# pytorch extension for [pandas-ml-utils][ghl1]
Adds a `PytorchModel` to the pandas ml utils suite. While a regular class extending `nn.Module` is sufficient,
there is also a special class `PytorchNN` which can be extended as well. Using this class has the following
advantages:
* allows to use L1, L2 regularization -> [example][ghl2]
* allows different forward path for training and prediction (useful i.e. for reparameterization trick) -> [example][ghl3]
* allows to implement auto-encoders easily by just providing the encode/decode functions
* added loss functions like `SoftDTW` (fit time series) loss or `HeteroscedasticityLoss` (fit Normal Distribution) -> [example][ghl3]
<br/><br/>
![Fitting Example][ghi1]
[ghl1]: https://github.com/KIC/pandas-ml-quant/tree/0.2.1/pandas-ml-utils-torch/../pandas-ml-utils
[ghl2]: https://github.com/KIC/pandas-ml-quant/tree/0.2.1/pandas-ml-utils-torch/./examples/regression_with_regularization.ipynb
[ghl3]: https://github.com/KIC/pandas-ml-quant/tree/0.2.1/pandas-ml-utils-torch/./examples/probabilistic_model.ipynb
[ghi1]: https://github.com/KIC/pandas-ml-quant/raw/0.2.1/pandas-ml-utils-torch/../.readme/videos/probabilistic-model-fit.gif
pandas-ml-utils-torch-0.2.1.tar.gz
需积分: 1 104 浏览量
2024-03-07
12:44:43
上传
评论
收藏 36KB GZ 举报
pandas-ml-utils-torch-0.2.1.tar.gz （59个子文件） 
pandas-ml-utils-torch-0.2.1 
noxfile.py 2KB setup.py 2KB PKG-INFO 2KB pandas_ml_utils_torch utils.py 747B __init__.py 256B loss __init__.py 248B simple_losses.py 2KB soft_dtw.py 3KB classificaion.py 3KB __pycache__ classificaion.cpython-38.pyc 3KB simple_losses.cpython-38.pyc 3KB soft_dtw.cpython-38.pyc 4KB classificaion.cpython-37.pyc 3KB probabilistic.cpython-37.pyc 1KB simple_losses.cpython-37.pyc 3KB __init__.cpython-37.pyc 511B probabilistic.cpython-38.pyc 1KB soft_dtw.cpython-37.pyc 4KB __init__.cpython-38.pyc 515B probabilistic.py 670B layers.py 3KB merging_cross_folds.py 582B pytorch_base.py 8KB __pycache__ pytorch_model.cpython-37.pyc 6KB layers.cpython-37.pyc 4KB pytorch_base.cpython-38.pyc 8KB pytorch_model.cpython-38.pyc 6KB merging_cross_folds.cpython-38.pyc 800B utils.cpython-38.pyc 1KB merging_cross_folds.cpython-37.pyc 799B __init__.cpython-37.pyc 513B utils.cpython-37.pyc 1KB pytorch_base.cpython-37.pyc 8KB layers.cpython-38.pyc 4KB __init__.cpython-38.pyc 553B pytorch_model.py 5KB pandas_ml_utils_torch_test test_multiple_feature_sets.py 2KB __init__.py 0B check_links.py 173B test_pytochmodel_accuracy.py 3KB test_layer.py 2KB test_pytorch_base.py 3KB test_custom_loss.py 6KB test_notebooks.py 350B test_loss.py 5KB test_pytorch_model.py 4KB scratch.py 2KB test_cross_folding.py 2KB pandas_ml_utils_torch.egg-info 
SOURCES.txt 3KB top_level.txt 49B PKG-INFO 2KB requires.txt 126B dependency_links.txt 1B licence.txt 1KB requirements.txt 19B dev-requirements.txt 52B MANIFEST.in 52B setup.cfg 79B Readme.md 1KB资源评论
