pytorch常见问题

单机多卡训练 及多机多卡训练

https://zhuanlan.zhihu.com/p/86441879

 

网络初始化 1

def weights_init(m):
    if isinstance(m, nn.Conv2d) or isinstance(m, nn.Linear):
        nn.init.xavier_uniform_(m.weight.data)
        nn.init.constant_(m.bias, 0.1)

net.apply(weights_init)

 

网络初始化 2

    def weights_init(self, m):
        classname = m.__class__.__name__
        if classname.find('Conv') != -1:
            nn.init.kaiming_normal_(m.weight.data)
            nn.init.kaiming_normal_(m.bias.data)
        if classname.find("Batch") != -1:
            nn.init.normal_(m.weight.data, 1.0, 0.02)  # 6
            nn.init.constant_(m.bias.data, 0)

 

给训练踩踩油门——Pytorch加速数据读取

https://zhuanlan.zhihu.com/p/80695364

https://zhuanlan.zhihu.com/p/80695364

 

pytorch 使用tensorboard

https://blog.csdn.net/wsp_1138886114/article/details/87602112