-
Notifications
You must be signed in to change notification settings - Fork 364
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.
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
feature(pu): add td3_vae algorithm #152
Merged
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
Codecov Report
@@ Coverage Diff @@
## main #152 +/- ##
==========================================
- Coverage 84.73% 84.37% -0.36%
==========================================
Files 430 436 +6
Lines 33019 33619 +600
==========================================
+ Hits 27979 28367 +388
- Misses 5040 5252 +212
Flags with carried forward coverage won't be shown. Click here to find out more.
Continue to review full report at Codecov.
|
PaParaZz1
requested changes
Dec 15, 2021
…presentation shift correction
…ddings of obs and action, use tanh after sample z and after the reconstruction_action head
…ot empty and will leave over the data in random collect phase
PaParaZz1
requested changes
Dec 30, 2021
puyuan1996
changed the title
WIP: feature(pu): add td3_vae algorithm
feature(pu): add td3_vae algorithm
Dec 30, 2021
PaParaZz1
force-pushed
the
main
branch
2 times, most recently
from
January 4, 2022 06:27
ee876e0
to
c6947cd
Compare
PaParaZz1
approved these changes
Jan 4, 2022
puyuan1996
added a commit
to puyuan1996/DI-engine
that referenced
this pull request
Apr 18, 2022
* feature(pu): add td3_vae * fix(pu): fix log typo * polish(pu): vae and rl update alternately * polish(pu):polish td3_vae config * test(pu): delete noise and change the data for updating vae * fix(pu): use latent action relabel * polish(pu):polish td3_vae config * feature(pu): add lunarlander continuous env(ci skip) * polish(pu):polish config * feature(pu): add ddpg lunarlander_cont config * feature(pu): add noise in original action in collect phase and add representation shift correction * fix(pu): using decode_with_obs to use the obs_encoding generating from current obs * feature(pu): representaion shift correction for each transition * feature(pu): add latent space constraint and tanh operation to sample z in vae * polish(pu): polish config * polish(pu): polish as review * polish(pu):polish td3_vae config * polish(pu): polish vae structure, use add not concat between the embeddings of obs and action, use tanh after sample z and after the reconstruction_action head * polish(pu):polish kl weight and prediction weight * polish(pu):polish td3_vae using the best setting * style(pu): yapf format * polish(pu):polish config * fix(pu): fix bug when collector_env_num>1, the self._traj_buffer is not empty and will leave over the data in random collect phase * polish(pu): update the current best config * polish(pu): polish config * polish(pu):polish as review * polish(pu):revert unwanted changes * polish(pu): polish vae obs encoding * polish(pu): return dict-type result and update comment format in vae * feature(pu): add save vae_model state_dict * polish(pu): add test_vae and yapf format Co-authored-by: niuyazhe <niuyazhe@sensetime.com>
SolenoidWGT
pushed a commit
to SolenoidWGT/DI-engine
that referenced
this pull request
Aug 22, 2023
* style(internlm): fix lint error * feat(utils/logger.py): support uniscale logger * fix(utils/logger.py): fix import circular error * feat(train.py): support dashboard metric panel and fix ci train config * fix(ci_scripts/train/slurm_train.sh): fix ci train error * fix(ci_scripts/train/torchrun.sh): fix ci train error * fix(ci_scripts/train): restore ci update * fix(config.json): delete alert webhook * feat(train.py): optimize func init logger * feat(config.json): delete config.json --------- Co-authored-by: 黄婷 <huangting3@CN0014010744M.local> Co-authored-by: huangting.p <huangting@sensetime.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in 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
Add the td3_vae algorithm, train the vae encoder-decoder to encode the original continuous action into the latent action space, and test it in env lunarlander-cont.
Related Issue
TODO
Check List