文章MLP-Mixer 代码维度问题

Question

model = MLPMixer(in_channels=3, dim=512, num_classes=1000, patch_size=16, image_size=224, depth=2, token_dim=256,channel_dim=2048)



  self.to_patch_embedding = nn.Sequential(
            nn.Conv2d(in_channels, dim, patch_size, patch_size),
            Rearrange('b c h w->b (h w) c')
        )

这里想问一下 dim ,token_dim,channel_dim 的数值是真么来的