VGG-19-encoded Style Features for AdaIN

1000 randomly selected image features from the painter-by-numbers dataset (small yellow duck and Wendy Kan. Painter by Numbers.  https://kaggle.com/competitions/painter-by-numbers, 2016. Kaggle.).

Features are extracted from a VGG-19 network pretrained on imagenet before its final layer. They can be used for StyleTransfer via the AdaIN method, so that it is more efficient. 

Reference repository: https://github.com/Georgsiedel/model-based-data-augmentation