Additional internal Settings
Compared to the original playground, several UI elements have been hidden for simplicity. The following settings are used internally:
- Learning rate: 0.03
- Activation function: Rectified Linear Unit (ReLU)
- Ratio of training to test data: 50%
- Noise: 0
- Batch size: 10
- Problem type: classification
- No Regularization, Regularization rate 0
Credits
Repurposed for university teaching at Princeton University (SPI-353) and University of Hamburg (66-656 / CSD) with cosmetic modifications. Taken from https://github.com/tensorflow/playground. Credit for everything but errors goes to Daniel Smilkov and Shan Carter!
This is a continuation of many people’s previous work — most notably Andrej Karpathy’s convnet.js demo and Chris Olah’s articles about neural networks. Many thanks also to D. Sculley for help with the original idea and to Fernanda Viégas and Martin Wattenberg and the rest of the Big Picture and Google Brain teams for feedback and guidance.