Update README.md

README.md

@@ -31,19 +31,23 @@ The model is created by Dongxu Li, Junnan Li, Steven C.H. Hoi.
 ```python
 from diffusers.pipelines import BlipDiffusionPipeline
 from diffusers.utils import load_image
+import torch
 
-
-blip_diffusion_pipe= BlipDiffusionPipeline.from_pretrained('ayushtues/blipdiffusion')
-blip_diffusion_pipe.to('cuda')
+blip_diffusion_pipe = BlipDiffusionPipeline.from_pretrained(
+    "ayushtues/blipdiffusion", torch_dtype=torch.float16
+).to("cuda")
 
 cond_subject = "dog"
 tgt_subject = "dog"
 text_prompt_input = "swimming underwater"
 
+cond_image = load_image(
+    "https://huggingface.co/datasets/ayushtues/blipdiffusion_images/resolve/main/dog.jpg"
+)
 
-cond_image = load_image("https://huggingface.co/datasets/ayushtues/blipdiffusion_images/resolve/main/dog.jpg")
+iter_seed = 88888
 guidance_scale = 7.5
-num_inference_steps = 50
+num_inference_steps = 25
 negative_prompt = "over-exposure, under-exposure, saturated, duplicate, out of frame, lowres, cropped, worst quality, low quality, jpeg artifacts, morbid, mutilated, out of frame, ugly, bad anatomy, bad proportions, deformed, blurry, duplicate"
 
 output = blip_diffusion_pipe(
@@ -56,8 +60,8 @@ output = blip_diffusion_pipe(
     neg_prompt=negative_prompt,
     height=512,
     width=512,
-)
-output[0][0].save("image.png")
+).images
+output[0].save("image.png")
 ```
 Input Image : <img src="https://huggingface.co/datasets/ayushtues/blipdiffusion_images/resolve/main/dog.jpg"  style="width:500px;"/>
 
@@ -70,17 +74,22 @@ from diffusers.pipelines import BlipDiffusionControlNetPipeline
 from diffusers.utils import load_image
 from controlnet_aux import CannyDetector
 
-blip_diffusion_pipe= BlipDiffusionControlNetPipeline.from_pretrained("ayushtues/blipdiffusion-controlnet")
-blip_diffusion_pipe.to('cuda')
+blip_diffusion_pipe = BlipDiffusionControlNetPipeline.from_pretrained(
+    "ayushtues/blipdiffusion-controlnet", torch_dtype=torch.float16
+).to("cuda")
 
-style_subject = "flower" # subject that defines the style
+style_subject = "flower"  # subject that defines the style
 tgt_subject = "teapot"  # subject to generate.
 text_prompt = "on a marble table"
-cldm_cond_image = load_image("https://huggingface.co/datasets/ayushtues/blipdiffusion_images/resolve/main/kettle.jpg").resize((512, 512))
-canny = CannyDetector()
-cldm_cond_image = canny(cldm_cond_image, 30, 70, output_type='pil')
 
-style_image = load_image("https://huggingface.co/datasets/ayushtues/blipdiffusion_images/resolve/main/flower.jpg")
+cldm_cond_image = load_image(
+    "https://huggingface.co/datasets/ayushtues/blipdiffusion_images/resolve/main/kettle.jpg"
+).resize((512, 512))
+canny = CannyDetector()
+cldm_cond_image = canny(cldm_cond_image, 30, 70, output_type="pil")
+style_image = load_image(
+    "https://huggingface.co/datasets/ayushtues/blipdiffusion_images/resolve/main/flower.jpg"
+)
 
 guidance_scale = 7.5
 num_inference_steps = 50
@@ -89,7 +98,7 @@ negative_prompt = "over-exposure, under-exposure, saturated, duplicate, out of f
 output = blip_diffusion_pipe(
     text_prompt,
     style_image,
-     cldm_cond_image,
+    cldm_cond_image,
     style_subject,
     tgt_subject,
     guidance_scale=guidance_scale,
@@ -97,8 +106,8 @@ output = blip_diffusion_pipe(
     neg_prompt=negative_prompt,
     height=512,
     width=512,
-    )
-output[0][0].save("image.png")
+).images
+output[0].save("image.png")
 ```
 
 Input Style Image : <img src="https://huggingface.co/datasets/ayushtues/blipdiffusion_images/resolve/main/flower.jpg"  style="width:500px;"/>
@@ -111,19 +120,24 @@ from diffusers.pipelines import BlipDiffusionControlNetPipeline
 from diffusers.utils import load_image
 from controlnet_aux import HEDdetector
 
-blip_diffusion_pipe= BlipDiffusionControlNetPipeline.from_pretrained("ayushtues/blipdiffusion-controlnet")
+blip_diffusion_pipe = BlipDiffusionControlNetPipeline.from_pretrained(
+    "ayushtues/blipdiffusion-controlnet"
+)
 controlnet = ControlNetModel.from_pretrained("lllyasviel/sd-controlnet-scribble")
 blip_diffusion_pipe.controlnet = controlnet
-blip_diffusion_pipe.to('cuda')
+blip_diffusion_pipe.to("cuda")
 
-style_subject = "flower" # subject that defines the style
+style_subject = "flower"  # subject that defines the style
 tgt_subject = "bag"  # subject to generate.
 text_prompt = "on a table"
-cldm_cond_image = load_image("https://huggingface.co/lllyasviel/sd-controlnet-scribble/resolve/main/images/bag.png" ).resize((512, 512))
+cldm_cond_image = load_image(
+    "https://huggingface.co/lllyasviel/sd-controlnet-scribble/resolve/main/images/bag.png"
+).resize((512, 512))
 hed = HEDdetector.from_pretrained("lllyasviel/Annotators")
 cldm_cond_image = hed(cldm_cond_image)
-
-style_image = load_image("https://huggingface.co/datasets/ayushtues/blipdiffusion_images/resolve/main/flower.jpg")
+style_image = load_image(
+    "https://huggingface.co/datasets/ayushtues/blipdiffusion_images/resolve/main/flower.jpg"
+)
 
 guidance_scale = 7.5
 num_inference_steps = 50
@@ -132,7 +146,7 @@ negative_prompt = "over-exposure, under-exposure, saturated, duplicate, out of f
 output = blip_diffusion_pipe(
     text_prompt,
     style_image,
-     cldm_cond_image,
+    cldm_cond_image,
     style_subject,
     tgt_subject,
     guidance_scale=guidance_scale,
@@ -140,8 +154,8 @@ output = blip_diffusion_pipe(
     neg_prompt=negative_prompt,
     height=512,
     width=512,
-)
-output[0][0].save("image.png"')
+).images
+output[0].save("image.png")
 ```
 
 Input Style Image : <img src="https://huggingface.co/datasets/ayushtues/blipdiffusion_images/resolve/main/flower.jpg"  style="width:500px;"/>