mirror of https://github.com/open-mmlab/mmyolo.git
19 lines
1.1 KiB
Markdown
19 lines
1.1 KiB
Markdown
|
# Set the random seed
|
||
|
|
|
||
|
|
If you want to set the random seed during training, you can use the following command.
|
||
|
|
|
||
|
|
```shell
|
||
|
|
python ./tools/train.py \
|
||
|
|
${CONFIG} \ # path of the config file
|
||
|
|
--cfg-options randomness.seed=2023 \ # set seed to 2023
|
||
|
|
[randomness.diff_rank_seed=True] \ # set different seeds according to global rank
|
||
|
|
[randomness.deterministic=True] # set the deterministic option for CUDNN backend
|
||
|
|
# [] stands for optional parameters, when actually entering the command line, you do not need to enter []
|
||
|
|
```
|
||
|
|
|
||
|
|
`randomness` has three parameters that can be set, with the following meanings.
|
||
|
|
|
||
|
|
- `randomness.seed=2023`, set the random seed to 2023.
|
||
|
|
- `randomness.diff_rank_seed=True`, set different seeds according to global rank. Defaults to False.
|
||
|
|
- `randomness.deterministic=True`, set the deterministic option for cuDNN backend, i.e., set `torch.backends.cudnn.deterministic` to True and `torch.backends.cudnn.benchmark` to False. Defaults to False. See https://pytorch.org/docs/stable/notes/randomness.html for more details.
|