Carbon footprints and performance for different KD approaches.
收藏数据链接：
官方服务：
资源简介：
Here, for image recognition, we use ResNet18 (teacher) and MobileNetV2 (student) models. For object detection, we use VGG16 (teacher) and MobileNetV2 (student). We report average results after running the same program five times. ↑ (↓) means higher (lower) is better.
创建时间：
2023-05-15
