Merge branch 'main' of https://github.com/huggingface/diffusers into main

models/vision/glide/modeling_glide.py

@@ -124,6 +124,7 @@ class GLIDE(DiffusionPipeline):
             - _extract_into_tensor(self.noise_scheduler.sqrt_recipm1_alphas_cumprod, t, x_t.shape) * eps
         )
 
+    @torch.no_grad()
     def __call__(self, prompt, generator=None, torch_device=None):
         torch_device = "cuda" if torch.cuda.is_available() else "cpu"
 
@@ -164,4 +165,6 @@ class GLIDE(DiffusionPipeline):
             nonzero_mask = (t != 0).float().view(-1, *([1] * (len(image.shape) - 1)))  # no noise when t == 0
             image = mean + nonzero_mask * torch.exp(0.5 * log_variance) * noise
 
+        image = image[0].permute(1, 2, 0)
+
         return image

models/vision/glide/run_glide.py

 import torch
 
 from modeling_glide import GLIDE
+import matplotlib
+import matplotlib.pyplot as plt
+matplotlib.rcParams['interactive'] = True
 
 
 generator = torch.Generator()
@@ -10,5 +13,8 @@ generator = generator.manual_seed(0)
 pipeline = GLIDE.from_pretrained("fusing/glide-base")
 
 img = pipeline("an oil painting of a corgi", generator)
+img = ((img + 1)*127.5).round().clamp(0, 255).to(torch.uint8).cpu().numpy()
 
-print(img)
+plt.figure(figsize=(8, 8))
+plt.imshow(img)
+plt.show()

src/diffusers/__init__.py

@@ -9,6 +9,7 @@ from .models.clip_text_transformer import CLIPTextModel
 from .models.unet import UNetModel
 from .models.unet_glide import UNetGLIDEModel
 from .models.unet_ldm import UNetLDMModel
+from .models.vqvae import VQModel
 from .pipeline_utils import DiffusionPipeline
 from .schedulers.classifier_free_guidance import ClassifierFreeGuidanceScheduler
 from .schedulers.gaussian_ddpm import GaussianDDPMScheduler

src/diffusers/models/__init__.py

@@ -20,3 +20,4 @@ from .clip_text_transformer import CLIPTextModel
 from .unet import UNetModel
 from .unet_glide import UNetGLIDEModel
 from .unet_ldm import UNetLDMModel
+from .vqvae import VQModel
\ No newline at end of file