pytorch 确保实验结果的复现性

 根据官方文档整理的代码块

def setup_seed(seed):
    # Make sure the algorithm results are reproducible
    torch.manual_seed(seed)  # set CPU seed
    torch.cuda.manual_seed(seed)   # set GPU seed
    torch.cuda.manual_seed_all(seed)  # if you are using multi-GPU
    np.random.seed(seed)  # Numpy module
    random.seed(seed)  # Python random module
    # causes cuDNN to deterministically select an algorithm, possibly at the cost of reduced performance
    torch.backends.cudnn.benchmark = False
    # avoid algorithm itself may be nondeterministic
    torch.backends.cudnn.deterministic = True

if __name__ == '__main__':
    setup_seed(1)
    # todo

参考： pytoch官网文档