[Algorithm] Update TD3 Example #1523
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
facebook-github-bot
added
the
CLA Signed
This was referenced Sep 14, 2023
vmoens
approved these changes
Sep 14, 2023
vmoens
reviewed
Sep 20, 2023
BY571
commented
Sep 21, 2023
test/test_cost.py
Outdated
| for i in loss_val: | ||
| assert i in loss_val_td.values(), f"{i} not in {loss_val_td.values()}" | ||
| # for i, key in enumerate(loss_val_td.keys()): | ||
| # torch.testing.assert_close(loss_val_td.get(key), loss_val[i]) |
There was a problem hiding this comment.
This is dangerous as keys in the tensordict get ordered by name but output tuple loss_val doesn't. So for now im just checking if all values in the loss_val tuple are also in the loss_val_td.
# actor metadata
metadata = {
"state_action_value_actor": state_action_value_actor.mean().detach(),
}
# value metadata
metadata = {
"td_error": td_error,
"next_state_value": next_target_qvalue.mean().detach(),
"pred_value": current_qvalue.mean().detach(),
"target_value": target_value.mean().detach(),
}
# out tensordict
td_out = TensorDict(
source={
"loss_actor": loss_actor,
"loss_qvalue": loss_qval,
**metadata_actor,
**metadata_value,
},
batch_size=[],
)
loss_vals will be in that order (loss_actor, loss_qvalue, state_action_value_actor, next_state_value, pred_value, target_value)
However, as the items are getting ordered in the TD by the keys the output tensordict has actually this order:
(loss_actor, loss_qvalue, next_state_value, pred_value, state_action_value_actor, target_value)
There was a problem hiding this comment.
dispatch returns the keys in the order of out_keys.
So this is predictable, we can just do
for i, key in enumerate(loss.out_keys):
torch.testing.assert_close(loss_val_td.get(key), loss_val[i])does that solve the problem?
# Conflicts: # examples/td3/utils.py
vmoens
approved these changes
Oct 3, 2023
vmoens
pushed a commit
to hyerra/rl
that referenced
this pull request
Oct 10, 2023
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
Updated TD3 script similar to PPO Update, added time logging, more comments, cleaner structure, and fixes here and there. Running some tests right now to verify performance.
What do you think @vmoens, @albertbou92 @matteobettini, how could we improve the example further?
Motivation and Context
Why is this change required? What problem does it solve?
If it fixes an open issue, please link to the issue here.
You can use the syntax
close #15213if this solves the issue #15213Types of changes
What types of changes does your code introduce? Remove all that do not apply:
Checklist
Go over all the following points, and put an
xin all the boxes that apply.If you are unsure about any of these, don't hesitate to ask. We are here to help!