a27fb5d54c
enhanced optimizer return options ( #120 )
...
* added smarter optimizer options
* added smarter optimizer options
* added smarter optimizer options tests
* added smarter optimizer options tests
* added smarter optimizer options tests
* added smarter optimizer options tests
* added smarter optimizer options tests
* added smarter optimizer options tests
* added smarter optimizer options tests
* added single gpu data transfer recursive
* added single gpu data transfer recursive
* added single gpu data transfer recursive
* added single gpu data transfer recursive
* added single gpu data transfer recursive
2019-08-15 11:31:56 -04:00
db9254acbe
enable recursive parsing for single gpu inputs ( #121 )
...
* added tests
* added single gpu data transfer recursive
* added single gpu data transfer recursive
* added single gpu data transfer recursive
* added single gpu data transfer recursive
* added single gpu data transfer recursive
* added single gpu data transfer recursive
2019-08-15 09:39:09 -04:00
2f984c9971
enable returning only opt list ( #114 )
2019-08-14 09:02:11 -04:00
7f53e7bfb3
Val idx optional in validation_step ( #108 )
...
* made dataset_i only available with multiple datasets
* updated interface signature
* updated tests
2019-08-13 11:37:37 -04:00
905a2e5a12
allow user to control optimizer step for every optimizer
...
* added custom hook for user defined optimizer step
* refactored to allow multiple optimizers different training_step
* refactored to allow multiple optimizers different training_step
* refactored to allow multiple optimizers different training_step
* refactored to allow multiple optimizers different training_step
* refactored to allow multiple optimizers different training_step
* pep8
2019-08-13 09:32:45 -04:00
5d5968033f
LR scheduler + train refactor ( #103 )
...
* split __train up for clarity
* split __train up for clarity
* added lr scheduler after epoch completes
2019-08-12 16:07:42 -04:00
511f7ecb9a
Support for multiple val_dataloaders ( #97 )
...
* Added support for multiple validation dataloaders
* Fix typo in README.md
* Update trainer.py
* Add support for multiple dataloaders
* Rename dataloader_index to dataloader_i
* Added warning to check val_dataloaders
Added a warning to ensure that all val_dataloaders were DistributedSamplers if ddp is enabled
* Updated DistributedSampler warning
* Fixed typo
* Added multiple val_dataloaders
* Multiple val_dataloader test
* Update lightning_module_template.py
Added dataloader_i to validation_step parameters
* Update trainer.py
* Reverted template changes
* Create multi_val_module.py
* Update no_val_end_module.py
* New MultiValModel
* Rename MultiValModel to MultiValTestModel
* Revert to LightningTestModel
* Update test_models.py
* Update trainer.py
* Update test_models.py
* multiple val_dataloaders in test template
* Fixed flake8 warnings
* Update trainer.py
* Fix flake errors
* Fixed Flake8 errors
* Update lm_test_module.py
keep this test model with a single dataset for val
* Update trainer.py
* Update trainer.py
* Update trainer.py
* Update trainer.py
* Update trainer.py
* Update test_models.py
* Update trainer.py
* Update trainer.py
* Update trainer.py
* Update trainer.py
* Update trainer.py
* Update trainer.py
* Update RequiredTrainerInterface.md
* Update RequiredTrainerInterface.md
* Update test_models.py
* Update trainer.py
dont need the else clause, val_dataloader is either a list or none because of get_dataloaders()
* Update trainer.py
fixed flake errors
* Update trainer.py
2019-08-12 15:23:11 -04:00
e5805bf8ff
val and test are optional now ( #95 )
...
* made validation step optional
* added no val model
* val_step can be implemented but not validation_end
* added no val end model
* added tests
* added tests
* remove class
* remove class
* remove class
* remove class
* remove class
* remove class
* remove class
* remove class
* remove class
* remove class
* remove class
* updated docs
* updated docs
* updated test
* updated test
* updated test
* updated test
* updated test
* updated test
* updated test
* updated test
* updated test
* fix pep8
2019-08-11 10:01:57 -04:00
996b1f9a6d
When running DDP without DistributedSampler, throw warning instead of exception ( #91 )
2019-08-10 15:58:12 -04:00
73d08557ba
fix accumulated grad norm fixes #87 ( #88 )
...
* Update trainer.py
* Update trainer.py
* Update trainer.py
* Update trainer.py
* Update trainer.py
* Update trainer.py
* Update trainer.py
2019-08-10 08:32:45 -04:00
8cd764a151
removed reduce on non-loss outputs from dp ( #78 )
...
* removed reduce on non-loss outputs from dp
* fixed val reduce
* fixed val reduce
* fixed val reduce
* fixed val reduce
2019-08-08 12:06:29 -04:00
3d23a56ed2
make experiment param in trainer optional ( #77 )
...
* removed forced exp
* modified test to also run without exp
2019-08-08 10:59:16 -04:00
aa7245d9db
Load fix ( #74 )
...
* skip weight load without callback
* added simple cpu test
* fixed pep
2019-08-08 06:00:04 -04:00
f3dea818f2
added single gpu train test
2019-08-07 13:54:32 -04:00
ab499573a1
added single gpu train test
2019-08-07 13:53:44 -04:00
56f16694c4
added single gpu train test
2019-08-07 13:50:12 -04:00
2f7a9ad40d
added single gpu train test
2019-08-07 13:49:01 -04:00
9ecb1f2aee
added single gpu train test
2019-08-07 13:46:06 -04:00
e30514922b
added single gpu train test
2019-08-07 13:43:28 -04:00
afa4548b12
added single gpu train
2019-08-07 13:39:40 -04:00
a2d6d514d5
cleaned up pep8 issues
2019-08-07 10:19:03 -04:00
10e4b18452
made imports absolute
2019-08-07 10:14:59 -04:00
35f23bbc82
Merge pull request #55 from williamFalcon/continue
...
add training restore
2019-08-07 09:02:16 -04:00
2f1df17371
updated test
2019-08-07 09:01:19 -04:00
d539e49085
updated test
2019-08-07 08:14:52 -04:00
df5e10d0aa
updated test
2019-08-07 08:12:54 -04:00
27e88fde31
debug
2019-08-07 07:57:03 -04:00
b8ce4adfa8
debug
2019-08-07 07:55:47 -04:00
cdbcbad352
added hook on_sanity_check_start
2019-08-07 07:51:55 -04:00
b0fae55571
fixed restore location
2019-08-07 07:42:14 -04:00
d5fd16a478
removed bad hook call
2019-08-07 07:38:45 -04:00
82d63a9677
fixed none name
2019-08-07 07:25:20 -04:00
a895bf1b71
fixed none name
2019-08-07 07:15:23 -04:00
d3f19c8321
added auto restore
2019-08-07 06:55:05 -04:00
97a9a0f6c1
Update trainer.py
2019-08-07 10:25:51 +02:00
a1bb6237a6
review changes #44
2019-08-06 22:45:46 +02:00
d9bfe964f9
update by flake8
2019-08-06 22:45:46 +02:00
632d07b490
fix prints for py3.5
2019-08-06 22:45:46 +02:00
778149735c
fix tests
2019-08-06 22:45:27 +02:00
c44966a8bf
apply PEP8
2019-08-06 22:45:27 +02:00
eacd93e2f0
fix rltv imports
2019-08-06 22:45:27 +02:00
f915e3657f
Merge pull request #42 from williamFalcon/warning
...
fixed clip grad warning
2019-08-05 14:18:57 -04:00
9af4267056
fixed clip grad warning
2019-08-05 14:01:10 -04:00
469941a528
pkg relative imports
...
* split requirements.txt
* pytest verbose
2019-08-05 10:52:09 +02:00
92f8c57ff5
cutout examples
2019-08-05 09:51:47 +02:00
40e5f1b4c8
formatting
2019-08-04 13:06:25 -05:00
1d78b40a00
Merge branch 'master' of https://github.com/williamFalcon/pytorch-lightning
2019-08-03 09:16:25 -05:00
ef6d5a412c
proc 0 only for save hpc. all procs for hpc load
2019-08-01 16:19:04 -04:00
5e4728b2a7
only proc 0 can submit a continuation slurm job
2019-08-01 10:15:28 -04:00
d372f9a2e2
updated dict keys
2019-07-28 11:46:26 -04:00
William Falcon