1

#!/usr/bin/env python3
from diffusers import DiffusionPipeline, EulerDiscreteScheduler, StableDiffusionPipeline, KDPM2DiscreteScheduler, StableDiffusionImg2ImgPipeline, HeunDiscreteScheduler, KDPM2AncestralDiscreteScheduler, DDIMScheduler
from diffusers import StableDiffusionXLPipeline, StableDiffusionXLImg2ImgPipeline
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

if use_diffusers:
    start_time = time.time()
    pipe = StableDiffusionXLPipeline.from_pretrained("stabilityai/stable-diffusion-xl-base-0.9", torch_dtype=torch.float16, variant="fp16", use_safetensors=True, local_files_only=True)
    pipe.to("cuda")

    if use_refiner:
        refiner = StableDiffusionXLImg2ImgPipeline.from_pretrained("stabilityai/stable-diffusion-xl-refiner-0.9", torch_dtype=torch.float16, use_safetensors=True, variant="fp16")
        refiner.to("cuda")
        # refiner.enable_sequential_cpu_offload()
else:
    pipe = StableDiffusionXLPipeline.from_single_file("https://huggingface.co/stabilityai/stable-diffusion-xl-base-0.9/blob/main/sd_xl_base_0.9.safetensors", torch_dtype=torch.float16, use_safetensors=True)
    pipe.to("cuda")

    if use_refiner:
        refiner = StableDiffusionXLImg2ImgPipeline.from_single_file("https://huggingface.co/stabilityai/stable-diffusion-xl-refiner-0.9/blob/main/sd_xl_refiner_0.9.safetensors", torch_dtype=torch.float16, use_safetensors=True)
        refiner.to("cuda")


prompt = "An astronaut riding a green horse on Mars"
for steps in [24, 27, 31]:
    for denoising_end in [0.63, 0.66, 0.67, 0.71]:
        seed = 0
        seed_everything(seed)
        image = pipe(prompt=prompt, num_inference_steps=40, denoising_end=0.675, output_type="latent" if use_refiner else "pil").images[0]
# image = pipe(prompt=prompt, output_type="latent" if use_refiner else "pil").images[0]

        if use_refiner:
            image = refiner(prompt=prompt, num_inference_steps=40, denoising_start=0.675, image=image[None, :]).images[0]

# pipe.unet.to(memory_format=torch.channels_last)
# pipe(prompt=prompt, num_inference_steps=2).images[0]

# image = pipe(prompt=prompt, num_images_per_prompt=1, num_inference_steps=40, output_type="latent").images

        file_name = f"aaa_{seed}"
        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/ediffi_sdxl/{file_name}.png")