence_holdout.npy')
+    charge_holdout = np.load('data/intensity/precursor_charge_holdout.npy')
+    spectra_holdout = np.load('data/intensity/intensity_holdout.npy')
+
+    dataset_train = pd.DataFrame({'Sequence':list(seq_train), 'Retention time':list(irt_train), 'Charge':list(charge_train), 'Spectra' : list(spectra_train)},index=list(range(6787933)))
+    dataset_train.to_pickle('database/data_prosit_merged_train.pkl')
+    dataset_test = pd.DataFrame({'Sequence':list(seq_holdout), 'Retention time':list(irt_holdout), 'Charge':list(charge_holdout), 'Spectra' : list(spectra_holdout)},index=list(range(754215)))
+    dataset_test.to_pickle('database/data_prosit_merged_holdout.pkl')
\ No newline at end of file
\ No newline at end of file

main_custom.py

@@ -26,6 +26,7 @@ def train(model, data_train, epoch, optimizer, criterion_rt, criterion_intensity
         param.requires_grad = True
     if forward == 'both':
         for seq, charge, rt, intensity in data_train:
+            print(seq, charge, rt, intensity )
             rt, intensity = rt.float(), intensity.float()
             if torch.cuda.is_available():
                 seq, charge, rt, intensity = seq.cuda(), charge.cuda(), rt.cuda(), intensity.cuda()