charlie880624/STDC
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[Fix] Fix iter bug when resuming checkpoint in distributed train (#866)

Browse Source 
* [Fix] Fix iter bug when resuming checkpoint in distributed train

* fix lint error

Signed-off-by: FreyWang <wangwxyz@qq.com>
This commit is contained in:
FreyWang 2021-09-11 14:33:01 +08:00 committed by GitHub
parent 872e54497e
commit c4c2fdcff3
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 4 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
tools/train.py
View File

@ -7,7 +7,7 @@ import time
import mmcv
import torch
from mmcv.runner import init_dist
from mmcv.runner import get_dist_info, init_dist
from mmcv.utils import Config, DictAction, get_git_hash
from mmseg import __version__
@ -94,6 +94,9 @@ def main():
else:
distributed = True
init_dist(args.launcher, **cfg.dist_params)
# gpu_ids is used to calculate iter when resuming checkpoint,
_, world_size = get_dist_info()
cfg.gpu_ids = range(world_size)
# create work_dir
mmcv.mkdir_or_exist(osp.abspath(cfg.work_dir))