add debugging statements

interactive_demo.py

@@ -257,8 +257,11 @@ def interactive_demo(cfg: DemoConfig):
     # Build the VLM --> Download/Load Pretrained Model from Checkpoint
     overwatch.info("Initializing VLM =>> Bundling Models, Image Processors, and Tokenizer")
     hf_token = cfg.hf_token.read_text().strip() if isinstance(cfg.hf_token, Path) else os.environ[cfg.hf_token]
+
+    overwatch.info(f"*** hf_token {hf_token[:4]}")
+
     vlm = load_vlm(cfg.model_family, cfg.model_id, cfg.run_dir, hf_token=hf_token)
-    print("model loaded")
+    overwatch.info("*** model loaded")
 
     global worker
     global limit_model_concurrency
@@ -266,6 +269,7 @@ def interactive_demo(cfg: DemoConfig):
     worker = ModelWorker(
         cfg.controller_address, cfg.worker_address, worker_id, cfg.no_register, vlm, cfg.model_name
     )
+    overwatch.info("*** starting uvicorn")
     uvicorn.run(app, host=cfg.host, port=cfg.port, log_level="debug")
 
 

serve/controller.py

@@ -246,6 +246,7 @@ app = FastAPI()
 @app.post("/register_worker")
 async def register_worker(request: Request):
     data = await request.json()
+    logger.info(f"data = {data}")
     controller.register_worker(data["worker_name"], data["check_heart_beat"], data.get("worker_status", None))
 
 
@@ -295,4 +296,4 @@ if __name__ == "__main__":
     logger.info(f"args: {args}")
 
     controller = Controller(args.dispatch_method)
-    uvicorn.run(app, host=args.host, port=args.port, log_level="info")
+    uvicorn.run(app, host=args.host, port=args.port, log_level="debug")