dev

app.py

@@ -54,7 +54,7 @@ def fine_tune_model(model, dataset_name, hub_id, api_key, num_epochs, batch_size
             num_train_epochs=int(num_epochs),
             weight_decay=0.01,
             #gradient_accumulation_steps=int(grad),
-            max_grad_norm = 1.0, 
+            #max_grad_norm = 1.0, 
             load_best_model_at_end=True,
             metric_for_best_model="accuracy",
             greater_is_better=True,
@@ -156,7 +156,7 @@ def run_train(dataset_name, hub_id, api_key, num_epochs, batch_size, lr, grad):
     
     config = AutoConfig.from_pretrained("google/t5-efficient-tiny")
     model = AutoModelForSeq2SeqLM.from_config(config)
-    print(model.named_parameters())
+    print(list(model.named_parameters()))
     initialize_weights(model)
     lora_config = LoraConfig(
         r=16,  # Rank of the low-rank adaptation
@@ -165,7 +165,6 @@ def run_train(dataset_name, hub_id, api_key, num_epochs, batch_size, lr, grad):
         bias="none"  # Bias handling
     )
     model = get_peft_model(model, lora_config)
-    model.gradient_checkpointing_enable()    
     result = fine_tune_model(model, dataset_name, hub_id, api_key, num_epochs, batch_size, lr, grad)
     return result
 # Create Gradio interface