In the Jupyter notebook titled simple_reweighting_example.ipynb, you can find a simple example using pytorch and numpy on how to train a network to learn to reweight a distribution into another one. On this example, you will learn to weight data generated from a Gaussian probability distribution centered at 10 with a standard deviation of 1 to get data distributed following a Gaussian distribution centered at 11 with a standard deviation of 1.
Follow the instructions here
Follow the instructions here
Follow the instructions here