In this repository, I have written some code on understanding softmax-regression. In addition, I also tried to get familiar with Pytorch by programming softmax regression in the Pytorch flow as well as using Dataset
and DataLoader
.
The source code, in my view, is easy to read. You can change the mean values of the data or the number of classes in the main.py
and softmax_regression.py
, respectively. However, the boundaries drawn after training are valid only if the data is 2-dimensional.