deb_control_files:
- control
- md5sums
- postinst
- prerm
deb_fields:
Architecture: all
Depends: python3-torch (>= 1.3.1), python3-numpy, python3-scipy, python3-sklearn,
python3-tabulate, python3-tqdm, python3.11, python3:any
Description: "scikit-learn compatible neural network library that wraps PyTorch\n\
\ The goal of skorch is to make it possible to use PyTorch with sklearn. This\
\ is\n achieved by providing a wrapper around PyTorch that has an sklearn interface.\n\
\ In that sense, skorch is the spiritual successor to nolearn, but instead of\n\
\ using Lasagne and Theano, it uses PyTorch.\n .\n skorch does not re-invent the\
\ wheel, instead getting as much out of your way as\n possible. If you are familiar\
\ with sklearn and PyTorch, you don\u2019t have to learn\n any new concepts, and\
\ the syntax should be well known. (If you\u2019re not familiar\n with those libraries,\
\ it is worth getting familiarized.)\n .\n Additionally, skorch abstracts away\
\ the training loop, making a lot of\n boilerplate code obsolete. A simple net.fit(X,\
\ y) is enough. Out of the box,\n skorch works with many types of data, be it\
\ PyTorch Tensors, NumPy arrays,\n Python dicts, and so on. However, if you have\
\ other data, extending skorch is\n easy to allow for that.\n .\n Overall, skorch\
\ aims at being as flexible as PyTorch while having a clean\n interface as sklearn."
Homepage: https://github.com/skorch-dev/skorch
Installed-Size: '796'
Maintainer: Debian Deep Learning Team <debian-ai@lists.debian.org>
Package: python3-skorch
Priority: optional
Section: science
Source: skorch
Version: 0.12.1-2
srcpkg_name: skorch
srcpkg_version: 0.12.1-2