search

faustomorales / vit-keras

Keras implementation of ViT (Vision Transformer)
Apache License 2.0
336 stars 80 forks source link

Support for rectangular images #26

Closed dakshthapar closed 3 years ago

dakshthapar commented 3 years ago

Changed the embedding layer weight load to accumulate rectangular images.

dakshthapar commented 3 years ago

Thank you for your review. I have changed the notation for the y-axis and x-axis with respect to height and width as suggested.