Skip to content

Commit 7e57581

Browse files
SeanNarenlexierule
SeanNaren
authored and
lexierule
committed
v1.2.8 & Update CHANGELOG.md
1 parent 75bcb91 commit 7e57581

File tree

2 files changed

+15
-211
lines changed

2 files changed

+15
-211
lines changed

‎CHANGELOG.md

Lines changed: 14 additions & 210 deletions
Original file line numberDiff line numberDiff line change
@@ -5,226 +5,31 @@ All notable changes to this project will be documented in this file.
55
The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
66

77

8-
## [1.3.0] - 2021-MM-DD
8+
## [1.2.8] - 2021-04-14
99

1010
### Added
1111

12-
- Added utils for NaN/Inf detection for gradients and parameters ([#6834](https://github.com/PyTorchLightning/pytorch-lightning/pull/6834/))
13-
14-
15-
- Added more explicit exception message when trying to execute `trainer.test()` or `trainer.validate()` with `fast_dev_run=True` ([#6667](https://github.com/PyTorchLightning/pytorch-lightning/pull/6667))
16-
17-
18-
- Added `LightningCLI` class to provide simple reproducibility with minimum boilerplate training cli. ([#4492](https://github.com/PyTorchLightning/pytorch-lightning/pull/4492))
19-
20-
21-
- Trigger warning when non-metric logged value with multi processes hasn't been reduced ([#6417](https://github.com/PyTorchLightning/pytorch-lightning/pull/6417))
22-
23-
24-
- Added `gradient_clip_algorithm` argument to Trainer for gradient clipping by value ([#6123](https://github.com/PyTorchLightning/pytorch-lightning/pull/6123)).
25-
26-
27-
- Added a way to print to terminal without breaking up the progress bar ([#5470](https://github.com/PyTorchLightning/pytorch-lightning/pull/5470))
28-
29-
30-
- Added support to checkpoint after training steps in `ModelCheckpoint` callback ([#6146](https://github.com/PyTorchLightning/pytorch-lightning/pull/6146))
31-
32-
33-
- Added `checkpoint` parameter to callback's `on_save_checkpoint` hook ([#6072](https://github.com/PyTorchLightning/pytorch-lightning/pull/6072))
34-
35-
36-
- Added arg to `self.log` that enables users to give custom names when dealing with multiple dataloaders ([#6274](https://github.com/PyTorchLightning/pytorch-lightning/pull/6274))
37-
38-
39-
- Added `teardown` method to `BaseProfiler` to enable subclasses defining post-profiling steps outside of `__del__` ([#6370](https://github.com/PyTorchLightning/pytorch-lightning/pull/6370))
40-
41-
42-
- Added `setup` method to `BaseProfiler` to enable subclasses defining pre-profiling steps for every process ([#6633](https://github.com/PyTorchLightning/pytorch-lightning/pull/6633))
43-
44-
45-
- Added no return warning to predict ([#6139](https://github.com/PyTorchLightning/pytorch-lightning/pull/6139))
46-
47-
48-
- Added `Trainer.predict` config validation ([#6543](https://github.com/PyTorchLightning/pytorch-lightning/pull/6543))
49-
50-
51-
- Added `AbstractProfiler` interface ([#6621](https://github.com/PyTorchLightning/pytorch-lightning/pull/6621))
52-
53-
54-
- Added support for including module names for forward in the autograd trace of `PyTorchProfiler` ([#6349](https://github.com/PyTorchLightning/pytorch-lightning/pull/6349))
55-
56-
57-
- Added support for the PyTorch 1.8.1 autograd profiler ([#6618](https://github.com/PyTorchLightning/pytorch-lightning/pull/6618))
58-
59-
60-
- Added `outputs` parameter to callback's `on_validation_epoch_end` & `on_test_epoch_end` hooks ([#6120](https://github.com/PyTorchLightning/pytorch-lightning/pull/6120))
61-
62-
63-
- Added `configure_sharded_model` hook ([#6679](https://github.com/PyTorchLightning/pytorch-lightning/pull/6679))
64-
65-
66-
- Added support for `precision=64`, enabling training with double precision ([#6595](https://github.com/PyTorchLightning/pytorch-lightning/pull/6595))
67-
68-
69-
- Added support for DDP communication hooks ([#6736](https://github.com/PyTorchLightning/pytorch-lightning/issues/6736))
70-
71-
72-
- Added `artifact_location` argument to `MLFlowLogger` which will be passed to the `MlflowClient.create_experiment` call ([#6677](https://github.com/PyTorchLightning/pytorch-lightning/pull/6677))
73-
74-
75-
- Added `model` parameter to precision plugins' `clip_gradients` signature ([#6764](https://github.com/PyTorchLightning/pytorch-lightning/pull/6764))
76-
77-
78-
### Changed
79-
80-
- Renamed `pytorch_lightning.callbacks.swa` to `pytorch_lightning.callbacks.stochastic_weight_avg` ([#6259](https://github.com/PyTorchLightning/pytorch-lightning/pull/6259))
81-
82-
83-
- Refactor `RunningStage` and `TrainerState` usage ([#4945](https://github.com/PyTorchLightning/pytorch-lightning/pull/4945))
84-
85-
86-
- Changed `trainer.evaluating` to return `True` if validating or testing ([#4945](https://github.com/PyTorchLightning/pytorch-lightning/pull/4945))
87-
88-
89-
- Changed `setup()` and `teardown()` stage argument to take any of `{fit,validate,test,predict}` ([#6386](https://github.com/PyTorchLightning/pytorch-lightning/pull/6386))
90-
91-
92-
- Changed profilers to save separate report files per state and rank ([#6621](https://github.com/PyTorchLightning/pytorch-lightning/pull/6621))
93-
94-
95-
- Changed `PyTorchProfiler` to use `torch.autograd.profiler.record_function` to record functions ([#6349](https://github.com/PyTorchLightning/pytorch-lightning/pull/6349))
96-
97-
98-
### Deprecated
99-
100-
- Deprecated `TrainerTrainingTricksMixin` in favor of a separate utilities module for NaN/Inf detection for gradients and parameters ([#6834](https://github.com/PyTorchLightning/pytorch-lightning/pull/6834/))
101-
102-
103-
- `period` has been deprecated in favor of `every_n_val_epochs` in the `ModelCheckpoint` callback ([#6146](https://github.com/PyTorchLightning/pytorch-lightning/pull/6146))
104-
105-
106-
- Deprecated `trainer.running_sanity_check` in favor of `trainer.sanity_checking` ([#4945](https://github.com/PyTorchLightning/pytorch-lightning/pull/4945))
107-
108-
109-
- Deprecated `Profiler(output_filename)` in favor of `dirpath` and `filename` ([#6621](https://github.com/PyTorchLightning/pytorch-lightning/pull/6621))
110-
111-
112-
- Deprecated `PytorchProfiler(profiled_functions)` in favor of `record_functions` ([#6349](https://github.com/PyTorchLightning/pytorch-lightning/pull/6349))
113-
114-
115-
- Deprecated metrics in favor of `torchmetrics` ([#6505](https://github.com/PyTorchLightning/pytorch-lightning/pull/6505),
116-
[#6530](https://github.com/PyTorchLightning/pytorch-lightning/pull/6530),
117-
[#6540](https://github.com/PyTorchLightning/pytorch-lightning/pull/6540),
118-
[#6547](https://github.com/PyTorchLightning/pytorch-lightning/pull/6547),
119-
[#6515](https://github.com/PyTorchLightning/pytorch-lightning/pull/6515),
120-
[#6572](https://github.com/PyTorchLightning/pytorch-lightning/pull/6572),
121-
[#6573](https://github.com/PyTorchLightning/pytorch-lightning/pull/6573),
122-
[#6584](https://github.com/PyTorchLightning/pytorch-lightning/pull/6584),
123-
[#6636](https://github.com/PyTorchLightning/pytorch-lightning/pull/6636),
124-
[#6637](https://github.com/PyTorchLightning/pytorch-lightning/pull/6637),
125-
[#6649](https://github.com/PyTorchLightning/pytorch-lightning/pull/6649),
126-
[#6659](https://github.com/PyTorchLightning/pytorch-lightning/pull/6659),
127-
)
128-
129-
130-
### Removed
131-
132-
- Removed support for passing a bool value to `profiler` argument of Trainer ([#6164](https://github.com/PyTorchLightning/pytorch-lightning/pull/6164))
133-
134-
135-
- Removed no return warning from val/test step ([#6139](https://github.com/PyTorchLightning/pytorch-lightning/pull/6139))
136-
137-
138-
- Removed passing a `ModelCheckpoint` instance to `Trainer(checkpoint_callback)` ([#6166](https://github.com/PyTorchLightning/pytorch-lightning/pull/6166))
139-
140-
141-
- Removed deprecated Trainer argument `enable_pl_optimizer` and `automatic_optimization` ([#6163](https://github.com/PyTorchLightning/pytorch-lightning/pull/6163))
142-
143-
144-
- Removed deprecated metrics ([#6161](https://github.com/PyTorchLightning/pytorch-lightning/pull/6161))
145-
* from `pytorch_lightning.metrics.functional.classification` removed `to_onehot`, `to_categorical`, `get_num_classes`, `roc`, `multiclass_roc`, `average_precision`, `precision_recall_curve`, `multiclass_precision_recall_curve`
146-
* from `pytorch_lightning.metrics.functional.reduction` removed `reduce`, `class_reduce`
147-
148-
149-
- Removed deprecated `ModelCheckpoint` arguments `prefix`, `mode="auto"` ([#6162](https://github.com/PyTorchLightning/pytorch-lightning/pull/6162))
150-
151-
152-
- Removed `mode='auto'` from `EarlyStopping` ([#6167](https://github.com/PyTorchLightning/pytorch-lightning/pull/6167))
153-
154-
155-
- Removed legacy references for magic keys in the `Result` object ([#6016](https://github.com/PyTorchLightning/pytorch-lightning/pull/6016))
156-
157-
158-
- Removed deprecated `LightningModule` `hparams` setter ([#6207](https://github.com/PyTorchLightning/pytorch-lightning/pull/6207))
159-
160-
161-
- Removed legacy code to log or include metrics in the progress bar by returning them in a dict with the `"log"/"progress_bar"` magic keys. Use `self.log` instead ([#6734](https://github.com/PyTorchLightning/pytorch-lightning/pull/6734))
162-
163-
164-
- Removed `optimizer_idx` argument from `training_step` in manual optimization ([#6093](https://github.com/PyTorchLightning/pytorch-lightning/pull/6093))
165-
12+
- Added TPUSpawn + IterableDataset error message ([#6875](https://github.com/PyTorchLightning/pytorch-lightning/pull/6875))
16613

16714
### Fixed
16815

16+
- Fixed process rank not being available right away after `Trainer` instantiation ([#6941](https://github.com/PyTorchLightning/pytorch-lightning/pull/6941))
17+
- Fixed `sync_dist` for tpus ([#6950](https://github.com/PyTorchLightning/pytorch-lightning/pull/6950))
18+
- Fixed `AttributeError` for `require_backward_grad_sync` when running manual optimization with sharded plugin ([#6915](https://github.com/PyTorchLightning/pytorch-lightning/pull/6915))
19+
- Fixed `--gpus` default for parser returned by `Trainer.add_argparse_args` ([#6898](https://github.com/PyTorchLightning/pytorch-lightning/pull/6898))
20+
- Fixed TPU Spawn all gather ([#6896](https://github.com/PyTorchLightning/pytorch-lightning/pull/6896))
21+
- Fixed `EarlyStopping` logic when `min_epochs` or `min_steps` requirement is not met ([#6705](https://github.com/PyTorchLightning/pytorch-lightning/pull/6705))
22+
- Fixed csv extension check ([#6436](https://github.com/PyTorchLightning/pytorch-lightning/pull/6436))
23+
- Fixed checkpoint issue when using Horovod distributed backend ([#6958](https://github.com/PyTorchLightning/pytorch-lightning/pull/6958))
24+
- Fixed tensorboard exception raising ([#6901](https://github.com/PyTorchLightning/pytorch-lightning/pull/6901))
25+
- Fixed setting the eval/train flag correctly on accelerator model ([#6983](https://github.com/PyTorchLightning/pytorch-lightning/pull/6983))
26+
- Fixed DDP_SPAWN compatibility with bug_report_model.py ([#6892](https://github.com/PyTorchLightning/pytorch-lightning/pull/6892))
27+
- Fixed bug where `BaseFinetuning.flatten_modules()` was duplicating leaf node parameters ([#6879](https://github.com/PyTorchLightning/pytorch-lightning/pull/6879))
16928
- Set better defaults for `rank_zero_only.rank` when training is launched with SLURM and torchelastic:
17029
* Support SLURM and torchelastic global rank environment variables ([#5715](https://github.com/PyTorchLightning/pytorch-lightning/pull/5715))
17130
* Remove hardcoding of local rank in accelerator connector ([#6878](https://github.com/PyTorchLightning/pytorch-lightning/pull/6878))
17231

17332

174-
- Sanitize `None` params during pruning ([#6836](https://github.com/PyTorchLightning/pytorch-lightning/pull/6836))
175-
176-
177-
- Made the `Plugin.reduce` method more consistent across all Plugins to reflect a mean-reduction by default ([#6011](https://github.com/PyTorchLightning/pytorch-lightning/pull/6011))
178-
179-
180-
- Move lightning module to correct device type when using LightningDistributedWrapper ([#6070](https://github.com/PyTorchLightning/pytorch-lightning/pull/6070))
181-
182-
183-
- Do not print top-k verbose log with `ModelCheckpoint(monitor=None)` ([#6109](https://github.com/PyTorchLightning/pytorch-lightning/pull/6109))
184-
185-
186-
- Fixed csv extension check ([#6436](https://github.com/PyTorchLightning/pytorch-lightning/pull/6436))
187-
188-
189-
- Fixed `ModelCheckpoint(monitor=None, save_last=True)` not saving checkpoints ([#6136](https://github.com/PyTorchLightning/pytorch-lightning/pull/6136))
190-
191-
192-
- Fixed `ModelCheckpoint(save_top_k=0, save_last=True)` not saving the `last` checkpoint ([#6136](https://github.com/PyTorchLightning/pytorch-lightning/pull/6136))
193-
194-
195-
- Fixed `.teardown(stage='fit')` getting called during `trainer.test` ([#6386](https://github.com/PyTorchLightning/pytorch-lightning/pull/6386))
196-
197-
198-
- Fixed `.on_fit_{start,end}()` getting called during `trainer.test` ([#6386](https://github.com/PyTorchLightning/pytorch-lightning/pull/6386))
199-
200-
201-
- Fixed LightningModule `all_gather` on cpu tensors ([#6416](https://github.com/PyTorchLightning/pytorch-lightning/pull/6416))
202-
203-
204-
- Fixed torch distributed not available in setup hook for DDP ([#6506](https://github.com/PyTorchLightning/pytorch-lightning/pull/6506))
205-
206-
207-
- Fixed bug where `BaseFinetuning.flatten_modules()` was duplicating leaf node parameters ([#6879](https://github.com/PyTorchLightning/pytorch-lightning/pull/6879))
208-
209-
210-
- Fixed `EarlyStopping` logic when `min_epochs` or `min_steps` requirement is not met ([#6705](https://github.com/PyTorchLightning/pytorch-lightning/pull/6705))
211-
212-
213-
- Fixed TPU Spawn all gather ([#6896](https://github.com/PyTorchLightning/pytorch-lightning/pull/6896))
214-
215-
216-
- Fixed `--gpus` default for parser returned by `Trainer.add_argparse_args` ([#6898](https://github.com/PyTorchLightning/pytorch-lightning/pull/6898))
217-
218-
219-
- Fixed `AttributeError for `require_backward_grad_sync` when running manual optimization with sharded plugin ([#6915](https://github.com/PyTorchLightning/pytorch-lightning/pull/6915))
220-
221-
222-
- Fixed `sync_dist` for tpus ([#6950](https://github.com/PyTorchLightning/pytorch-lightning/pull/6950))
223-
224-
225-
- Fixed process rank not being available right away after `Trainer` instantiation ([#6941](https://github.com/PyTorchLightning/pytorch-lightning/pull/6941))
226-
227-
22833
## [1.2.7] - 2021-04-06
22934

23035
### Fixed
@@ -336,7 +141,6 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
336141
- Fixed incorrect yield logic for the amp autocast context manager ([#6080](https://github.com/PyTorchLightning/pytorch-lightning/pull/6080))
337142
- Fixed priority of plugin/accelerator when setting distributed mode ([#6089](https://github.com/PyTorchLightning/pytorch-lightning/pull/6089))
338143
- Fixed error message for AMP + CPU incompatibility ([#6107](https://github.com/PyTorchLightning/pytorch-lightning/pull/6107))
339-
- Disabled batch transfer in DP mode ([#6093](https://github.com/PyTorchLightning/pytorch-lightning/pull/6093))
340144

341145

342146
## [1.2.0] - 2021-02-18

‎pytorch_lightning/info.py

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,7 +1,7 @@
11
import time
22

33
_this_year = time.strftime("%Y")
4-
__version__ = '1.2.7'
4+
__version__ = '1.2.8'
55
__author__ = 'William Falcon et al.'
66
__author_email__ = '[email protected]'
77
__license__ = 'Apache-2.0'

0 commit comments

Comments
 (0)