Real-world uses

Created for this library

1. 1.

   A synthetic data startup uses a generator in its GAN setup to produce realistic transaction streams without exposing real customer records.

2. 2.

   A computer vision team trains a generator network to produce synthetic images of rare defects for augmenting a quality-control dataset.

3. 3.

   A research team uses a generator to produce paraphrased variants of customer support tickets so its classifier sees a more diverse training set.