facebookresearch/fairseq
1
1
Fork 0
mirror of https://github.com/facebookresearch/fairseq.git synced 2024-08-16 20:10:40 +03:00
Code Issues Packages Projects Releases Wiki Activity

Update getting_started.rst (#1188)

Browse Source 
Summary:
Hi,

I think there is a minor mistake in the doc. `--distributed-no-spawn` argument is needed for distributed training on multiple machines without `slurm`. Otherwise, the program will start 8 jobs on each GPU, when `nproc_per_node=8`.
Pull Request resolved: https://github.com/pytorch/fairseq/pull/1188

Differential Revision: D17627778

Pulled By: myleott

fbshipit-source-id: 35ab6b650dc1132d7cb2d150e80d2ebf0caf3e69
This commit is contained in:
Zhanghao Wu 2019-09-27 07:26:12 -07:00 committed by Facebook Github Bot
parent e073ddfe46
commit 2314979ea5
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
docs/getting_started.rst
View File

@ -189,4 +189,4 @@ replacing ``node_rank=0`` with ``node_rank=1`` on the second node:
--lr 0.0005 --min-lr 1e-09 \
--dropout 0.3 --weight-decay 0.0 --criterion label_smoothed_cross_entropy --label-smoothing 0.1 \
--max-tokens 3584 \
--fp16
--fp16 --distributed-no-spawn