分享

A Python implementation of the Replicated Softmax model.

 Clay*more 2014-02-16
This is an implementation of R. Salakhutdinov and G.E. Hinton's Replicated Softmax model,

see http://www./~rsalakhu/papers/repsoft.pdf
You will need Numpy and SciPy installed. I don't provide the dataset.
Also feel free to change the initialization of the biases, it just worked for me.
It is part of my M.Sc. Thesis, advised by Christian Osendorfer at TUMI6.
You can reach me here: joerg [-at-] fylance [-.-] de

  • rsm.py: contains the RSM implementation encapsulated in a class.
  • rsm_driver_20newsgroups.py: shows an exemplary usage of the class.

Download: rsm.tar.gz (downloaded 517 times)

This software comes with no warranty at all. J鰎g Landthaler - Last modified: Wed Dec 28 15:15:21 2011 CET

    本站是提供个人知识管理的网络存储空间,所有内容均由用户发布,不代表本站观点。请注意甄别内容中的联系方式、诱导购买等信息,谨防诈骗。如发现有害或侵权内容,请点击一键举报。
    转藏 分享 献花(0

    0条评论

    发表

    请遵守用户 评论公约

    类似文章 更多