A issus of MAEViTEncoder's code

[wangbaoyuanGUET]

Hi!Developers of SlefMedMAE!
While I checking the code of MAEViTEncoder, I see self.patch_embed = embed_layer(img_size=patch_size, patch_size=patch_size, in_chans=in_chans, embed_dim=embed_dim). I'd like to know why img_size uses patch_size instead of the size of the img?

[HadiHammoud44]

same concern, i guess because the image was already reshaped as (Batch * Sequence Length, in_channels, new_image_size) before applying the 3D Conv, where the new shape equals the patch size.