Skip to content
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

[Algorithm] CrossQ #2033

Merged
merged 49 commits into from
Jul 10, 2024
Merged

[Algorithm] CrossQ #2033

merged 49 commits into from
Jul 10, 2024

Conversation

BY571
Copy link
Contributor

@BY571 BY571 commented Mar 21, 2024

Description

Adding CrossQ

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 #15213 if this solves the issue #15213

  • I have raised an issue to propose this change (required for new features and bug fixes)

Types of changes

What types of changes does your code introduce? Remove all that do not apply:

  • Bug fix (non-breaking change which fixes an issue)
  • New feature (non-breaking change which adds core functionality)
  • Breaking change (fix or feature that would cause existing functionality to change)
  • Documentation (update in the documentation)
  • Example (update in the folder of examples)

Checklist

Go over all the following points, and put an x in all the boxes that apply.
If you are unsure about any of these, don't hesitate to ask. We are here to help!

  • I have read the CONTRIBUTION guide (required)
  • My change requires a change to the documentation.
  • I have updated the tests accordingly (required for a bug fix or a new feature).
  • I have updated the documentation accordingly.

Copy link

pytorch-bot bot commented Mar 21, 2024

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2033

Note: Links to docs will display an error until the docs builds have been completed.

❌ 4 New Failures, 1 Unrelated Failure

As of commit c010e39 with merge base a151923 (image):

NEW FAILURES - The following jobs have failed:

BROKEN TRUNK - The following job failed but were present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

This comment was automatically generated by Dr. CI and updates every 15 minutes.

@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Mar 21, 2024
torchrl/objectives/crossq.py Outdated Show resolved Hide resolved
@BY571
Copy link
Contributor Author

BY571 commented Mar 21, 2024

Performance with separate target_computation looks good:

image

But we need to check for speed. It should be similar to our sac implementation.

@BY571 BY571 marked this pull request as ready for review March 26, 2024 18:40
@vmoens vmoens added the new algo New algorithm request or PR label Apr 8, 2024
# Conflicts:
#	.github/unittest/linux_examples/scripts/run_test.sh
Copy link
Contributor

@vmoens vmoens left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

thanks for this
There are just a couple of things to fix before merging

sota-implementations/crossq/crossq.py Outdated Show resolved Hide resolved
sota-implementations/crossq/crossq.py Outdated Show resolved Hide resolved
sota-implementations/crossq/crossq.py Show resolved Hide resolved
sota-implementations/crossq/utils.py Outdated Show resolved Hide resolved
sota-implementations/crossq/utils.py Outdated Show resolved Hide resolved
torchrl/objectives/crossq.py Show resolved Hide resolved
torchrl/objectives/crossq.py Outdated Show resolved Hide resolved
torchrl/objectives/crossq.py Outdated Show resolved Hide resolved
torchrl/objectives/crossq.py Outdated Show resolved Hide resolved
torchrl/objectives/crossq.py Show resolved Hide resolved
@vmoens
Copy link
Contributor

vmoens commented Jun 12, 2024

@BY571 we should also add it to the sota benchmarks

sota-implementations/crossq/batchrenorm.py Outdated Show resolved Hide resolved
import torch.nn as nn


class BatchRenorm(nn.Module):
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Let's put this in the modules no?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

and add it to the doc.
Happy to write a couple of tests.
Is it a copy paste? If so, can we check the license?

sota-implementations/crossq/batchrenorm.py Outdated Show resolved Hide resolved
sota-implementations/crossq/batchrenorm.py Outdated Show resolved Hide resolved
sota-implementations/crossq/batchrenorm.py Outdated Show resolved Hide resolved
torchrl/objectives/crossq.py Show resolved Hide resolved
torchrl/objectives/crossq.py Show resolved Hide resolved
torchrl/objectives/crossq.py Show resolved Hide resolved
next_tensordict.set(self.tensor_keys.action, next_action)
next_sample_log_prob = next_dist.log_prob(next_action)

# TODO: separate forward pass seems faster than the combined.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Keep?

torchrl/objectives/crossq.py Show resolved Hide resolved
loss_function="l2",
**kwargs,
)
sd = loss_fn.state_dict()
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

we should check that this contains the values of the actor and qvalue nets

torchrl/objectives/crossq.py Show resolved Hide resolved
torchrl/objectives/crossq.py Show resolved Hide resolved
torchrl/objectives/crossq.py Outdated Show resolved Hide resolved
torchrl/objectives/crossq.py Outdated Show resolved Hide resolved
torchrl/objectives/crossq.py Outdated Show resolved Hide resolved
torchrl/objectives/crossq.py Show resolved Hide resolved
@vmoens
Copy link
Contributor

vmoens commented Jul 9, 2024

@BY571 examples CI is failing:

  File "/pytorch/rl/sota-implementations/crossq/utils.py", line 244, in make_loss_module
    loss_module = CrossQLoss(
TypeError: __init__() got an unexpected keyword argument 'delay_actor'

@BY571
Copy link
Contributor Author

BY571 commented Jul 9, 2024

@vmoens updated :)

Copy link
Contributor

@vmoens vmoens left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM thanks a million for this

@vmoens vmoens merged commit a0a47a9 into pytorch:main Jul 10, 2024
59 of 64 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. new algo New algorithm request or PR
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants