James Betker fba29d7dcc Move to apex distributeddataparallel and add switch all_reduce ... Torch's distributed_data_parallel is missing "delay_allreduce", which is necessary to get gradient checkpointing to work with recurrent models.		2020-10-08 11:20:05 -06:00
..
.idea
data	Update validate_data to work with SingleImageDataset	2020-10-02 08:58:34 -06:00
data_scripts	Import switched_conv as a submodule	2020-10-07 23:10:54 -06:00
metrics
models	Move to apex distributeddataparallel and add switch all_reduce	2020-10-08 11:20:05 -06:00
options	Move loaded_options to util	2020-10-03 20:29:06 -06:00
scripts
switched_conv@004dda04e3	Update switched_conv submodule	2020-10-07 23:11:50 -06:00
temp
utils	Clone and detach in recursively_detach	2020-10-07 12:41:00 -06:00
process_video.py	Add distributed_checkpoint for more efficient checkpoints	2020-10-06 20:38:38 -06:00
recover_tensorboard_log.py
requirements.txt
run_scripts.sh
test.py	Import switched_conv as a submodule	2020-10-07 23:10:54 -06:00
train.py	Move gpu_ids out of if statement	2020-10-06 20:40:20 -06:00
train2.py	Add concatenate injector	2020-10-07 09:02:42 -06:00