wassname/ray
1
0
Fork 0
mirror of https://github.com/wassname/ray.git synced 2026-08-12 12:20:11 +08:00
Code Issues Packages Projects Releases Wiki Activity

[tune/sgd] Document func_trainable and add checkpoint context (#9739)

Browse Source 
Co-authored-by: krfricke <krfricke@users.noreply.github.com>
Co-authored-by: Amog Kamsetty <amogkam@users.noreply.github.com>
This commit is contained in:
Richard Liaw
2020-07-30 09:46:37 -07:00
committed by GitHub
co-authored by krfricke Amog Kamsetty
parent e540e425e4
commit 0c3b9ebeef
23 changed files with 619 additions and 452 deletions
Download Patch File Download Diff File Expand all files Collapse all files
python/ray/util/sgd/torch/torch_trainer.py
+2
View File
@@ -140,6 +140,8 @@ class TorchTrainer:
Defaults to True.
wrap_ddp (bool): Whether to automatically wrap DistributedDataParallel
over each model. If False, you are expected to call it yourself.
timeout_s (float): Seconds before the torch process group
times out. Useful when machines are unreliable.
add_dist_sampler (bool): Whether to automatically add a
DistributedSampler to all created dataloaders. Only applicable
if num_workers > 1.