Merge branch 'main' of https://huggingface.co/diffusers/tools

run_local_xl.py

@@ -17,7 +17,7 @@ api = HfApi()
 start_time = time.time()
 
 use_refiner = bool(int(sys.argv[1]))
-use_diffusers = True
+use_diffusers = False
 
 vae = AutoencoderKL.from_pretrained("madebyollin/sdxl-vae-fp16-fix", torch_dtype=torch.float16, force_upcast=True)
 if use_diffusers:

run_local_xl_inpaint.py

@@ -0,0 +1,67 @@
+#!/usr/bin/env python3
+from diffusers import DiffusionPipeline, EulerDiscreteScheduler, StableDiffusionPipeline, KDPM2DiscreteScheduler, StableDiffusionImg2ImgPipeline, HeunDiscreteScheduler, KDPM2AncestralDiscreteScheduler, DDIMScheduler
+from diffusers import StableDiffusionXLPipeline, StableDiffusionXLImg2ImgPipeline, StableDiffusionXLInpaintPipeline, AutoencoderKL
+import time
+from pytorch_lightning import seed_everything
+import os
+from huggingface_hub import HfApi
+# from compel import Compel
+import torch
+import sys
+from pathlib import Path
+import requests
+from PIL import Image
+from io import BytesIO
+
+api = HfApi()
+start_time = time.time()
+
+use_refiner = bool(int(sys.argv[1]))
+use_diffusers = True
+
+vae = AutoencoderKL.from_pretrained("madebyollin/sdxl-vae-fp16-fix", torch_dtype=torch.float16, force_upcast=True)
+pipe = StableDiffusionXLInpaintPipeline.from_pretrained("stabilityai/stable-diffusion-xl-base-0.9", vae=vae, torch_dtype=torch.float16, variant="fp16", use_safetensors=True, local_files_only=True)
+print(time.time() - start_time)
+pipe.to("cuda")
+
+def download_image(url):
+    response = requests.get(url)
+    return Image.open(BytesIO(response.content)).convert("RGB")
+
+
+img_url = "https://raw.githubusercontent.com/CompVis/latent-diffusion/main/data/inpainting_examples/overture-creations-5sI6fQgYIuo.png"
+mask_url = "https://raw.githubusercontent.com/CompVis/latent-diffusion/main/data/inpainting_examples/overture-creations-5sI6fQgYIuo_mask.png"
+
+init_image = download_image(img_url).resize((1024, 1024))
+mask_image = download_image(mask_url).resize((1024, 1024))
+
+if use_refiner:
+    start_time = time.time()
+    refiner = StableDiffusionXLInpaintPipeline.from_pretrained("stabilityai/stable-diffusion-xl-refiner-0.9", vae=vae, torch_dtype=torch.float16, use_safetensors=True, variant="fp16")
+    refiner.to("cuda")
+    # refiner.enable_sequential_cpu_offload()
+
+prompt = "A majestic tiger sitting on a bench"
+steps = 50
+seed = 3
+denoising_end = None
+
+seed_everything(seed)
+start_time = time.time()
+image = pipe(prompt=prompt, image=init_image, mask_image=mask_image, num_inference_steps=steps, denoising_end=denoising_end, strength=0.80, output_type="latent").images
+print(time.time() - start_time)
+
+if use_refiner:
+    image = refiner(prompt=prompt, image=image, mask_image=mask_image, num_inference_steps=steps, denoising_start=denoising_end).images[0]
+
+file_name = f"aaa_1"
+path = os.path.join(Path.home(), "images", "ediffi_sdxl", f"{file_name}.png")
+image.save(path)
+
+api.upload_file(
+    path_or_fileobj=path,
+    path_in_repo=path.split("/")[-1],
+    repo_id="patrickvonplaten/images",
+    repo_type="dataset",
+)
+print(f"https://huggingface.co/datasets/patrickvonplaten/images/blob/main/{file_name}.png")

run_xl_ediffi.py

@@ -1,5 +1,5 @@
 #!/usr/bin/env python3
-from diffusers import DiffusionPipeline, EulerDiscreteScheduler, StableDiffusionPipeline, KDPM2DiscreteScheduler, StableDiffusionImg2ImgPipeline, HeunDiscreteScheduler, KDPM2AncestralDiscreteScheduler, DDIMScheduler
+from diffusers import DPMSolverMultistepScheduler, DiffusionPipeline, EulerDiscreteScheduler, StableDiffusionPipeline, KDPM2DiscreteScheduler, StableDiffusionImg2ImgPipeline, HeunDiscreteScheduler, KDPM2AncestralDiscreteScheduler, DDIMScheduler, HeunDiscreteScheduler, DEISMultistepScheduler
 from diffusers import DiffusionPipeline
 import time
 from pytorch_lightning import seed_everything
@@ -18,25 +18,39 @@ from torch.nn.functional import fractional_max_pool2d_with_indices
 api = HfApi()
 start_time = time.time()
 
+scheduler = DPMSolverMultistepScheduler(
+    beta_start=0.00085,
+    beta_end=0.012,
+    beta_schedule="scaled_linear",
+    prediction_type="epsilon",
+    num_train_timesteps=1000,
+    trained_betas=None,
+    thresholding=False,
+    algorithm_type="dpmsolver++",
+    solver_type="midpoint",
+    lower_order_final=True,
+    use_karras_sigmas=True,
+)
+
 model_id = "stabilityai/stable-diffusion-xl-base-0.9"
-model_id = "runwayml/stable-diffusion-v1-5"
 pipe_high_noise = DiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.float16, variant="fp16", use_safetensors=True, local_files_only=True)
-pipe_high_noise.scheduler = EulerDiscreteScheduler.from_config(pipe_high_noise.scheduler.config)
+pipe_high_noise.scheduler = scheduler
 pipe_high_noise.to("cuda")
 
 pipe_low_noise = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-refiner-0.9", torch_dtype=torch.float16, use_safetensors=True, variant="fp16")
-pipe_low_noise.scheduler = EulerDiscreteScheduler.from_config(pipe_low_noise.scheduler.config)
+pipe_low_noise.scheduler = scheduler
 pipe_low_noise.to("cuda")
 
-prompt = "A majestic lion jumping from a big stone at night"
+prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
 
-num_inference_steps = 40
-high_noise_frac = 0.8
 
-# seed = 0
-# seed_everything(seed)
+random_generator = torch.Generator()
+random_generator.manual_seed(0)
+
+num_inference_steps = 100
+high_noise_frac = 0.8
 
-image = pipe_high_noise(prompt=prompt, num_inference_steps=num_inference_steps, denoising_end=high_noise_frac, output_type="pt").images
+image = pipe_high_noise(prompt=prompt, num_inference_steps=num_inference_steps, denoising_end=high_noise_frac, output_type="latent").images
 image = pipe_low_noise(prompt=prompt, num_inference_steps=num_inference_steps, denoising_start=high_noise_frac, image=image).images[0]
 
 file_name = f"aaa_1"