Initial WorkWise backend GPU

Dockerfile.txt

@@ -0,0 +1,13 @@
+FROM python:3.11-slim
+
+WORKDIR /app
+
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+COPY app/ ./app/
+COPY data/ ./data/
+
+EXPOSE 8000
+
+CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "8000"]

app copy 2.py

@@ -1,57 +0,0 @@
-import gradio as gr
-import spaces
-
-# Import your services
-from app.config import settings
-from app.services.vector_store import vector_store
-from app.utils.logger import setup_logger
-
-logger = setup_logger(__name__)
-
-@spaces.GPU
-def query_jira(question: str):
-    try:
-        from app.services.rag_service import process_query
-        return process_query(question)
-    except Exception as e:
-        return f"Error: {str(e)}"
-
-# Create Gradio app with API
-with gr.Blocks() as demo:
-    gr.Markdown("# WorkWise - Jira RAG Assistant")
-    
-    input_text = gr.Textbox(label="Question")
-    output_text = gr.Textbox(label="Answer")
-    submit_btn = gr.Button("Ask")
-    
-    submit_btn.click(fn=query_jira, inputs=input_text, outputs=output_text, api_name="ask")
-
-# Access FastAPI app from Gradio
-app = demo.app
-
-# CORS Middleware
-app.add_middleware(
-    CORSMiddleware,
-    allow_origins=settings.ALLOWED_ORIGINS,
-    allow_credentials=True,
-    allow_methods=["*"],
-    allow_headers=["*"],
-)
-
-# Add your custom routes to Gradio's underlying FastAPI app
-from app.routes import ingest_routes, ask_routes, metrics_routes
-
-app.include_router(ingest_routes.router, prefix="/api", tags=["Ingestion"])
-app.include_router(ask_routes.router, prefix="/api", tags=["Query"])  
-app.include_router(metrics_routes.router, prefix="/api", tags=["Metrics"])
-
-@app.get("/health")
-async def health_check():
-    try:
-        info = vector_store.get_collection_info()
-        return {"status": "healthy", "vectors_count": info.get("vectors_count", 0)}
-    except Exception as e:
-        return {"status": "error", "message": str(e)}
-
-if __name__ == "__main__":
-    demo.launch(ssr_mode=False)

app copy.py

@@ -1,87 +0,0 @@
-import gradio as gr
-from fastapi import FastAPI
-from fastapi.middleware.cors import CORSMiddleware
-import spaces
-
-# Import your existing routes and services
-from app.config import settings
-from app.routes import ingest_routes, ask_routes, metrics_routes
-from app.services.vector_store import vector_store
-from app.utils.logger import setup_logger
-
-logger = setup_logger(__name__)
-
-# Create FastAPI app
-app = FastAPI(
-    title="WorkWise Backend",
-    description="RAG-powered Jira analytics application",
-    version="1.0.0"
-)
-
-# CORS Middleware
-app.add_middleware(
-    CORSMiddleware,
-    allow_origins=settings.ALLOWED_ORIGINS,
-    allow_credentials=True,
-    allow_methods=["*"],
-    allow_headers=["*"],
-)
-
-# Include your existing routers
-app.include_router(ingest_routes.router, prefix="/api", tags=["Ingestion"])
-app.include_router(ask_routes.router, prefix="/api", tags=["Query"])
-app.include_router(metrics_routes.router, prefix="/api", tags=["Metrics"])
-
-@app.get("/")
-async def root():
-    return {
-        "status": "online",
-        "service": "WorkWise API",
-        "version": "1.0.0"
-    }
-
-@app.get("/health")
-async def health_check():
-    try:
-        info = vector_store.get_collection_info()
-        return {
-            "status": "healthy",
-            "index_path": settings.FAISS_INDEX_PATH,
-            "payloads_path": settings.FAISS_PAYLOADS_PATH,
-            "vectors_count": info.get("vectors_count", 0)
-        }
-    except Exception as e:
-        logger.error(f"Health check failed: {e}")
-        return {"status": "error", "message": str(e)}
-
-# Minimal Gradio UI (required for ZeroGPU)
-@spaces.GPU
-def query_jira(question: str):
-    """Simple wrapper for Gradio - calls your actual RAG logic"""
-    try:
-        # Import your actual query function here
-        from app.services.rag_service import process_query  # adjust import as needed
-        result = process_query(question)
-        return result
-    except Exception as e:
-        return f"Error: {str(e)}"
-
-# Create Gradio interface
-with gr.Blocks(title="WorkWise - Ask Jira") as demo:
-    gr.Markdown("# WorkWise - Jira RAG Assistant")
-    gr.Markdown("Ask questions about your Jira data")
-    
-    with gr.Row():
-        input_text = gr.Textbox(label="Your Question", placeholder="e.g., What are the top issues this sprint?")
-        output_text = gr.Textbox(label="Answer")
-    
-    submit_btn = gr.Button("Ask")
-    submit_btn.click(fn=query_jira, inputs=input_text, outputs=output_text)
-
-# Mount Gradio to FastAPI
-#app = gr.mount_gradio_app(app, demo, path="/")
-demo = gr.mount_gradio_app(demo, app, path="/")
-
-# Launch
-if __name__ == "__main__":
-    demo.launch(ssr_mode=False)

app.py

@@ -1,36 +0,0 @@
-import spaces
-import gradio as gr
-from fastapi import FastAPI
-from fastapi.responses import JSONResponse
-
-# Step 1: Create a simple function
-@spaces.GPU
-def simple_function(text):
-    return f"Echo: {text}"
-
-# Step 2: Create Gradio interface
-demo = gr.Interface(
-    fn=simple_function,
-    inputs=gr.Textbox(label="Input"),
-    outputs=gr.Textbox(label="Output")
-)
-
-# Step 3: Get FastAPI app from Gradio
-fastapi_app = demo.app
-
-# Step 4: Add FastAPI endpoints AFTER getting the app
-@fastapi_app.get("/health")
-def health_check():
-    return JSONResponse({"status": "healthy", "message": "API is working"})
-
-@fastapi_app.get("/api/test")
-def test_endpoint():
-    return JSONResponse({"message": "FastAPI endpoint works!"})
-
-@fastapi_app.post("/api/echo")
-def echo_endpoint(data: dict):
-    return JSONResponse({"echo": data})
-
-# Step 5: Launch
-if __name__ == "__main__":
-    demo.launch()

app.py.bak

@@ -1,16 +0,0 @@
-import gradio as gr
-import spaces
-from fastapi import FastAPI
-import uvicorn
-from threading import Thread
-
-@spaces.GPU
-def process(data):
-    # Your GPU logic
-    return result
-
-# Minimal Gradio UI
-demo = gr.Interface(fn=process, inputs="text", outputs="text")
-
-# Launch Gradio (keeps space alive)
-demo.launch(server_name="0.0.0.0", server_port=7860)

app.py.min.working

@@ -1,27 +0,0 @@
-import gradio as gr
-from fastapi import FastAPI
-import spaces
-
-app = FastAPI()
-
-@spaces.GPU
-def your_gpu_function(input_data):
-    # Your GPU computation
-    return {"message": "FastAPI is working"}
-    #return result
-
-# FastAPI endpoint
-@app.get("/api/predict")
-async def predict(input: str):
-    result = your_gpu_function(input)
-    return {"result": result}
-
-# Gradio interface (required for ZeroGPU)
-with gr.Blocks() as demo:
-    gr.Interface(fn=your_gpu_function, inputs="text", outputs="text")
-
-# Mount FastAPI to Gradio
-app = gr.mount_gradio_app(app, demo, path="/")
-
-if __name__ == "__main__":
-    demo.launch(ssr_mode=False)

app/{main.py.bak → main.py}

@@ -9,7 +9,7 @@ from app.utils.logger import setup_logger
 logger = setup_logger(__name__)
 
 app = FastAPI(
-    title="WorkWise Backend",
+    title="WorkWise Backend GPU",
     description="RAG-powered Jira analytics application",
     version="1.0.0"
 )
@@ -17,7 +17,7 @@ app = FastAPI(
 # CORS Middleware
 app.add_middleware(
     CORSMiddleware,
-    allow_origins=settings.ALLOWED_ORIGINS,
+    allow_origins=settings.ALLOWED_ORIGINS if hasattr(settings, "ALLOWED_ORIGINS") else ["*"],
     allow_credentials=True,
     allow_methods=["*"],
     allow_headers=["*"],
@@ -57,7 +57,6 @@ async def health_check():
         return {"status": "error", "message": str(e)}
 
 # This is needed only when this was a Docker Space. Remove for Gradio
-'''
 if __name__ == "__main__":
     import uvicorn
     uvicorn.run(
@@ -67,4 +66,3 @@ if __name__ == "__main__":
         reload=True,
         log_level=settings.LOG_LEVEL
     )
-'''

requirements.txt

@@ -1,12 +1,25 @@
-spaces
 fastapi==0.109.0
 uvicorn[standard]==0.27.0
 python-dotenv==1.0.0
+python-multipart==0.0.6    # if you accept file uploads
+
+# === Data / utilities ===
 pandas==2.2.0
 numpy==1.26.3
-sentence-transformers==2.3.1
-faiss-cpu==1.7.4
-qdrant-client==1.7.3
+requests==2.31.0
 pydantic==2.5.3
 python-multipart==0.0.6
-requests==2.31.0
+tqdm==4.66.4                  # progress bars
+loguru==0.7.2                 # clean logging
+
+# Vector/Embedding stack (keep exactly what you use)
+faiss-cpu==1.7.4           # keep if you use FAISS locally
+qdrant-client==1.7.3       # keep if using Qdrant
+
+
+# === ML / AI stack (GPU-compatible) ===
+# Using torch for embeddings/models:
+torch #-- choose the right wheel for your CUDA (see Dockerfile notes)
+transformers
+sentence-transformers
+

space.yaml

@@ -0,0 +1,33 @@
+# === Hugging Face Space Configuration ===
+
+# We’re using a custom container because this is a GPU FastAPI backend.
+sdk: "docker"
+
+# (Optional metadata)
+title: "WorkWise Backend GPU"
+emoji: "⚙️"
+colorFrom: "blue"
+colorTo: "indigo"
+app_port: 7860        # HF routes traffic through port 7860 inside the container
+pinned: false
+
+# Hardware (HF will spin up a GPU runtime)
+hardware: "gpu"
+
+# Docker build context
+# Hugging Face will automatically build your Dockerfile at repo root
+# (Make sure you have Dockerfile and requirements.txt)
+---
+
+# Optional environment variables (you can also set these from HF UI)
+env:
+  - name: HF_HOME
+    value: /data/huggingface
+  - name: HF_HUB_DISABLE_TELEMETRY
+    value: "1"
+  - name: CUDA_VISIBLE_DEVICES
+    value: "0"
+
+# If your app uses secrets (API keys, vector DB URL), define them via
+# “Settings → Repository secrets” in Hugging Face UI (not here)
+