This is not a fork of somebody else's code. I, @BobMcDear, am the original creator of this project but due to problems with Git was forced to delete and restore it. In other words, anushmite/PyTorch-EfficientNet is a fork of this repository and not vice versa.
This is an implementation of EfficientNet in PyTorch. You can find a Colab version of the code here.
efficientnet.py/EfficientNet: This is a generic form of EfficientNet that takes in width and depth scale factors and returns a corresponding scaled version of EfficientNet-B0. These two can be passed in as w_factor and d_factor respectively, with default values of 1. For instance, if you set them to 1.1 and 1.2, that would give EfficneNet-B2, while 2 and 3.1 would give EfficientNet-B7. Moreover, out_sz can be passed to set the output dimension of the final fully-connected layer, with a default of 1000.
For convenience, EfficientNet B0 through B7 can be found under the same file, where each one's name is EfficientNetB[number].