img-embedding

1. Loads random 1000 images from a folder,
2. Creates embedding by using a pre-trained vgg11_bn model,
3. Lowers dimensionality using t-SNE to two dimensions,
4. Plots the images interactively, one can browse and zoom.

Run

1. Add images to imgs folder
2. python main.py