Implement async generator #648
Conversation
facebook-github-bot
added
the
CLA Signed
JasonKChow
force-pushed
the
pr648
branch
6 times, most recently
from
8401e78 to
f3c7074
Compare
Summary: Rewrite the server to use asyncio and be multithreaded. Allows handling multiple clients. Test Plan: Tests updated. Later test also tests multiple db writes simultaneously.
Summary: Use scoped sessions that is shared between threads to allow async db operations. Test Plan: Tests should pass
Summary: Logging levels were used inconsistently and would cause problems with multithreaded servers. We remove this ability. Test Plan: Tests updated.
Summary: All message handlers will emit a info logging message to support better live tracking of server functions. Test Plan: Tests should pass
Summary: To support multi client server, strategy methods will not act on copies of models to avoid changing tensor gradients between two threads. Test Plan: New test
Summary: A subclass of the aepsych server with methods specifically to run the server in a background process. This will be used to ensured that even within the same main script, the server will run like an actual server and does not do anything sneaky like bypassing the async queue. Test Plan: New test
Summary: Add the ___reduce__ method to transform wrappers (model/generator). This allows transform wrapped objects to be pickled (with original pickle module, not just dill) Test Plan: tests should pass
Summary: A generator that allows asynchronous generation of points. Uses a different process to generate points. This doesn't fix the situation where modeling fitting takes a long time. Test Plan: New test
|
@JasonKChow has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
|
Hi @JasonKChow! Thank you for your pull request. We require contributors to sign our Contributor License Agreement, and yours needs attention. You currently have a record in our system, but the CLA is no longer valid, and will need to be resubmitted. ProcessIn order for us to review and merge your suggested changes, please sign at https://code.facebook.com/cla. If you are contributing on behalf of someone else (eg your employer), the individual CLA may not be sufficient and your employer may need to sign the corporate CLA. Once the CLA is signed, our tooling will perform checks and validations. Afterwards, the pull request will be tagged with If you have received this in error or have any questions, please contact us at [email protected]. Thanks! |
Summary: A generator that allows asynchronous generation of points. Uses a different process to generate points. This doesn't fix the situation where modeling fitting takes a long time.
Test Plan: New test
Stack created with Sapling. Best reviewed with ReviewStack.