Update app.py

app.py

@@ -1,64 +1,97 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+import PyPDF2
+import os
+from openai import OpenAI
 
-"""
-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")
+# Function to extract text from PDF
+def extract_text_from_pdf(pdf_file):
+    reader = PyPDF2.PdfReader(pdf_file)
+    text = ""
+    for page in reader.pages:
+        if page == reader.pages[0]:
+            continue
+        text += page.extract_text() + "\n"
+    return text
 
+# Mock LLM response function
+# Replace this with your actual LLM integration
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+client = OpenAI(
+    base_url="https://api.studio.nebius.com/v1/",
+    api_key=os.environ.get("NEBIUS_API_KEY")
+)
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+def answer_question(pdf_text, question, history):
+    # You may want to truncate pdf_text if it's too long for the model
+    context = pdf_text[:16000]  # adjust as needed
+    messages = [
+        {"role": "system", "content": '''You are a helpful assistant who specializes in body composition, diet, and exercise. 
+         Answer questions based on the provided document. Encourage the user to seek a professional if they have serious concerns whenever appropriate.'''},
+        {"role": "user", "content": f"Document: {context}"},
+        {"role": "user", "content": f"Question: {question}"}
+    ]
+    if history:
+        for msg in history:
+            if msg["role"] in ["user", "assistant"]:
+                messages.append(msg)
+    # Add the new user question
+    messages.append({"role": "user", "content": question}) 
 
-    messages.append({"role": "user", "content": message})
+    response = client.chat.completions.create(
+        model="meta-llama/Meta-Llama-3.1-70B-Instruct",
+        temperature=0.6,
+        top_p=0.95,
+        messages=messages
+    )
+    return response.choices[0].message.content
 
-    response = ""
+# Gradio chatbot logic
+class PDFChatbot:
+    def __init__(self):
+        self.pdf_text = None
 
-    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
+    def upload_pdf(self, pdf_file):
+        if pdf_file is None:
+            return "No PDF uploaded."
+        self.pdf_text = extract_text_from_pdf(pdf_file)
+        return "PDF uploaded and processed. You can now ask questions."
 
-        response += token
-        yield response
+    def chat(self, message, history):
+        if self.pdf_text is None:
+            return "Please upload a PDF first.", history
+        # Ensure history is a list of dicts with 'role' and 'content'
+        if history is None:
+            history = []
+        # Only keep user/assistant messages for context
+        context_history = [msg for msg in history if msg["role"] in ["user", "assistant"]]
+        answer = answer_question(self.pdf_text, message, context_history)
+        print("User:", message)
+        print("LLM Answer:", answer)
+        history = history + [
+            {"role": "user", "content": message},
+            {"role": "assistant", "content": answer}
+        ]
+        return "", history
 
+pdf_bot = PDFChatbot()
 
-"""
-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)",
-        ),
-    ],
-)
+with gr.Blocks() as demo:
+    gr.Markdown("Body Composition Agent\nUpload a PDF of your recent body spec or any other document you use to track your body composition and ask questions about its content.")
+    pdf_file = gr.File(label="Upload PDF", file_types=[".pdf"])
+    upload_btn = gr.Button("Process PDF")
+    upload_status = gr.Textbox(label="Status", interactive=False)
+    chatbot = gr.Chatbot(type="messages")
+    msg = gr.Textbox(label="Your question:")
+    submit_btn = gr.Button("Submit")
+
+    def handle_upload(pdf):
+        return pdf_bot.upload_pdf(pdf)
+
+    def handle_chat(message, history):
+        return pdf_bot.chat(message, history)
 
+    upload_btn.click(handle_upload, inputs=pdf_file, outputs=upload_status)
+    submit_btn.click(handle_chat, inputs=[msg, chatbot], outputs=[msg, chatbot])
+    msg.submit(handle_chat, inputs=[msg, chatbot], outputs=[msg, chatbot])
 
-if __name__ == "__main__":
-    demo.launch()
+demo.launch()