update RedirectResponse

m3_server.py

@@ -11,6 +11,7 @@ from fastapi import FastAPI, HTTPException, Request
 from fastapi.responses import JSONResponse
 from FlagEmbedding import BGEM3FlagModel
 from pydantic import BaseModel
+from starlette.responses import RedirectResponse
 from starlette.status import HTTP_504_GATEWAY_TIMEOUT
 
 _ = """
@@ -27,7 +28,9 @@ max_q_length = 256  # max context lenght for questions in re-ranker
 request_flush_timeout = 0.1  # flush time out for future improvements on api calls / gpu batches (for now is pretty basic)
 rerank_weights = [0.4, 0.2, 0.4]  # re-rank score weights
 request_time_out = 30  # Timeout threshold
+request_time_out = 1200  # Timeout threshold
 gpu_time_out = 5  # gpu processing timeout threshold
+gpu_time_out = 600  # gpu processing timeout threshold
 port = 3000
 port = 7860
 DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
@@ -174,7 +177,7 @@ class RequestProcessor:
             await self.queue.put((request_data, request_type, request_id))
             return await self.response_futures[request_id]
         except Exception as e:
-            raise HTTPException(status_code=500, detail=f"Internal Server Error {e}")
+            raise HTTPException(status_code=500, detail=f"Internal Server Error: {e}")
 
 
 description = dedent(
@@ -232,7 +235,8 @@ async def timeout_middleware(request: Request, call_next):
 @app.get("/")
 async def landing():
     """Define landing page."""
-    return "Swagger UI at https://mikeee-baai-m3.hf.space/docs"
+    # return "Swagger UI at https://mikeee-baai-m3.hf.space/docs"
+    return RedirectResponse("/docs")
 
 
 @app.post("/embed/", response_model=EmbedResponse)