karlinjfandJosh Karlin 41b6cbb3ca Don't copy the batch when training on a single gpu (#1576) ...

* fix

* whitespace

Co-authored-by: Josh Karlin <karlinjf@gmail.com>

2020-04-23 14:28:20 -04:00

..

callbacks

Default value for ModelCheckpoint filepath (#1548)

2020-04-23 11:50:58 -04:00

core

Add SLURM check in ddp_train() and init_ddp_connection() (#1387)

2020-04-19 17:08:19 -04:00

loggers

Improved docs for Loggers (#1484)

2020-04-16 12:04:12 -04:00

logging

add rank warning (#1428)

2020-04-09 14:05:46 -04:00

overrides

Add ddp_cpu backend for testing ddp without GPUs (#1158)

2020-04-15 23:17:31 -04:00

profiler

Profiler summary (#1259)

2020-03-31 08:57:48 -04:00

pt_overrides

add rank warning (#1428)

2020-04-09 14:05:46 -04:00

root_module

add rank warning (#1428)

2020-04-09 14:05:46 -04:00

trainer

Don't copy the batch when training on a single gpu (#1576)

2020-04-23 14:28:20 -04:00

utilities

Revert namespace package search to normal package search (#1545)

2020-04-21 08:26:47 -04:00

__init__.py

Update __init__.py

2020-04-23 11:04:57 -04:00