Pytorch set random seed

---

In the use of PyTorch when , If you want to seed random numbers , stay GPU or CPU Fix each training result on the , You need to add the following code at the beginning of program execution ：

def seed_everything():
    '''  Set the overall development environment seed :param seed: :param device: :return: '''
    import os
    import random
    import numpy as np

    random.seed(seed)
    os.environ['PYTHONHASHSEED'] = str(seed)
    np.random.seed(seed)
    torch.manual_seed(seed)
    torch.cuda.manual_seed(seed)
    torch.cuda.manual_seed_all(seed)
    
    # some cudnn methods can be random even after fixing the seed
    # unless you tell it to be deterministic
    torch.backends.cudnn.deterministic = True