A demo project for an auto-diff numerical operation library. Designed for course "Advanced Apps of Machine Learning'' at Tsinghua Univ.
A demo project for a tiny auto-diff symbolic numerical operation library.
This project is a school project for course Advanced Apps of Machine Learning at Tsinghua University
All numerical operations are actually performed by the python lib numpy, the lib kaleido aims at building symbolic computation
graphs and performing auto differentiation. The core are well designed and concise (< 1000 lines in total). So it can be a good example
for those who are interested in symbolic computation libs.
Finally, it is a demo project, so do NOT use it in production environment.
You can find two simple examples in folder examples.