Resnet模型如何共享权重

最新推荐文章于 2024-01-26 14:13:46 发布

小小她爹

最新推荐文章于 2024-01-26 14:13:46 发布

阅读量763

点赞数

分类专栏：架构、自然语言处理、金融风控、信用评分深度学习文章标签：深度学习

原文链接：https://discuss.pytorch.org/t/how-to-implement-two-networks-with-shared-weights-and-with-separate-batch-normalizations/6902

版权

架构、自然语言处理、金融风控、信用评分同时被 2 个专栏收录

39 篇文章

订阅专栏

深度学习

6 篇文章

订阅专栏

引用：https://discuss.pytorch.org/t/share-weight-in-resnet-model/21371

Share weight in Resnet model

提问：

答案：

I’m not sure if you will train the share weight twice in one forward action, if it is , I think the forward code should be something like following.


def forward(self, x):
    input_image = x
    ...
    x = self.share_weight_layer(x)
    ...
    pool = x = self.avg_pool(x)
    product = pool * input_image
    x = self.some_layer(product)
    ...
    x = self.share_weight_layer(x) # this is the same layer as previous one
    ...
    return x