cross_attention_kwargs ['attention_map'] are not expected by AttnProcessor2_0 and will be ignored.

[dingbang777]

cross_attention_kwargs ['attention_map'] are not expected by AttnProcessor2_0 and will be ignored. I want to define a custom cross_attention_kwargs for AttnProcessor2_0 but before I put it into the attnprocessor (at the diffusers repo /diffusers/models/unets/unet_2d_blocks.py

 class CrossAttnDownBlock2D(nn.Module      def forward(
        self,
        hidden_states: torch.Tensor,
        temb: Optional[torch.Tensor] = None,
        encoder_hidden_states: Optional[torch.Tensor] = None,
        attention_mask: Optional[torch.Tensor] = None,
        cross_attention_kwargs: Optional[Dict[str, Any]] = None,
        encoder_attention_mask: Optional[torch.Tensor] = None,
        additional_residuals: Optional[torch.Tensor] = None,
    ) -> Tuple[torch.Tensor, Tuple[torch.Tensor, ...]]:
        if cross_attention_kwargs is not None:
            if cross_attention_kwargs.get("scale", None) is not None:
                logger.warning("Passing `scale` to `cross_attention_kwargs` is deprecated. `scale` will be ignored.")
        print(86594619)
        print('cross_attention_kwargs', cross_attention_kwargs)  ):    )))))))  

I print the dict 'cross_attention_kwargs' and got the error. And finally, cross_attention_kwargs was not successfully passed into the attnprocessor

[yiyixuxu]

hi!
I think you need to set your model to use your custom attention processor that accept this argument,

unet.set_attn_processor()

[dingbang777]

thank you, I've solved the problem