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

[Feature] Log pbar rate in SOTA implementations #2662

Merged
merged 9 commits into from
Dec 18, 2024
Merged

[Feature] Log pbar rate in SOTA implementations #2662

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
a8c6615
Update
vmoens Dec 17, 2024
4f41d4d
Update
vmoens Dec 17, 2024
1a8360b
Update
vmoens Dec 17, 2024
e78ab4a
Update
vmoens Dec 17, 2024
7c044b3
Update
vmoens Dec 17, 2024
7444057
Update
vmoens Dec 17, 2024
ce2de43
Update
vmoens Dec 17, 2024
595808e
Update
vmoens Dec 17, 2024
ace76f3
Update
vmoens Dec 18, 2024
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
Update 
[ghstack-poisoned]
  • Loading branch information
@vmoens
vmoens committed Dec 17, 2024
commit 7444057268ca59bee5395ebec1aeeb85cc42467c
4 changes: 3 additions & 1 deletion sota-implementations/decision_transformer/dt.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -76,7 +76,9 @@ def main(cfg: "DictConfig"): # noqa: F821
loss_module = make_dt_loss(cfg.loss, actor, device=model_device)

# Create optimizer
transformer_optim, scheduler = make_dt_optimizer(cfg.optim, loss_module)
transformer_optim, scheduler = make_dt_optimizer(
cfg.optim, loss_module, model_device
)

# Create inference policy
inference_policy = DecisionTransformerInferenceWrapper(
Expand Down
4 changes: 2 additions & 2 deletions sota-implementations/decision_transformer/utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -511,10 +511,10 @@ def make_odt_optimizer(optim_cfg, loss_module):
return dt_optimizer, log_temp_optimizer, scheduler


def make_dt_optimizer(optim_cfg, loss_module):
def make_dt_optimizer(optim_cfg, loss_module, device):
dt_optimizer = torch.optim.Adam(
loss_module.actor_network_params.flatten_keys().values(),
lr=torch.as_tensor(optim_cfg.lr),
lr=torch.tensor(optim_cfg.lr, device=device),
weight_decay=optim_cfg.weight_decay,
eps=1.0e-8,
)
Expand Down
2 changes: 1 addition & 1 deletion sota-implementations/dqn/config_atari.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -42,5 +42,5 @@ loss:

compile:
compile: False
compile_mode:
compile_mode: default
cudagraphs: False
Loading