Update app.py

app.py

@@ -104,7 +104,7 @@ if (runModel=='1'):
     # Create an instance of the custom loss function
     training_args = TrainingArguments(
         output_dir='./results_' + modelNameToUse,
-        num_train_epochs=5,
+        num_train_epochs=10,
         per_device_train_batch_size=8,
         per_device_eval_batch_size=8,
         warmup_steps=500,
@@ -112,8 +112,7 @@ if (runModel=='1'):
         logging_dir='./logs_' + modelNameToUse,
         logging_steps=10,
         evaluation_strategy="epoch",  # Evaluation strategy is 'epoch'
-        save_strategy="epoch",       # Save strategy should also be 'epoch'
-        load_best_model_at_end=True,  # Load the best model based on evaluation
+
     )
     
     trainer = Trainer(
@@ -133,7 +132,6 @@ if (runModel=='1'):
         0: "lastmonth",
         1: "nextweek",
         2: "sevendays"
-        
     }
 
     def evaluate_and_report_errors(model, dataloader, tokenizer):
@@ -227,15 +225,17 @@ if (runModel=='1'):
         path_in_repo="data-timeframe_model",
         repo_id=repo_name,
         token=api_token,
-        commit_message="Update fine-tuned model"
+        commit_message="Update fine-tuned model for test"
     )
+    
     upload_folder(
         folder_path=tokenizer_path,
         path_in_repo="data-timeframe_tokenizer",
         repo_id=repo_name,
         token=api_token,
-        commit_message="Update fine-tuned tokenizer"
+        commit_message="Update fine-tuned tokenizer for test"
     )
+    print("tokenizer folder: ", tokenizer_path)
     
 else:
     print('Load Pre-trained')