Download Latest Version
Sweep code for studying model population stats source code.tar.gz (688.6 kB)
Get an email when there's a new version of PyCls
| Name | Modified | Size | Downloads / Week |
|---|---|---|---|
| Parent folder | |||
| Initial release of Model Zoo source code.tar.gz | 2020-04-15 | 646.9 kB | |
| Initial release of Model Zoo source code.zip | 2020-04-15 | 705.6 kB | |
| README.md | 2020-04-15 | 957 Bytes | |
| Totals: 3 Items | 1.4 MB | 0 | |
We have added a large set of baseline results and pretrained models available for download in the pycls Model Zoo; including the simple, fast, and effective RegNet models that we hope can serve as solid baselines across a wide range of flop regimes.
New features included in this release: - Cache model weight URLs provided in configs locally | [4e470e] - Allow optional weight decay fine-tuning for BN params, changes the default BN weight decay | [4e470e] - Support Squeeze & Excitation in AnyNet | [ec1886] - RegNet model abstraction | 4f5b5dafe4f4274f7cba774923b8d1ba81d9904c, 708d429e67b1a43dfd7e22cb754ecaaf234ba308, [24a280] - Run precise time in test_net and count acts for a model | [ad01b8]
Other changes: - Updated README | [4acac2] - New Model Zoo | 4acac2b06945f6b0722f20748e2104bd6a771468, [4f5b5d] - New configs for models from Designing Network Design Spaces paper | 44bcfa71ebd7b9cd8e81f4d28ce319e814cbbb73, [e9c1ef]