img2img.ipynb

img2img_nb.py

"""make variations of input image"""

import argparse
import glob
import os
import sys
import time
from contextlib import nullcontext
from itertools import islice

import numpy as np
import PIL
import torch
from einops import rearrange, repeat
from ldm.models.diffusion.ddim import DDIMSampler
from ldm.models.diffusion.plms import PLMSSampler
from ldm.util import instantiate_from_config
from omegaconf import OmegaConf
from PIL import Image
from pytorch_lightning import seed_everything
from torch import autocast
from torchvision.utils import make_grid
from tqdm import tqdm, trange


def chunk(it, size):
    it = iter(it)
    return iter(lambda: tuple(islice(it, size)), ())


def load_model_from_config(config, ckpt, verbose=False):
    print(f"Loading model from {ckpt}")
    pl_sd = torch.load(ckpt, map_location="cpu")
    if "global_step" in pl_sd:
        print(f"Global Step: {pl_sd['global_step']}")
    sd = pl_sd["state_dict"]
    model = instantiate_from_config(config.model)
    m, u = model.load_state_dict(sd, strict=False)
    if len(m) > 0 and verbose:
        print("missing keys:")
        print(m)
    if len(u) > 0 and verbose:
        print("unexpected keys:")
        print(u)

    model.cuda()
    model.eval()
    return model


def load_img(path):
    image = Image.open(path).convert("RGB")
    w, h = image.size
    print(f"loaded input image of size ({w}, {h}) from {path}")
    # resize to integer multiple of 32
    w, h = map(lambda x: x - x % 32, (w, h))
    image = image.resize((w, h), resample=PIL.Image.LANCZOS)
    image = np.array(image).astype(np.float32) / 255.0
    image = image[None].transpose(0, 3, 1, 2)
    image = torch.from_numpy(image)
    return 2.*image - 1.


def load_model(config="configs/stable-diffusion/v1-inference.yaml", ckpt="models/ldm/stable-diffusion-v1/model.ckpt"):
    config = OmegaConf.load(f"{config}")
    model = load_model_from_config(config, f"{ckpt}")
    model = model.to(torch.float16)

    return model


def main(model, prompt: str, init_img: str, skip_grid=True, ddim_steps=50, plms=True, fixed_code=False, ddim_eta=0.0, n_iter=1, C=4, f=8, n_samples=1, n_rows=0, scale=5.0, strength=0.75, from_file=None, seed=42, precision="autocast", prompt_correction=[]):
    seed_everything(seed)

    device = torch.device(
        "cuda") if torch.cuda.is_available() else torch.device("cpu")
    model = model.to(device)

    if plms:
        raise NotImplementedError("PLMS sampler not (yet) supported")
        sampler = PLMSSampler(model)
    else:
        sampler = DDIMSampler(model)

    batch_size = n_samples
    n_rows = n_rows if n_rows > 0 else batch_size
    if not from_file:
        prompt = prompt
        assert prompt is not None
        data = [batch_size * [prompt]]

    else:
        print(f"reading prompts from {from_file}")
        with open(from_file, "r") as f:
            data = f.read().splitlines()
            data = list(chunk(data, batch_size))

    assert os.path.isfile(init_img)
    init_image = load_img(init_img).to(torch.float16).to(device)
    init_image = repeat(init_image, '1 ... -> b ...', b=batch_size)
    init_latent = model.get_first_stage_encoding(
        model.encode_first_stage(init_image))  # move to latent space

    sampler.make_schedule(ddim_num_steps=ddim_steps,
                          ddim_eta=ddim_eta, verbose=False)

    assert 0. <= strength <= 1., 'can only work with strength in [0.0, 1.0]'
    t_enc = int(strength * ddim_steps)
    print(f"target t_enc is {t_enc} steps")

    images = []
    precision_scope = autocast if precision == "autocast" else nullcontext
    with torch.no_grad():
        with precision_scope("cuda"):
            with model.ema_scope():
                tic = time.time()
                all_samples = list()
                for n in trange(n_iter, desc="Sampling"):
                    for prompts in tqdm(data, desc="data"):
                        uc = None
                        if scale != 1.0:
                            uc = model.get_learned_conditioning(
                                batch_size * [""])
                        if isinstance(prompts, tuple):
                            prompts = list(prompts)
                        c = model.get_learned_conditioning(prompts)
                        for pw in prompt_correction:
                            pw = pw.split('::')
                            p, weight = pw[:-1], float(pw[-1])
                            c += weight * \
                                model.get_learned_conditioning(list(p))

                        # encode (scaled latent)
                        z_enc = sampler.stochastic_encode(
                            init_latent, torch.tensor([t_enc]*batch_size).to(device))
                        # decode it
                        samples = sampler.decode(z_enc, c, t_enc, unconditional_guidance_scale=scale,
                                                 unconditional_conditioning=uc,)

                        x_samples = model.decode_first_stage(samples)
                        x_samples = torch.clamp(
                            (x_samples + 1.0) / 2.0, min=0.0, max=1.0)

                        for x_sample in x_samples:
                            x_sample = 255. * \
                                rearrange(x_sample.cpu().numpy(),
                                          'c h w -> h w c')
                            images.append(Image.fromarray(
                                x_sample.astype(np.uint8)))

                toc = time.time()

    return images


if __name__ == "__main__":
    main()