Skip to content

Add support for async method and remove context PythonServer #16453

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 4 commits into from
Jan 23, 2023
Merged

Add support for async method and remove context PythonServer #16453

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
1228799
update
Jan 20, 2023
f8aedfc
add changelog
Jan 20, 2023
12f4d94
Merge branch 'master' into add_support_for_async
tchaton Jan 23, 2023
377fac9
Merge branch 'master' into add_support_for_async
tchaton Jan 23, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion src/lightning_app/CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,7 +14,7 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).

### Changed

-
- Add support for async predict method in PythonServer and remove torch context ([#16453](https://github.com/Lightning-AI/lightning/pull/16453))


### Deprecated
Expand Down
17 changes: 9 additions & 8 deletions src/lightning_app/components/serve/python_server.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,4 +1,5 @@
import abc
import asyncio
import base64
import os
import platform
Expand Down Expand Up @@ -252,19 +253,19 @@ def _get_sample_dict_from_datatype(datatype: Any) -> dict:
return out

def _attach_predict_fn(self, fastapi_app: FastAPI) -> None:
from torch import inference_mode, no_grad

input_type: type = self.configure_input_type()
output_type: type = self.configure_output_type()

device = _get_device()
context = no_grad if device.type == "mps" else inference_mode
def predict_fn_sync(request: input_type): # type: ignore
return self.predict(request)

def predict_fn(request: input_type): # type: ignore
with context():
return self.predict(request)
async def async_predict_fn(request: input_type): # type: ignore
return await self.predict(request)

fastapi_app.post("/predict", response_model=output_type)(predict_fn)
if asyncio.iscoroutinefunction(self.predict):
fastapi_app.post("/predict", response_model=output_type)(async_predict_fn)
else:
fastapi_app.post("/predict", response_model=output_type)(predict_fn_sync)

def get_code_sample(self, url: str) -> Optional[str]:
input_type: Any = self.configure_input_type()
Expand Down