摘要:
print(torch.arange(0, 20)) tensor([ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19])>>> print(torch.arange(0, 20).view(20, 1, 1) 阅读全文
摘要:
https://www.jb51.net/article/177713.htm 阅读全文
摘要:
方法一:https://www.pytorchtutorial.com/pytorch-learning-rate-decay/ 方法二: # lr_step = [30,80] if epoch in opt.lr_step: save_model(os.path.join(opt.save_di 阅读全文
摘要:
check_set_list = ['zheng','xinying','zhang'] >>> sample_set_list = ['xinying','zhangchunmei','zhiwu'] >>> check_set_list & sample_set_list # &求交集必须对集合 阅读全文