Update app.py

app.py

@@ -2,35 +2,35 @@ import streamlit as st
 from transformers import AutoModelForCausalLM, AutoTokenizer
 from peft import PeftModel
 
-# Load the base model and the fine-tuned model
+# Load the base model and adapter
 @st.cache_resource
 def load_model():
-    base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-Coder-3B-Instruct")
-    model = PeftModel.from_pretrained(base_model, "mohamedyd/Natural-Coder-3B-Instruct-V1")
-    tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen2.5-Coder-3B-Instruct")
+    base_model_name = "Qwen/Qwen2.5-Coder-3B-Instruct"  # Ensure this is the correct base model
+    adapter_model_name = "mohamedyd/Natural-Coder-3B-Instruct-V1"
+
+    # Load the base model
+    base_model = AutoModelForCausalLM.from_pretrained(base_model_name, trust_remote_code=True)
+    
+    # Load the PEFT adapter on top of the base model
+    model = PeftModel.from_pretrained(base_model, adapter_model_name)
+
+    # Load tokenizer from base model
+    tokenizer = AutoTokenizer.from_pretrained(base_model_name, trust_remote_code=True)
+
     return model, tokenizer
 
 model, tokenizer = load_model()
 
-# Streamlit app
+# Streamlit App UI
 st.title("Natural-Coder-3B-Instruct-V1 Model Interaction")
 
-# Text input for user prompt
 user_input = st.text_area("Enter your prompt here:", height=150)
 
-# Button to generate response
 if st.button("Generate Response"):
     if user_input:
-        # Tokenize the input
         inputs = tokenizer(user_input, return_tensors="pt")
-        
-        # Generate response
         outputs = model.generate(**inputs, max_length=512, num_return_sequences=1)
-        
-        # Decode the output
         response = tokenizer.decode(outputs[0], skip_special_tokens=True)
-        
-        # Display the response
         st.write("Model Response:")
         st.write(response)
     else: