You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: src/lightning/pytorch/CHANGELOG.md
+19-19Lines changed: 19 additions & 19 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -22,53 +22,53 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
22
22
23
23
- Removed the limitation to call `self.trainer.model.parameters()` in `LightningModule.configure_optimizers()` ([#17309](https://github.com/Lightning-AI/lightning/pull/17309))
24
24
25
-
- Generalized `Optimizer` validation to accommodate both FSDP 1.x and 2.x ([#16733](https://github.com/Lightning-AI/lightning/pull/16733))
26
-
27
25
28
26
- Changes to the `NeptuneLogger` ([#16761](https://github.com/Lightning-AI/lightning/pull/16761)):
29
27
* It now supports neptune-client 0.16.16 and neptune >=1.0, and we have replaced the `log()` method with `append()` and `extend()`.
30
28
* It now accepts a namespace `Handler` as an alternative to `Run` for the `run` argument. This means that you can call it like `NeptuneLogger(run=run["some/namespace"])` to log everything to the `some/namespace/` location of the run.
31
29
32
-
-`Trainer(accelerator="tpu", devices=[i])"` now selects the i-th TPU core (0-based, previously it was 1-based) ([#17227](https://github.com/Lightning-AI/lightning/pull/17227))
33
-
34
-
35
-
- Pickling the `LightningModule` no longer pickles the `Trainer` ([#17133](https://github.com/Lightning-AI/lightning/pull/17133))
36
-
37
-
38
-
- Disable `torch.inference_mode` with `torch.compile` in PyTorch 2.0 ([#17215](https://github.com/Lightning-AI/lightning/pull/17215))
39
-
40
30
41
-
-Changed the `is_picklable` util function to handle the edge case that throws a `TypeError` ([#17270](https://github.com/Lightning-AI/lightning/pull/17270))
31
+
-`Trainer(accelerator="tpu", devices=[i])"` now selects the i-th TPU core (0-based, previously it was 1-based) ([#17227](https://github.com/Lightning-AI/lightning/pull/17227))
42
32
43
33
44
34
- Allow using iterable-style datasets with TPUs ([#17331](https://github.com/Lightning-AI/lightning/pull/17331))
45
35
46
-
### Depercated
36
+
### Deprecated
47
37
48
38
-
49
39
50
-
51
40
### Removed
52
41
53
42
-
54
43
55
-
56
44
### Fixed
57
45
58
-
- Fixed issue where pickling the module instance would fail with a DataLoader error ([#17130](https://github.com/Lightning-AI/lightning/pull/17130))
46
+
-
59
47
60
48
61
-
- Fixed WandbLogger not showing "best" aliases for model checkpoints when `ModelCheckpoint(save_top_k>0)` is used ([#17121](https://github.com/Lightning-AI/lightning/pull/17121))
49
+
## [2.0.1.post0] - 2023-04-11
62
50
51
+
### Fixed
63
52
64
-
-Fixed parsing the precision config for inference in `DeepSpeedStrategy` ([#16973](https://github.com/Lightning-AI/lightning/pull/16973))
53
+
-Changed the `is_picklable` util function to handle the edge case that throws a `TypeError` ([#17270](https://github.com/Lightning-AI/lightning/pull/17270))
65
54
66
55
67
-
- Fixed the availability check for `rich` that prevented Lightning to be imported in Google Colab ([#17156](https://github.com/Lightning-AI/lightning/pull/17156))
68
56
57
+
## [2.0.1] - 2023-03-30
69
58
70
-
- Fixed issue where `torch.compile` would fail when logging to WandB ([#17216](https://github.com/Lightning-AI/lightning/pull/17216))
59
+
### Changed
71
60
61
+
- Pickling the `LightningModule` no longer pickles the `Trainer` ([#17133](https://github.com/Lightning-AI/lightning/pull/17133))
62
+
- Generalized `Optimizer` validation to accommodate both FSDP 1.x and 2.x ([#16733](https://github.com/Lightning-AI/lightning/pull/16733))
63
+
- Disable `torch.inference_mode` with `torch.compile` in PyTorch 2.0 ([#17215](https://github.com/Lightning-AI/lightning/pull/17215))
64
+
65
+
### Fixed
66
+
67
+
- Fixed issue where pickling the module instance would fail with a DataLoader error ([#17130](https://github.com/Lightning-AI/lightning/pull/17130))
68
+
- Fixed WandbLogger not showing "best" aliases for model checkpoints when `ModelCheckpoint(save_top_k>0)` is used ([#17121](https://github.com/Lightning-AI/lightning/pull/17121))
69
+
- Fixed the availability check for `rich` that prevented Lightning to be imported in Google Colab ([#17156](https://github.com/Lightning-AI/lightning/pull/17156))
70
+
- Fixed parsing the precision config for inference in `DeepSpeedStrategy` ([#16973](https://github.com/Lightning-AI/lightning/pull/16973))
71
+
- Fixed issue where `torch.compile` would fail when logging to WandB ([#17216](https://github.com/Lightning-AI/lightning/pull/17216))
0 commit comments