Delete app.py

app.py

@@ -1,161 +0,0 @@
-import os
-import json
-import uuid
-import httpx
-import gradio as gr
-from fastapi import FastAPI, HTTPException, Request
-from transformers import pipeline, AutoTokenizer, AutoModelForCausalLM
-import uvicorn
-import asyncio
-
-# ✅ Securely Load Hugging Face Token
-HF_TOKEN = os.getenv("HF_TOKEN")
-if not HF_TOKEN:
-    raise ValueError("❌ HF_TOKEN not found! Set it in Hugging Face Secrets.")
-
-# ✅ Load Model Configuration
-MODEL_NAME = "hpyapali/tinyllama-workout"
-event_store = {}  # Store AI responses with event_id
-
-app = FastAPI()
-
-# ✅ Log server restart
-print("🔄 Restarting Hugging Face AI Model Server...")
-
-# ✅ Load AI Model
-try:
-    print("🔄 Loading AI Model...")
-    tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME, token=HF_TOKEN)
-    model = AutoModelForCausalLM.from_pretrained(MODEL_NAME, token=HF_TOKEN)
-    pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)
-    print("✅ AI Model Loaded Successfully!")
-except Exception as e:
-    print(f"❌ Error loading model: {e}")
-    pipe = None
-
-
-# ✅ AI Function - Analyzes workout data
-def analyze_workouts(last_workouts: str):
-    """Generates AI-based workout rankings based on heart rate recovery."""
-    if pipe is None:
-        return "❌ AI model is not loaded."
-
-    if not last_workouts.strip():
-        return "❌ No workout data provided."
-
-    instruction = (
-        "You are a fitness AI assistant. Rank the following workouts based on heart rate recovery after 2 minutes."
-        "\n\n### Ranking Rules:"
-        "\n- A **larger heart rate dip** indicates better recovery."
-        "\n- If two workouts have the same HR dip, **rank by highest peak HR**."
-        "\n\n### Workouts Data:\n"
-        f"{last_workouts}"
-        "\n\n### Output Format (Rank from best to worst, no explanation, just rankings):"
-        "\n1. Best: Running - HR dip: 28 bpm"
-        "\n2. Cycling - HR dip: 25 bpm"
-        "\n3. Rowing - HR dip: 22 bpm"
-        "\n4. Strength Training - HR dip: 18 bpm"
-        "\n5. Walking - HR dip: 12 bpm"
-        "\n6. Yoga - HR dip: 8 bpm"
-    )
-
-    try:
-        result = pipe(
-            instruction,
-            max_new_tokens=250,
-            temperature=0.3,
-            top_p=0.9,
-            do_sample=True,
-            return_full_text=False
-        )
-
-        if not result or not result[0].get("generated_text", "").strip():
-            return "❌ AI did not generate a valid response."
-
-        return result[0]["generated_text"].strip()
-
-    except Exception as e:
-        return f"❌ Error generating workout recommendation: {str(e)}"
-
-
-# ✅ API Route for Processing Workout Data
-@app.post("/gradio_api/call/predict")
-async def process_workout_request(request: Request):
-    try:
-        req_body = await request.json()
-        print("📩 RAW REQUEST FROM HF:", req_body)  
-
-        if "data" not in req_body or not isinstance(req_body["data"], list):
-            raise HTTPException(status_code=400, detail="Invalid request format: 'data' must be a list.")
-
-        last_workouts = req_body["data"][0]
-        event_id = str(uuid.uuid4())  
-
-        print(f"✅ Processing AI Request - Event ID: {event_id}")
-
-        response_text = analyze_workouts(last_workouts)
-
-        event_store[event_id] = response_text  
-
-        webhook_url = req_body.get("webhook_url")
-        if webhook_url:
-            print(f"📡 Sending response to Webhook: {webhook_url}")
-            async with httpx.AsyncClient() as client:
-                await client.post(webhook_url, json={"event_id": event_id, "data": [response_text]})
-
-        return {"event_id": event_id}  
-
-    except Exception as e:
-        print(f"❌ Error processing request: {e}")
-        raise HTTPException(status_code=500, detail=str(e))
-
-
-# ✅ Polling API (If Webhook Fails)
-@app.get("/gradio_api/poll/{event_id}")
-async def poll(event_id: str):
-    """Fetches stored AI response for a given event ID."""
-    if event_id in event_store:
-        return {"data": [event_store.pop(event_id)]}
-    return {"detail": "Not Found"}
-
-
-# ✅ Webhook Receiver (For Debugging Webhook Calls)
-@app.post("/fineTuneModel")
-async def receive_webhook(request: Request):
-    """Handles webhook responses (useful for debugging webhook calls)."""
-    try:
-        req_body = await request.json()
-        print("📩 Webhook Received:", req_body)
-        return {"status": "success", "received": req_body}
-    except Exception as e:
-        return {"error": str(e)}
-
-
-# ✅ Health Check
-@app.get("/")
-async def root():
-    return {"message": "Workout Analysis & Ranking AI is running!"}
-
-
-# ✅ Gradio UI for Testing
-iface = gr.Interface(
-    fn=analyze_workouts,
-    inputs="text",
-    outputs="text",
-    title="Workout Analysis & Ranking AI",
-    description="Enter workout data to analyze effectiveness, rank workouts, and receive improvement recommendations."
-)
-
-
-# ✅ Start Both FastAPI & Gradio
-def start_gradio():
-    iface.launch(server_name="0.0.0.0", server_port=7860, share=True)
-
-def start_fastapi():
-    uvicorn.run(app, host="0.0.0.0", port=7861)
-
-# ✅ Run both servers in parallel
-if __name__ == "__main__":
-    import threading
-    threading.Thread(target=start_gradio).start()
-    threading.Thread(target=start_fastapi).start()