Upload app.py

app.py

@@ -20,9 +20,9 @@ lora_config = LoraConfig(
     lora_dropout=0.1,  # Dropout for LoRA layers
     bias="none"  # Bias handling
 )
-model = AutoModelForSeq2SeqLM.from_pretrained('google/t5-efficient-tiny-nh8', num_labels=2, force_download=True)
-model = get_peft_model(model, lora_config)
-model.gradient_checkpointing_enable()   
+model = AutoModelForSeq2SeqLM.from_pretrained('google/t5-efficient-tiny', num_labels=2, force_download=True)
+#model = get_peft_model(model, lora_config)
+#model.gradient_checkpointing_enable()   
 
 @spaces.GPU(duration=120)
 def fine_tune_model(model, dataset_name, hub_id, api_key, num_epochs, batch_size, lr, grad):
@@ -53,8 +53,8 @@ def fine_tune_model(model, dataset_name, hub_id, api_key, num_epochs, batch_size
             per_device_eval_batch_size=int(batch_size), 
             num_train_epochs=int(num_epochs),
             weight_decay=0.01,
-            gradient_accumulation_steps=int(grad),
-            max_grad_norm = 1.0, 
+            #gradient_accumulation_steps=int(grad),
+            #max_grad_norm = 1.0, 
             load_best_model_at_end=True,
             metric_for_best_model="accuracy",
             greater_is_better=True,
@@ -84,7 +84,6 @@ def fine_tune_model(model, dataset_name, hub_id, api_key, num_epochs, batch_size
                 train_dataset=tokenized_train_dataset,
                 eval_dataset=tokenized_test_dataset,
                 compute_metrics=compute_metrics,
-                #callbacks=[LoggingCallback()], 
             )            
         except:
             # Load the dataset