-
Gym removed a method used by torchbench
https://github.com/pytorch/benchmark/blob/main/torchbenchmark/models/soft_actor_critic/envs.py#L237
https://github.com/openai/gym/commit/3a8daafce113c61…
-
TorchBench CI has detected a performance signal.
Affected Tests:
- eval-cuda-fp32:
- alexnet[disc (latency)] status changed, 8.717 -> NotExist
- alexnet[blade (latency)] status changed, 7.224 -> …
-
TorchBench CI has detected a performance signal.
Affected Tests:
- eval-cuda-fp32:
- alexnet[disc (latency)] status changed, 8.717 -> NotExist
- alexnet[blade (latency)] status changed, 7.224 -> …
-
### What happened + What you expected to happen
I have the offline data in the path_out
I use the offline data for both of training and evaluation (I know it is not right, but just for the seek of…
-
downloading models from download.pytorch.org is flaky, causing mobile builds to fail, inductor tests
ex https://github.com/pytorch/pytorch/actions/runs/4852615634/jobs/8648441371
```
TIMING: en…
-
TorchBench CI has detected a performance signal or runtime regression.
Base PyTorch commit: f3e13d956729d4fa9ce35dca5beed588c914797d
Affected PyTorch commit: ed0910596db6fe7e40eff2d6681535cfe68cad4…
-
I have just gone through the paper "PowerGym: A Reinforcement Learning Environment for Volt-Var Control in Power Distribution Systems". I have found it insightful and thanks for sharing this reposito…
-
Gym deprecated the method `env.seed` and needs to use `env.reset(seed=seed)` instead
The method `end.seed` was removed in https://github.com/openai/gym/commit/3a8daafce113c6161190c0252c6a3a340cc38f…
-
Haarnoja, Tuomas et al.
https://arxiv.org/abs/1801.01290
-
https://github.com/pfnet/pfrl/blob/70f3da9163c047fa6981047e74836f1e138694e0/pfrl/agents/ddpg.py#L258-L262
Here, `self.policy` is of type nn.Module, so calling `self.policy(batch_xs)` returns a pytorc…