如何把imagenet预训练的模型，输入层的通道数随心所欲的修改，从而来适应自己的任务

#增加一个通道w = layers[0].weightlayers[0] = nn.Conv2d(4, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)layers[0].weight = torch.nn.Parameter(torch.cat((w, w[:, :1, :, :]), dim=1)) #方式2w = layers[0].weightlayers[0] = nn.Conv2d(4, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)layers[0].weight = torch.nn.Parameter(torch.cat((w, torch.zeros(64, 1, 7, 7)), dim=1)) #单通道输入layers[0] = nn.Conv2d(1, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)layers[0].weight = torch.nn.Parameter(w[:, :1, :, :])

以上这篇pytorch 实现在预训练模型的 input上增减通道就是小编分享给大家的全部内容了，希望能给大家一个参考，也希望大家多多支持。