fix w h

Author: lyogavin

examples/text_to_image/pipeline_stable_diffusion_img2img_render_text.py

@@ -177,7 +177,13 @@ def __call__(
         """
 
         # 1. Check inputs
-        self.check_inputs(prompt, strength, callback_steps)
+        # copied from https://github.com/huggingface/diffusers/blob/main/examples/community/img2img_inpainting.py
+        if isinstance(image, PIL.Image.Image):
+            w, h = image.size
+        elif isinstance(image, torch.Tensor):
+            *_, h, w = image.shape
+
+        self.check_inputs(prompt, h, w, callback_steps)
 
         # 2. Define call parameters
         batch_size = 1 if isinstance(prompt, str) else len(prompt)
@@ -207,11 +213,6 @@ def __call__(
         # 6.1 Prepare latent variables
         num_channels_latents = self.unet.in_channels // 2
 
-        # copied from https://github.com/huggingface/diffusers/blob/main/examples/community/img2img_inpainting.py
-        if isinstance(image, PIL.Image.Image):
-            w, h = image.size
-        elif isinstance(image, torch.Tensor):
-            *_, h, w = image.shape
 
         latents = self.prepare_latents(
             batch_size * num_images_per_prompt,