摘要:
论文地址:https://arxiv.org/abs/2007.11824 代码地址:https://github.com/megvii-model/FunnelAct 讲解:https://mp.weixin.qq.com/s/CCnx5WeL28bAFCM6QtFX4g 阅读全文
摘要:
来自ICML2020的一篇论文: Do We Need Zero Training Loss After Achieving Zero Training Error? 地址:https://arxiv.org/pdf/2002.08709.pdf 一行代码解决: 阅读全文
摘要:
读取图像一般是两个库:opencv和PIL 1、使用opencv读取图像 import cv2 image=cv2.imread("/content/drive/My Drive/colab notebooks/image/cat1.jpg") print(image.shape) (490, 41 阅读全文