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.

Sign up for GitHub

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

Jump to bottom

[RLlib] Don't add a cpu to bundle for learner when using gpu #35529

Merged
merged 17 commits into from
May 23, 2023
Merged

[RLlib] Don't add a cpu to bundle for learner when using gpu #35529

Changes from 1 commit
Commits
Show all changes
17 commits
Select commit Hold shift + click to select a range
bce003e
Initial commit
avnishn May 18, 2023
1a0c244
Don't add a cpu to bundle for learner when using gpu
avnishn May 18, 2023
0b75bf6
Fix indentation
avnishn May 18, 2023
d7ab6e4
Fix indentation
avnishn May 19, 2023
9ba46de
Move ppo kl infinite check to avoid tensorflow autograph errors
avnishn May 21, 2023
162470a
Release test cluster only has 2 gpus not 4, reduce number of gpus to …
avnishn May 21, 2023
668899d
Address comments
avnishn May 21, 2023
517f5c1
Add resource resolution back in
avnishn May 22, 2023
8a7954d
Merge branch 'master' of https://github.com/ray-project/ray into modi…
avnishn May 22, 2023
1a7820e
Make level with master, fix docstring
avnishn May 22, 2023
fa97a46
Configure the mulit agent cartpole ppo example to use resources more …
avnishn May 22, 2023
a81170f
Address lint
avnishn May 22, 2023
b2804d0
Reduce num cpus used for ci
avnishn May 22, 2023
c496c34
Move kl check out of traced update, torch, tf
avnishn May 22, 2023
b3b7c8f
Reduce num rollout workers
avnishn May 23, 2023
b2c947f
Merge branch 'master' of https://github.com/ray-project/ray into modi…
avnishn May 23, 2023
d199063
Remove inf kl changes in favor of svens
avnishn May 23, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
Release test cluster only has 2 gpus not 4, reduce number of gpus to … 
…avoid hanging experiment

Signed-off-by: Avnish <[email protected]>
  • Loading branch information
@avnishn
avnishn committed May 21, 2023
commit 162470a28a8e9f35d703216b7508d822cf7e8343
4 changes: 2 additions & 2 deletions release/rllib_tests/smoke_tests/smoke_test_basic_multi_node_training_learner.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,7 +36,7 @@ def run_tuner_n_rollout_workers_0_gpu(config):
config = config.rollouts(num_rollout_workers=5)
config = config.resources(
num_cpus_per_learner_worker=1,
num_learner_workers=4,
num_learner_workers=2,
)

tuner = tune.Tuner(
Expand Down Expand Up @@ -94,7 +94,7 @@ def test_multi_node_training_smoke():
.environment("CartPole-v1")
.resources(
num_gpus_per_learner_worker=1,
num_learner_workers=4,
num_learner_workers=2,
)
.rollouts(num_rollout_workers=2)
.reporting(min_time_s_per_iteration=0, min_sample_timesteps_per_iteration=10)
Expand Down