Use asyncio.sleep in AsyncInferenceClient (not time.sleep) (#2674)

huggingface · Nov 22, 2024 · 1c10aaa · 1c10aaa
1 parent fcecbb8
commit 1c10aaa
Show file tree

Hide file tree

Showing 2 changed files with 2 additions and 2 deletions.
diff --git a/src/huggingface_hub/inference/_generated/_async_client.py b/src/huggingface_hub/inference/_generated/_async_client.py
@@ -324,7 +324,7 @@ async def post(
                         logger.info(f"Waiting for model to be loaded on the server: {error}")
                         if "X-wait-for-model" not in headers and url.startswith(INFERENCE_ENDPOINT):
                             headers["X-wait-for-model"] = "1"
-                        time.sleep(1)
+                        await asyncio.sleep(1)
                         if timeout is not None:
                             timeout = max(self.timeout - (time.time() - t0), 1)  # type: ignore
                         continue

diff --git a/utils/generate_async_inference_client.py b/utils/generate_async_inference_client.py
@@ -226,7 +226,7 @@ def _rename_to_AsyncInferenceClient(code: str) -> str:
                         logger.info(f"Waiting for model to be loaded on the server: {error}")
                         if "X-wait-for-model" not in headers and url.startswith(INFERENCE_ENDPOINT):
                             headers["X-wait-for-model"] = "1"
-                        time.sleep(1)
+                        await asyncio.sleep(1)
                         if timeout is not None:
                             timeout = max(self.timeout - (time.time() - t0), 1)  # type: ignore
                         continue