Qlib RL framework (stage 2) - trainer #1125
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
you-n-g
reviewed
Jun 24, 2022
|
Please merge main branch to run the new version of CI |
|
It looks great now! |
qianyun210603
pushed a commit
to qianyun210603/qlib
that referenced
this pull request
Mar 23, 2023
* checkpoint (cherry picked from commit 1a8e0bd) * Not a workable version (cherry picked from commit 3498e18) * vessel * ckpt * . * vessel * . * . * checkpoint callback * . * cleanup * logger * . * test * . * add test * . * . * . * . * New reward * Add train API * fix mypy * fix lint * More comment * 3.7 compat * fix test * fix test * . * Resolve comments * fix typehint
# for free
to join this conversation on GitHub.
Already have an account?
# to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
This PR introduces a trainer implementation that is used to train a policy in Qlib RL framework.
This includes the following components:
The designs mimicks pytorch-lightning. The key difference is that it's tailored for RL paradigm, in which data comes from online collects.
The trainer relies on the full implementation of logger to be 100% powered. The logger will be refactored and implemented in the next stage.
Motivation and Context
It's an immediate follow-up of #1076.
How Has This Been Tested?
pytest qlib/tests/test_all_pipeline.pyunder upper directory ofqlib.Screenshots of Test Results (if appropriate):
Types of changes