Revision 99f8010e483d1564144b8eedc29d8f5c432793c0 authored by Amanpreet Singh on 17 February 2021, 22:52:49 UTC, committed by Facebook GitHub Bot on 17 February 2021, 22:54:16 UTC
Summary: Pull Request resolved: https://github.com/facebookresearch/mmf/pull/780 This will allow specifying batch size per device instead of default global batch size. If specified using `training.batch_size_per_device`, this will supersede the value specified in `training.batch_size` and information will be logged. Reviewed By: madian9, vedanuj Differential Revision: D26480897 fbshipit-source-id: 9ffb3d9b79a84c39db4000e6a3d051e3bd1b6eb5
1 parent daeaa95
requirements.txt
torch==1.6.0
torchvision==0.7.0
numpy>=1.16.6
tqdm>=4.43.0
demjson==2.2.4
torchtext==0.5.0
GitPython==3.1.0
requests==2.23.0
fasttext==0.9.1
nltk==3.4.5
editdistance==0.5.3
transformers==3.4.0
sklearn==0.0
omegaconf==2.0.6
lmdb==0.98
termcolor==1.1.0
iopath==0.1.3
pytorch_lightning==1.1.6
Computing file changes ...
