issues
search
google-deepmind
/
rlax
https://rlax.readthedocs.io
Apache License 2.0
1.24k
stars
85
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Create new version 0.1.3 of RLax.
#86
copybara-service[bot]
closed
2 years ago
0
Remove incremental_update from rlax: all usages ported to optax.incremental_update
#85
copybara-service[bot]
closed
2 years ago
0
Add utilities to support interruptions.
#84
copybara-service[bot]
closed
2 years ago
0
Minor edits to moving averages.
#83
copybara-service[bot]
closed
2 years ago
0
Add tree utilities to rlax.
#82
copybara-service[bot]
closed
2 years ago
0
Add utilities to extract overlapping subsequences from trajectories.
#81
copybara-service[bot]
closed
2 years ago
0
Update .pylintrc
#80
copybara-service[bot]
closed
2 years ago
0
Add moving averages helpers to rlax.
#79
copybara-service[bot]
closed
2 years ago
0
Add a pair of transforms where the scalar values are reparametrised as the linear combination of two-hot values on a non-linearly spaced discrete support.
#78
copybara-service[bot]
closed
2 years ago
1
Move usages of soon to be deprecated rlax.periodic_update to optax.periodic_update.
#77
copybara-service[bot]
closed
2 years ago
0
Move usages of rlax.periodic_update to optax.periodic_update.
#76
copybara-service[bot]
closed
1 year ago
0
Send deprecation warning for rlax nested_updates in favor of using optax.
#75
copybara-service[bot]
closed
2 years ago
0
Send deprecation warning for rlax.distributions in favor of using distrax.
#74
copybara-service[bot]
closed
2 years ago
0
Add a particular pair of transforms used by muzero that combine a non linear squashing function with a reparametrisation of the scalar as linear combination of two hot values in a discrete suppport.
#73
copybara-service[bot]
closed
2 years ago
0
internal change
#72
copybara-service[bot]
closed
1 year ago
1
Support Array lambda_ in Vtrace.
#71
copybara-service[bot]
closed
2 years ago
1
Migrate RLax squashed gaussian to use Distrax. Explicitly broadcast shapes in Distrax scalar affine to avoid rank promotion errors.
#70
copybara-service[bot]
closed
2 years ago
0
Update squashed gaussian distribution in rlax for prob and logprob to numerically match distrax's implementation.
#69
copybara-service[bot]
closed
2 years ago
0
Add test for squashed gaussian in rlax distributions.
#68
copybara-service[bot]
closed
2 years ago
0
Expose `use_jnp_split` arg in `tree_split_leaves`.
#67
copybara-service[bot]
closed
2 years ago
0
Update Jinja2 versioning to avoid Sphinx failures.
#66
copybara-service[bot]
closed
2 years ago
0
Add quantile_regression_loss to the rlax api.
#65
copybara-service[bot]
closed
2 years ago
1
Make documentation download-able as PDF?
#64
IanQS
opened
2 years ago
0
Bugfix to quantile_expected_sarsa.
#63
copybara-service[bot]
closed
2 years ago
1
Support for Bandits?
#62
IanQS
closed
2 years ago
0
Create new version 0.1.2 of RLax.
#61
copybara-service[bot]
closed
2 years ago
0
Release new version to loose jax version constraints
#60
ethanluoyc
closed
2 years ago
1
Use distrax distributions in epsilon_softmax.
#59
copybara-service[bot]
closed
2 years ago
0
RLax: Remove distribution validity checking from categorical_sample as it is already done in distrax.
#58
copybara-service[bot]
closed
2 years ago
0
rlax: Replace rlax categorical cross entropy computation with distrax components.
#57
copybara-service[bot]
closed
2 years ago
0
distrax: Use a safer log_prob in KL divergence between categoricals.
#56
copybara-service[bot]
closed
1 year ago
0
Move decoupled_multivariate_normal_kl_divergence out of distributions.py
#55
copybara-service[bot]
closed
2 years ago
0
Question of `logprob_fn` in the `squashed_gaussian`?
#54
fuyw
closed
2 years ago
2
Remove unconditional stop gradient bug in quantile q learning.
#53
copybara-service[bot]
closed
2 years ago
0
Remove the old venv directory before testing the package.
#52
copybara-service[bot]
closed
2 years ago
0
Bump ipython from 7.16.1 to 7.16.3 in /requirements
#51
dependabot[bot]
closed
1 year ago
1
Update requirements and allow new versions of JAX.
#50
copybara-service[bot]
closed
2 years ago
0
Add support for user defined number of splits in `tree_split_leaves` utility.
#49
copybara-service[bot]
closed
2 years ago
0
Change RLax citation to Jax Ecosystem citation.
#48
copybara-service[bot]
closed
2 years ago
0
Remove usages of apply_rng=True from Haiku code.
#47
copybara-service[bot]
closed
2 years ago
0
Add Sphinx build to CI test, point to documentation in README, and fix issues in doc strings that were causing CI test to fail.
#46
copybara-service[bot]
closed
2 years ago
0
Set up RLax sphinx documentation for readthedocs to build and serve documentation from the public github.
#45
copybara-service[bot]
closed
2 years ago
0
Add KNN Query to RLax public API.
#44
copybara-service[bot]
closed
2 years ago
0
Fix arg docstring for rho_tm1 and internal computations based on it to reflect time tm1 instead of t.
#43
copybara-service[bot]
closed
2 years ago
0
Drop python 3.6 support and release a new version.
#42
copybara-service[bot]
closed
2 years ago
0
Fix to ensure that policy loss gradients do not propagate into the temperature parameter if stop_gradient=True.
#41
copybara-service[bot]
closed
2 years ago
0
Create a new version 0.1.01 of RLax.
#40
copybara-service[bot]
closed
2 years ago
0
Drop python 3.6 support from v.0.1.0 per [JAX deprecation policy](https://www.google.com/url?sa=D&q=https%3A%2F%2Fjax.readthedocs.io%2Fen%2Flatest%2Fdeprecation.html). See [the issue](https://www.google.com/url?sa=D&q=https%3A%2F%2Fgithub.com%2Fdeepmind%2Foptax%2Fissues%2F222).
#39
copybara-service[bot]
closed
2 years ago
0
Internal change.
#38
copybara-service[bot]
closed
2 years ago
0
Need of central documentation
#37
EngineerKhan
closed
2 years ago
4
Previous
Next