num_images_per_prompt=num_samples FAILS after latest commit in SD pipelines

[ghpkishore]

Describe the bug

This used to wwork:

images = pipe(
prompt=prompt,
image=image,
mask_image=mask_image,
guidance_scale=guidance_scale,
generator=generator,
num_images_per_prompt=num_samples,
num_inference_steps =50,
height=height,
width=width,

).images

Now it doesn't.

Reproduction

Ue the RUNWAYML inpainting notebook. IT fails in that!

https://colab.research.google.com/github/huggingface/notebooks/blob/main/diffusers/in_painting_with_stable_diffusion_using_diffusers.ipynb

Logs

RuntimeError                              Traceback (most recent call last)
<ipython-input-8-7352d4d77608> in <module>
     11     guidance_scale=guidance_scale,
     12     generator=generator,
---> 13     num_images_per_prompt=num_samples,
     14 ).images

1 frames
/usr/local/lib/python3.7/dist-packages/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion_inpaint.py in __call__(self, prompt, image, mask_image, height, width, num_inference_steps, guidance_scale, negative_prompt, num_images_per_prompt, eta, generator, latents, output_type, return_dict, callback, callback_steps, **kwargs)
    567 
    568             # concat latents, mask, masked_image_latents in the channel dimension
--> 569             latent_model_input = torch.cat([latent_model_input, mask, masked_image_latents], dim=1)
    570 
    571             latent_model_input = self.scheduler.scale_model_input(latent_model_input, t)

RuntimeError: Sizes of tensors must match except in dimension 1. Expected size 6 but got size 2 for tensor number 1 in the list.

System Info

The colab notebook in the runwayml inpainting page fails.

[ghpkishore]

Marking @patrickvonplaten as the latest commits were made by you.

[ghpkishore]

It has to do something with num_samples because the error keeps changing with num_samples.

[ghpkishore]

Marking @patil-suraj @pcuenca Incase you guys can resolve this quickly

[patil-suraj]

Thanks for reporting, looking into it.