By Charlie Cheng, Ethan Qiu, Yuexi Shen, Siqi Zhu
- This is our final project for the course CS 182/282A: Deep Neural Networks at UC Berkeley.
This is an introductory tutorial to EfficientNet, a family of convolutional neural networks that have achieved state-of-the-art performance on image classification tasks. We designed various conceptual questions, including the intuition on compound scaling, stochastic depth training, squeeze & excitation, and more. We also implemented a PyTorch version of EfficientNet and trained it on the CIFAR-10 dataset. We hope this tutorial can help you understand the EfficientNet paper and its implementation better.