dev

app.py

@@ -97,6 +97,7 @@ def fine_tune_model(model, dataset_name, hub_id, api_key, num_epochs, batch_size
                     examples['text'], 
                     max_length=max_length,  # Set to None for dynamic padding
                     truncation=True,
+                    padding=True, 
                 )
             
                 # Setup the decoder input IDs (shifted right)
@@ -104,6 +105,7 @@ def fine_tune_model(model, dataset_name, hub_id, api_key, num_epochs, batch_size
                     examples['target'], 
                     max_length=max_length,  # Set to None for dynamic padding
                     truncation=True,
+                    padding=True, 
                     text_target=examples['target']  # Use text_target for target text
                 )
             
@@ -147,6 +149,14 @@ def predict(text):
 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)
+    lora_config = LoraConfig(
+        r=16,  # Rank of the low-rank adaptation
+        lora_alpha=32,  # Scaling factor
+        lora_dropout=0.1,  # Dropout for LoRA layers
+        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