|
1 | 1 | # History |
2 | 2 |
|
3 | | -## 0.0.1 (2021-01-26) |
4 | | - |
5 | | -- First release on PyPI. |
6 | | - |
7 | | -## 0.2.0 (2021-02-07) |
8 | | - |
9 | | -- Fixed an issue with torch.clip and torch version |
10 | | -- Fixed an issue with `gpus` parameter in TrainerConfig, by setting default value to `None` for CPU |
11 | | -- Added feature to use custom sampler in the training dataloader |
12 | | -- Updated documentation and added a new tutorial for imbalanced classification |
13 | | - |
14 | | -## 0.3.0 (2021-03-02) |
15 | | - |
16 | | -- Fixed a bug on inference |
17 | | - |
18 | | -## 0.4.0 (2021-03-18) |
19 | | - |
20 | | -- Added AutoInt Model |
21 | | -- Added Mixture Density Networks |
22 | | -- Refactored the classes to separate backbones from the head of the models |
23 | | -- Changed the saving and loading model to work for custom parameters that you pass in `fit` |
| 3 | +## 1.0.1 (2023-01-20) |
24 | 4 |
|
25 | | -## 0.5.0 (2021-03-18) |
| 5 | +- Bugfix for default metric for binary classification |
26 | 6 |
|
27 | | -- Added more documentation |
28 | | -- Added Zenodo citation |
29 | 7 |
|
30 | | -## 0.6.0 (2021-06-21) |
31 | 8 |
|
32 | | -- Upgraded versions of PyTorch Lightning to 1.3.6 |
33 | | -- Changed the way `gpus` parameter is handled to avoid confusion. `None` is CPU, `-1` is all GPUs, `int` is number of GPUs |
34 | | -- Added a few more Trainer Params like `deterministic`, `auto_select_gpus` |
35 | | -- Some bug fixes and changes to docs |
36 | | -- Added `seed_everything` to the fit method to ensure reproducibility |
37 | | -- Refactored data_aware_initialization to be part of the BaseModel. Inherited Models can override the method to implement data aware initialization techniques |
38 | | - |
39 | | -## 0.7.0 (2021-09-01) |
40 | | - |
41 | | -- Implemented TabTransformer and FTTransformer models |
42 | | -- Included capability to save a model using GPU an load in CPU |
43 | | -- Made the temp folder pytorch tabular specific to avoid conflicts with other tmp folders. |
44 | | -- Some bug fixes |
45 | | -- Edited an error out of Advanced Tutorial in docs |
46 | 9 |
|
47 | 10 | ## 1.0.0 (2023-01-18) |
48 | 11 |
|
|
71 | 34 | - Many CI/CD improvements including complete integration with GitHub Actions. |
72 | 35 | - Upgraded all dependencies, including PyTorch Lightning, pandas, to latest versions and added dependabot to manage it going forward. |
73 | 36 | - Added pre-commit to ensure code integrity and standardization. |
| 37 | + |
| 38 | +## 0.7.0 (2021-09-01) |
| 39 | + |
| 40 | +- Implemented TabTransformer and FTTransformer models |
| 41 | +- Included capability to save a model using GPU an load in CPU |
| 42 | +- Made the temp folder pytorch tabular specific to avoid conflicts with other tmp folders. |
| 43 | +- Some bug fixes |
| 44 | +- Edited an error out of Advanced Tutorial in docs |
| 45 | + |
| 46 | +## 0.6.0 (2021-06-21) |
| 47 | + |
| 48 | +- Upgraded versions of PyTorch Lightning to 1.3.6 |
| 49 | +- Changed the way `gpus` parameter is handled to avoid confusion. `None` is CPU, `-1` is all GPUs, `int` is number of GPUs |
| 50 | +- Added a few more Trainer Params like `deterministic`, `auto_select_gpus` |
| 51 | +- Some bug fixes and changes to docs |
| 52 | +- Added `seed_everything` to the fit method to ensure reproducibility |
| 53 | +- Refactored data_aware_initialization to be part of the BaseModel. Inherited Models can override the method to implement data aware initialization techniques |
| 54 | + |
| 55 | +## 0.5.0 (2021-03-18) |
| 56 | + |
| 57 | +- Added more documentation |
| 58 | +- Added Zenodo citation |
| 59 | + |
| 60 | +## 0.4.0 (2021-03-18) |
| 61 | + |
| 62 | +- Added AutoInt Model |
| 63 | +- Added Mixture Density Networks |
| 64 | +- Refactored the classes to separate backbones from the head of the models |
| 65 | +- Changed the saving and loading model to work for custom parameters that you pass in `fit` |
| 66 | + |
| 67 | +## 0.3.0 (2021-03-02) |
| 68 | + |
| 69 | +- Fixed a bug on inference |
| 70 | + |
| 71 | +## 0.2.0 (2021-02-07) |
| 72 | + |
| 73 | +- Fixed an issue with torch.clip and torch version |
| 74 | +- Fixed an issue with `gpus` parameter in TrainerConfig, by setting default value to `None` for CPU |
| 75 | +- Added feature to use custom sampler in the training dataloader |
| 76 | +- Updated documentation and added a new tutorial for imbalanced classification |
| 77 | + |
| 78 | +## 0.0.1 (2021-01-26) |
| 79 | + |
| 80 | +- First release on PyPI. |
0 commit comments