YAI_toyproject

Description

We use EfficientNetV2, which improves training speed, model size, and memory efficiency from the original EfficientNet by utilizing Compound Scaling, Fused MBConv, and Progressive Learning.