Skip to content

Conversation

@vmoens
Copy link
Collaborator

@vmoens vmoens commented Nov 23, 2022

Description

Integrates the MPPI planner as presented in
https://arxiv.org/pdf/2203.04955.pdf

cc @nicklashansen

vmoens and others added 5 commits November 18, 2022 11:39
* amend

* [BugFix] ConvNet forward method with tensors of more than 4 dimensions (#686)

* cnn forward fix

* more general code

* cnn testing

* precommit run check

* convnet tests

* [Feature] add `standard_normal` for RewardScaling (#682)

* Add standard_normal

* give attribute access

* Update standard_normal

* Update tests

* Fix tests

* Address in-place scaling of reward

* Improvise tests

* [Feature] Jumanji envs (#674)

* amend

* [Feature] Default collate_fn (#688)

* init

* amend

* amend

* [BugFix] Fix Examples (#687)

* amend

* [Refactoring] Replace direct gym version checks with decorated functions (#691)

* [Refactoring] Replace gym version checking with decorated functions (#)

Initial commit. Only tests.

* Refactoring in gym.py

* More refactoring in gym.py

* Completed refactoring

* amend

* amend

* Version 0.0.3 (#696)

* [Docs] Host TensorDict docs inside TorchRL docs (#693)

* Pull tensordict docs into TorchRL docs

* Add banner for tensordict docs

* [BugFix] Fix docs build (#698)

* [BugFix] Proper error messages for orphan transform creation (#697)

* amend

* [Feature] Append, init and insert transforms in ReplayBuffer (#695)

* lint

Co-authored-by: albertbou92 <albertbou92@users.noreply.github.com>
Co-authored-by: Aditya Gandhamal <61016383+adityagandhamal@users.noreply.github.com>
Co-authored-by: yingchenlin <yc.jon.lin@gmail.com>
Co-authored-by: Sergey Ordinskiy <113687736+ordinskiy@users.noreply.github.com>
Co-authored-by: Tom Begley <tomcbegley@gmail.com>
Co-authored-by: Alan Schelten <alan@schelten.net>
@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 Nov 23, 2022
@vmoens vmoens marked this pull request as draft November 23, 2022 11:45
@vmoens vmoens added the new algo New algorithm request or PR label Nov 23, 2022
@codecov
Copy link

codecov bot commented Nov 23, 2022

Codecov Report

Merging #701 (590e56f) into main (e4e485b) will increase coverage by 0.01%.
The diff coverage is 98.05%.

❗ Current head 590e56f differs from pull request most recent head 953b186. Consider uploading reports for the commit 953b186 to get more accurate results

@@            Coverage Diff             @@
##             main     #701      +/-   ##
==========================================
+ Coverage   88.80%   88.81%   +0.01%     
==========================================
  Files         123      123              
  Lines       21270    21228      -42     
==========================================
- Hits        18888    18853      -35     
+ Misses       2382     2375       -7     
Flag Coverage Δ
habitat-gpu 24.41% <7.40%> (-0.37%) ⬇️
linux-brax ?
linux-cpu 84.92% <98.05%> (-0.32%) ⬇️
linux-gpu 85.79% <98.05%> (-0.41%) ⬇️
linux-jumanji 29.31% <7.40%> (-0.82%) ⬇️
linux-outdeps-gpu 72.18% <98.05%> (-0.16%) ⬇️
linux-stable-cpu 84.79% <98.05%> (-0.31%) ⬇️
linux-stable-gpu 85.47% <98.05%> (-0.39%) ⬇️
linux_examples-gpu 42.74% <7.40%> (+0.07%) ⬆️
macos-cpu 84.60% <98.05%> (-0.40%) ⬇️
olddeps-gpu 74.84% <98.05%> (-1.24%) ⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files Coverage Δ
torchrl/modules/planners/cem.py 97.95% <96.29%> (+0.34%) ⬆️
torchrl/modules/planners/mppi.py 98.21% <98.21%> (ø)
test/test_modules.py 99.42% <100.00%> (+0.06%) ⬆️
torchrl/data/replay_buffers/writers.py 78.72% <0.00%> (-16.02%) ⬇️
torchrl/modules/tensordict_module/sequence.py 90.95% <0.00%> (-9.05%) ⬇️
examples/dreamer/dreamer.py 87.66% <0.00%> (-3.25%) ⬇️
torchrl/objectives/sac.py 88.52% <0.00%> (-3.21%) ⬇️
torchrl/data/replay_buffers/samplers.py 83.33% <0.00%> (-2.88%) ⬇️
torchrl/trainers/trainers.py 75.00% <0.00%> (-2.67%) ⬇️
torchrl/objectives/value/advantages.py 85.71% <0.00%> (-2.62%) ⬇️
... and 67 more

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

# Conflicts:
#	test/test_modules.py
#	torchrl/modules/planners/cem.py
@vmoens vmoens changed the title [Algorithm] TD-MPC [Algorithm] MPPI planner Jan 5, 2023
@vmoens vmoens marked this pull request as ready for review January 5, 2023 14:14
@vmoens vmoens merged commit 4c3fda9 into main Jan 5, 2023
@vmoens vmoens deleted the tdmpc branch January 5, 2023 15:18
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