T3 Task

entry.py

@@ -12,10 +12,10 @@ def prepare(config_path):
     parser = argparse.ArgumentParser()
     parser.add_argument('--process_data_mid', default=0)
     parser.add_argument('--process_data_ready', default=0)
-    parser.add_argument('--task', default='1')
+    parser.add_argument('--task', default='3')
     parser.add_argument('--base_model', default='MF')
     parser.add_argument('--seed', type=int, default=2020)
-    parser.add_argument('--ratio', default=[0.2, 0.8])
+    parser.add_argument('--ratio', default=[0.5, 0.5])
     parser.add_argument('--gpu', default='0')
     parser.add_argument('--epoch', type=int, default=10)
     parser.add_argument('--lr', type=float, default=0.01)

run.py

@@ -47,6 +47,8 @@ def __init__(self,
                         'aug_mae': 10, 'aug_rmse': 10,
                         'emcdr_mae': 10, 'emcdr_rmse': 10,
                         'ptupcdr_mae': 10, 'ptupcdr_rmse': 10}
+
+
 
     def seq_extractor(self, x):
         x = x.rstrip(']').lstrip('[').split(', ')
@@ -210,9 +212,7 @@ def eval_mae_last_epochs(self, model, data_loader, stage):
             print("Length of unique user_ids", len(set(user_ids)))
 
             # Calculate RMSE for each user
-            # user_ids = X[:, 0].unique().tolist() # Assuming UID is in the first column
             for i,user_id in enumerate(user_ids):
-                # user_indices = torch.where(X[:, 0] == user_id)[0]
                 user_targets = targets[i]
                 user_predicts = predicts[i]
                 if user_id not in user_targets_dict:
@@ -229,7 +229,7 @@ def eval_mae_last_epochs(self, model, data_loader, stage):
 
 
             # Store UID and RMSE values in a CSV file
-            filename = 'user_rmse_8_2_trial_2.csv'
+            filename = 'user_rmse_8_2_T3_MF.csv'
             with open(filename, 'w', newline='') as file:
                 writer = csv.writer(file)
                 writer.writerow(['UID', 'MAE', 'RMSE'])  # Write header
@@ -296,6 +296,7 @@ def CDR(self, model, data_src, data_map, data_meta, data_test,
             self.update_results(mae, rmse, 'ptupcdr')
             print('MAE: {} RMSE: {}'.format(mae, rmse))
 
+
     def main(self):
         model = self.get_model()
         data_src, data_tgt, data_meta, data_map, data_aug, data_test = self.get_data()