VMBNet

class GroupBlock(TimmBlock):
    def __init__(self, dim, num_heads, mlp_ratio=4., qkv_bias=False, qk_scale=None, drop=0., attn_drop=0.,
                 drop_path=0., act_layer=nn.GELU, norm_layer=nn.LayerNorm, sr_ratio=1, ws=1):

        #super(GroupBlock, self).__init__(dim, num_heads, mlp_ratio, qkv_bias, qk_scale, drop, attn_drop,
        #                                drop_path, act_layer, norm_layer)

        #delete the qk_scale
        super(GroupBlock, self).__init__(dim, num_heads, mlp_ratio, qkv_bias, drop, attn_drop,
                                        drop_path, act_layer, norm_layer)
        del self.attn
        if ws == 1:
            self.attn = Attention(dim, num_heads, qkv_bias, qk_scale, attn_drop, drop, sr_ratio)
        else:
            self.attn = GroupAttention(dim, num_heads, qkv_bias, qk_scale, attn_drop, drop, ws)

    def forward(self, x, H, W):
        x = x + self.drop_path(self.attn(self.norm1(x), H, W))
        x = x + self.drop_path(self.mlp(self.norm2(x)))

        return x