Hi! I think i have very noob question, but how can i handle diffusers (stable diffusion pipeline) output as numpy array and then convert to image? For now i use output_type=“pil” and then convert to numpy array and it works. But how to handle output_type=“np” and convert it to 3 dimensions (w,h,c)? All my attempts doesn’t work or give me black screen.
images = pipe(prompt=prompt, num_inference_steps=num_inference_steps, guidance_scale=8.0, lcm_origin_steps=50, output_type="np").images[0]