https://github.com/vdenberg/noisy-label-neural-network

Re-implementation of Bekker & Goldbergers Noisy Label Neural Network
https://github.com/vdenberg/noisy-label-neural-network

Last synced: 6 days ago
JSON representation

Re-implementation of Bekker & Goldbergers Noisy Label Neural Network

• Host: GitHub
• URL: https://github.com/vdenberg/noisy-label-neural-network
• Owner: vdenberg
• Created: 2016-08-04T13:43:23.000Z (over 8 years ago)
• Default Branch: master
• Last Pushed: 2016-09-04T17:16:46.000Z (about 8 years ago)
• Last Synced: 2024-05-21T13:36:17.160Z (6 months ago)
• Language: Python
• Size: 16.6 MB
• Stars: 3
• Watchers: 3
• Forks: 1
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Re-implementation of Bekker & Goldbergers Noisy Label Neural Network

Run experiment with:

$ python NNLNonMNIST.py [-pretrain] [-EM] [-plot]

One, two or all of three flags required.

-pretrain   trains neural networks on noisy data (noise fractions predetermined)

-EM         runs an EM module based on aforementioned models

-plot       !!!currently only prints rather than plots results!!!