小编Lin*_*hen的帖子

假设我创建了一个张量并将其放在 GPU 上并且以后不需要它并且想要释放分配给它的 GPU 内存；我该怎么做？

import torch
a=torch.randn(3,4).cuda() # nvidia-smi shows that some mem has been allocated.
# do something
# a does not exist and nvidia-smi shows that mem has been freed.

我试过了：

1. del a
2. del a; torch.cuda.empty_cache()

但它们都不起作用。

我有一堂课，里面有一个字典。我想通过直接“ instance.key”而不是“ instance.d [key]”来访问字典中的变量。怎么做？

class A:
    def __init__(self):
        self.a = 1
        self.b = 2
        self.fields = {'c': 3, 'd': 4}


def main():
    a = A()
    print(a.a) # 1
    print(a.b) # 2
    print(a.c) # currently raise Exception, I want it to print 3
    print(a.d) # currently raise Exception, I want it to print 4


if __name__ == '__main__':
    main()