improve the glide example

models/vision/glide/modeling_glide.py

@@ -223,6 +223,6 @@ class GLIDE(DiffusionPipeline):
             sampled_prev_image = prev_image + prev_variance
             image = sampled_prev_image
 
-        image = image[0].permute(1, 2, 0)
+        image = image.permute(0, 2, 3, 1)
 
         return image

models/vision/glide/run_glide.py

@@ -11,9 +11,10 @@ model_id = "fusing/glide-base"
 pipeline = DiffusionPipeline.from_pretrained(model_id)
 
 # run inference (text-conditioned denoising + upscaling)
-img = pipeline("a clip art of a hugging face", generator)
+img = pipeline("a crayon drawing of a corgi", generator)
 
 # process image to PIL
+img = img.squeeze(0)
 img = ((img + 1)*127.5).round().clamp(0, 255).to(torch.uint8).cpu().numpy()
 image_pil = PIL.Image.fromarray(img)