[BC]Weighted bc training #416
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
inlining for size.
or neither of them is set. Further, passes keep_temps to ```compilation_runner.get_workdir_context``` and ensures that the flag and argument are not set at the same time.
explicit_temps_dir is not set and persistent_objects_path is set.
code to train a new policy and to save it as a tf-policy and code to compute the re-weighting for the supervised learning problem. This required updates to SequenceExampleFeatureNames from generate_bc_trajectories_lib.
mtrofin
approved these changes
Jan 16, 2025
| int_labels = tf.cast(labels, tf.int32) | ||
| return tf.gather(weights_arr, int_labels) | ||
|
|
||
| def _loss_fn(self, y_true, y_pred, labels, weights_arr): |
algorithm for training.
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Outdated suggestions cannot be applied.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.
Training code for the BC-Max algorithm which includes the tensorflow
code to train a new policy and to save it as a tf-policy and code to
compute the re-weighting for the supervised learning problem. This
required updates to
SequenceExampleFeatureNamesfromgenerate_bc_trajectories_lib.