Skip to content

Latest commit

 

History

History
17 lines (8 loc) · 741 Bytes

readme.md

File metadata and controls

17 lines (8 loc) · 741 Bytes

ACW: a light-weight deep learning framework based on numpy

requires Python 3.x, numpy, matplotlib

It is still improving. The main design of this framework it to simplify the structure. I devide all stuff common in deep learning models into 3 categories:

  • Layer (LayerLib.py): perform some transform on input and have parameters to be optimize (such as Dense, Conv, and so on). The api is following Keras somewhat.
  • Op (OpLib.py): operation on input but with no parameters, such as add, multiply, maxpool, non-linear activation and so on.
  • Loss (LossLib.py): The loss function used to minimize. such as MSE, L1Loss, CrossEntropy and so on.

example

A MLP example has been provide in example/ dir.