Update app.py

app.py

@@ -1,64 +1,86 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+from transformers import GPT2LMHeadModel, GPT2Tokenizer
+import os
+import gdown
+from pathlib import Path
 
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+MODEL_FILE_ID = "1-4c80EuzAzXvmXsuA6pgVk7Drkv8Dh67"
+MODEL_FILE_NAME = "model.safetensors"  # Adjust based on your file name
+MODEL_DIR = "fine_tuned_gpt2_medium"  # Adjust based on your file name
 
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+# Download the model from Google Drive
+def download_model():
+    model_path = Path(MODEL_DIR) / MODEL_FILE_NAME
+    if not model_path.exists():
+        print("Downloading model from Google Drive...")
+        gdown.download(
+            gdown.download(
+            f"https://drive.google.com/uc?id={MODEL_FILE_ID}",
+            str(model_path),
+            quiet=False
+        ),
+            str(model_path),
+            quiet=False
+        )
+        print("Model downloaded successfully.")
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+# Load the model and tokenizer
+def load_model():
+    download_model()
+    tokenizer = GPT2Tokenizer.from_pretrained("fine_tuned_gpt2_medium")  # Use base tokenizer if necessary
+    model = GPT2LMHeadModel.from_pretrained(
+        pretrained_model_name_or_path=MODEL_DIR,
+        local_files_only=True  # Ensure it loads from local storage
+    )
+    return model, tokenizer
 
-    messages.append({"role": "user", "content": message})
+model, tokenizer = load_model()
 
-    response = ""
+# Functions for question generation and feedback
+def generate_question(response, history):
+    input_text = ""
+    for qa in history:
+        input_text += f"Question: {qa['question']}\nAnswer: {qa['answer']}\n"
+    input_text += f"Response: {response}\nNext Question:"
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
+    inputs = tokenizer.encode(input_text, return_tensors="pt", truncation=True, max_length=512)
+    outputs = model.generate(inputs, max_length=100, num_return_sequences=1, pad_token_id=tokenizer.eos_token_id)
+    generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    next_question = generated_text.split("Next Question:")[-1].strip()
+    return next_question
 
-        response += token
-        yield response
+def get_feedback(session_data):
+    input_text = "Session Feedback:\n"
+    for qa in session_data:
+        input_text += f"Question: {qa['question']}\nAnswer: {qa['answer']}\n"
 
+    inputs = tokenizer.encode(input_text, return_tensors="pt", truncation=True, max_length=512)
+    outputs = model.generate(inputs, max_length=150, num_return_sequences=1, pad_token_id=tokenizer.eos_token_id)
+    generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    feedback = generated_text.split("Session Feedback:")[-1].strip()
+    return feedback
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
+# Gradio Interface
+def question_interface(response, history):
+    history = eval(history)  # Convert string input to list
+    return generate_question(response, history)
+
+def feedback_interface(session_data):
+    session_data = eval(session_data)  # Convert string input to list
+    return get_feedback(session_data)
+
+iface = gr.Interface(
+    fn={
+        "generate_question": question_interface,
+        "get_feedback": feedback_interface
+    },
+    inputs=[
+        gr.inputs.Textbox(label="User Response"),
+        gr.inputs.Textbox(label="History (list of Q&A in string format)")
     ],
+    outputs="text",
+    live=True
 )
 
-
-if __name__ == "__main__":
-    demo.launch()
+iface.launch()