.. |
environments
|
(1/n) tests: Use strategy flag instead of accelerator for training strategies (#9931)
|
2021-10-16 20:40:25 +05:30 |
__init__.py
|
enable ddp as a plugin (#4285)
|
2020-10-22 05:15:51 -04:00 |
test_amp_plugins.py
|
Use `torch.autocast` (#10053)
|
2021-10-25 17:33:52 +00:00 |
test_checkpoint_io_plugin.py
|
Update strategy flag in docs (#10000)
|
2021-10-20 21:02:53 +05:30 |
test_cluster_integration.py
|
Update strategy flag in docs (#10000)
|
2021-10-20 21:02:53 +05:30 |
test_custom_plugin.py
|
Unify checkpoint load paths [redo #9693] (#10061)
|
2021-10-25 19:05:31 +00:00 |
test_ddp_fully_sharded_with_full_state_dict.py
|
Minor fixes related to clipping (#10130)
|
2021-10-25 16:40:22 +00:00 |
test_ddp_plugin.py
|
Update strategy flag in docs (#10000)
|
2021-10-20 21:02:53 +05:30 |
test_ddp_plugin_with_comm_hook.py
|
Update strategy flag in docs (#10000)
|
2021-10-20 21:02:53 +05:30 |
test_ddp_spawn_plugin.py
|
(1/n) tests: Use strategy flag instead of accelerator for training strategies (#9931)
|
2021-10-16 20:40:25 +05:30 |
test_deepspeed_plugin.py
|
Unify checkpoint load paths [redo #9693] (#10061)
|
2021-10-25 19:05:31 +00:00 |
test_double_plugin.py
|
(1/n) tests: Use strategy flag instead of accelerator for training strategies (#9931)
|
2021-10-16 20:40:25 +05:30 |
test_plugins_registry.py
|
Update strategy flag in docs (#10000)
|
2021-10-20 21:02:53 +05:30 |
test_sharded_plugin.py
|
Unify checkpoint load paths [redo #9693] (#10061)
|
2021-10-25 19:05:31 +00:00 |
test_single_device_plugin.py
|
Replace `yapf` with `black` (#7783)
|
2021-07-26 13:37:35 +02:00 |
test_tpu_spawn.py
|
Mark `trainer.data_connector` as protected (#10031)
|
2021-10-25 12:29:09 +01:00 |