Update app.py

app.py

@@ -1,27 +1,25 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+from transformers import pipeline
 
-# Initialize the Hugging Face Inference client with the chosen model
-client = InferenceClient(repo_id="meta-llama/Llama-2-7b-chat-hf")
+# Load the model
+model = pipeline("text-generation", model="meta-llama/Llama-2-7b-chat-hf")
 
-def evaluate_response(prompt):
-    # Send the prompt to the model and get the generated response
-    response = client.text_generation(prompt, max_new_tokens=200, temperature=0.7)
-    generated_text = response[0]['generated_text']
-    
-    # Simple quality evaluation: check length of the response
-    word_count = len(generated_text.split())
-    quality = "Good" if word_count > 10 else "Poor"
-    
-    return generated_text, f"Response Quality: {quality}"
+def generate_response(prompt):
+    response = model(prompt, max_length=100, num_return_sequences=1)
+    # Check if the response is not empty and contains the expected key
+    if response and isinstance(response, list) and 'generated_text' in response[0]:
+        return response[0]['generated_text']
+    else:
+        return "Error: No response generated."
 
-# Build the Gradio interface
+# Create the Gradio interface
 iface = gr.Interface(
-    fn=evaluate_response,
-    inputs=gr.Textbox(lines=4, label="Enter your prompt"),
-    outputs=[gr.Textbox(label="Model Response"), gr.Textbox(label="Quality Analysis")],
-    title="Model Understanding Test",
-    description="Enter instructions or a question, the app sends it to a language model and automatically evaluates the quality of the response."
+    fn=generate_response,
+    inputs=gr.Textbox(label="Enter your prompt or question:"),
+    outputs=gr.Textbox(label="Model Response:"),
+    title="Test Understanding Prompt",
+    description="Enter a prompt to see how well the model understands it."
 )
 
+# Launch the app
 iface.launch()