算力平台:
[[open-in-colab]]
训练扩散模型
无条件图像生成是扩散模型的一个流行应用，它能够生成看起来像训练数据集中图像的图像。通常，通过对预训练模型在特定数据集上进行微调可以获得最佳结果。你可以在Hub上找到许多这样的检查点，但如果你找不到喜欢的模型，你总是可以训练自己的模型！
本教程将教你如何在一个Smithsonian Butterflies数据集的子集上从头开始训练一个[UNet2DModel]，以生成你自己的🦋蝴蝶🦋。
在开始之前，请确保你已经安装了🤗 Datasets来加载和预处理图像数据集，以及🤗 Accelerate，以简化在任意数量的GPU上的训练。以下命令还将安装TensorBoard来可视化训练指标（你也可以使用Weights & Biases来跟踪你的训练）。
py
# uncomment to install the necessary libraries in Colab
#!pip install diffusers[training]我们鼓励你与社区分享你的模型，为此，你需要登录你的 Hugging Face 账户（如果你还没有账户，可以在这里创建一个！）。你可以从笔记本中登录，并在提示时输入你的令牌。确保你的令牌具有写入权限。
py
>>> from huggingface_hub import notebook_login
>>> notebook_login()或者从终端登录：
bash
huggingface-cli login由于模型检查点文件非常大，请安装 Git-LFS 来管理这些大文件的版本：
bash
!sudo apt -qq install git-lfs
!git config --global credential.helper store训练配置
为了方便，创建一个包含训练超参数的 TrainingConfig 类（可以根据需要调整这些参数）：
py
>>> from dataclasses import dataclass
>>> @dataclass
... class TrainingConfig:
... image_size = 128 # the generated image resolution
... train_batch_size = 16
... eval_batch_size = 16 # how many images to sample during evaluation
... num_epochs = 50
... gradient_accumulation_steps = 1
... learning_rate = 1e-4
... lr_warmup_steps = 500
... save_image_epochs = 10
... save_model_epochs = 30
... mixed_precision = "fp16" # `no` for float32, `fp16` for automatic mixed precision
... output_dir = "ddpm-butterflies-128" # the model name locally and on the HF Hub
... push_to_hub = True # whether to upload the saved model to the HF Hub
... hub_model_id = "<your-username>/<my-awesome-model>" # the name of the repository to create on the HF Hub
... hub_private_repo = False
... overwrite_output_dir = True # overwrite the old model when re-running the notebook
... seed = 0
>>> config = TrainingConfig()加载数据集
你可以使用 🤗 Datasets 库轻松加载 Smithsonian Butterflies 数据集：
py
>>> from datasets import load_dataset
>>> config.dataset_name = "huggan/smithsonian_butterflies_subset"
>>> dataset = load_dataset(config.dataset_name, split="train")🤗 Datasets 使用 [~datasets.Image] 功能自动解码图像数据并将其加载为 PIL.Image，我们可以对其进行可视化：
py
>>> import matplotlib.pyplot as plt
>>> fig, axs = plt.subplots(1, 4, figsize=(16, 4))
>>> for i, image in enumerate(dataset[:4]["image"]):
... axs[i].imshow(image)
... axs[i].set_axis_off()
>>> fig.show()
不过，这些图像的大小各不相同，因此你需要先对它们进行预处理：
Resize将图像大小调整为config.image_size中定义的大小。RandomHorizontalFlip通过随机镜像图像来增强数据集。Normalize很重要，它将像素值重新缩放到 [-1, 1] 范围内，这是模型所期望的。
py
>>> from torchvision import transforms
>>> preprocess = transforms.Compose(
... [
... transforms.Resize((config.image_size, config.image_size)),
... transforms.RandomHorizontalFlip(),
... transforms.ToTensor(),
... transforms.Normalize([0.5], [0.5]),
... ]
... )使用 🤗 Datasets 的 [~datasets.Dataset.set_transform] 方法在训练过程中动态应用 preprocess 函数：
py
>>> def transform(examples):
... images = [preprocess(image.convert("RGB")) for image in examples["image"]]
... return {"images": images}
>>> dataset.set_transform(transform)可以随意再次可视化图像，以确认它们已被调整大小。现在你可以将数据集包装在 DataLoader 中进行训练了！
py
>>> import torch
>>> train_dataloader = torch.utils.data.DataLoader(dataset, batch_size=config.train_batch_size, shuffle=True)创建一个 UNet2DModel
在 🧨 Diffusers 中，可以通过所需的参数轻松地从模型类创建预训练模型。例如，要创建一个 [UNet2DModel]：
py
>>> from diffusers import UNet2DModel
>>> model = UNet2DModel(
... sample_size=config.image_size, # the target image resolution
... in_channels=3, # the number of input channels, 3 for RGB images
... out_channels=3, # the number of output channels
... layers_per_block=2, # how many ResNet layers to use per UNet block
... block_out_channels=(128, 128, 256, 256, 512, 512), # the number of output channels for each UNet block
... down_block_types=(
... "DownBlock2D", # a regular ResNet downsampling block
... "DownBlock2D",
... "DownBlock2D",
... "DownBlock2D",
... "AttnDownBlock2D", # a ResNet downsampling block with spatial self-attention
... "DownBlock2D",
... ),
... up_block_types=(
... "UpBlock2D", # a regular ResNet upsampling block
... "AttnUpBlock2D", # a ResNet upsampling block with spatial self-attention
... "UpBlock2D",
... "UpBlock2D",
... "UpBlock2D",
... "UpBlock2D",
... ),
... )通常，快速检查样本图像形状是否与模型输出形状匹配是一个好主意：
py
>>> sample_image = dataset[0]["images"].unsqueeze(0)
>>> print("Input shape:", sample_image.shape)
Input shape: torch.Size([1, 3, 128, 128])
>>> print("Output shape:", model(sample_image, timestep=0).sample.shape)
Output shape: torch.Size([1, 3, 128, 128])太好了！接下来，你需要一个调度器来给图像添加一些噪声。
创建一个调度器
调度器的行为取决于你是将模型用于训练还是推理。在推理过程中，调度器从噪声中生成图像。在训练过程中，调度器从扩散过程中的特定点获取模型输出（或样本），并根据噪声调度和更新规则对图像应用噪声。
让我们来看看[DDPMScheduler]，并使用add_noise方法为之前的sample_image添加一些随机噪声：
py
>>> import torch
>>> from PIL import Image
>>> from diffusers import DDPMScheduler
>>> noise_scheduler = DDPMScheduler(num_train_timesteps=1000)
>>> noise = torch.randn(sample_image.shape)
>>> timesteps = torch.LongTensor([50])
>>> noisy_image = noise_scheduler.add_noise(sample_image, noise, timesteps)
>>> Image.fromarray(((noisy_image.permute(0, 2, 3, 1) + 1.0) * 127.5).type(torch.uint8).numpy()[0])
模型的训练目标是预测添加到图像中的噪声。这一步的损失可以通过以下方式计算：
py
>>> import torch.nn.functional as F
>>> noise_pred = model(noisy_image, timesteps).sample
>>> loss = F.mse_loss(noise_pred, noise)训练模型
到目前为止，你已经具备了开始训练模型的大部分组件，剩下的就是将所有内容整合在一起。
首先，你需要一个优化器和一个学习率调度器：
py
>>> from diffusers.optimization import get_cosine_schedule_with_warmup
>>> optimizer = torch.optim.AdamW(model.parameters(), lr=config.learning_rate)
>>> lr_scheduler = get_cosine_schedule_with_warmup(
... optimizer=optimizer,
... num_warmup_steps=config.lr_warmup_steps,
... num_training_steps=(len(train_dataloader) * config.num_epochs),
... )然后，你需要一种评估模型的方法。对于评估，你可以使用[DDPMPipeline]生成一批样本图像，并将其保存为网格：
py
>>> from diffusers import DDPMPipeline
>>> from diffusers.utils import make_image_grid
>>> import os
>>> def evaluate(config, epoch, pipeline):
... # Sample some images from random noise (this is the backward diffusion process).
... # The default pipeline output type is `List[PIL.Image]`
... images = pipeline(
... batch_size=config.eval_batch_size,
... generator=torch.Generator(device='cpu').manual_seed(config.seed), # Use a separate torch generator to avoid rewinding the random state of the main training loop
... ).images
... # Make a grid out of the images
... image_grid = make_image_grid(images, rows=4, cols=4)
... # Save the images
... test_dir = os.path.join(config.output_dir, "samples")
... os.makedirs(test_dir, exist_ok=True)
... image_grid.save(f"{test_dir}/{epoch:04d}.png")现在，你可以将所有这些组件一起包装在一个训练循环中，使用🤗 Accelerate 进行简单的 TensorBoard 日志记录、梯度累积和混合精度训练。要将模型上传到 Hub，编写一个函数来获取你的仓库名称和信息，然后将其推送到 Hub。
py
>>> from accelerate import Accelerator
>>> from huggingface_hub import create_repo, upload_folder
>>> from tqdm.auto import tqdm
>>> from pathlib import Path
>>> import os
>>> def train_loop(config, model, noise_scheduler, optimizer, train_dataloader, lr_scheduler):
... # Initialize accelerator and tensorboard logging
... accelerator = Accelerator(
... mixed_precision=config.mixed_precision,
... gradient_accumulation_steps=config.gradient_accumulation_steps,
... log_with="tensorboard",
... project_dir=os.path.join(config.output_dir, "logs"),
... )
... if accelerator.is_main_process:
... if config.output_dir is not None:
... os.makedirs(config.output_dir, exist_ok=True)
... if config.push_to_hub:
... repo_id = create_repo(
... repo_id=config.hub_model_id or Path(config.output_dir).name, exist_ok=True
... ).repo_id
... accelerator.init_trackers("train_example")
... # Prepare everything
... # There is no specific order to remember, you just need to unpack the
... # objects in the same order you gave them to the prepare method.
... model, optimizer, train_dataloader, lr_scheduler = accelerator.prepare(
... model, optimizer, train_dataloader, lr_scheduler
... )
... global_step = 0
... # Now you train the model
... for epoch in range(config.num_epochs):
... progress_bar = tqdm(total=len(train_dataloader), disable=not accelerator.is_local_main_process)
... progress_bar.set_description(f"Epoch {epoch}")
... for step, batch in enumerate(train_dataloader):
... clean_images = batch["images"]
... # Sample noise to add to the images
... noise = torch.randn(clean_images.shape, device=clean_images.device)
... bs = clean_images.shape[0]
... # Sample a random timestep for each image
... timesteps = torch.randint(
... 0, noise_scheduler.config.num_train_timesteps, (bs,), device=clean_images.device,
... dtype=torch.int64
... )
... # Add noise to the clean images according to the noise magnitude at each timestep
... # (this is the forward diffusion process)
... noisy_images = noise_scheduler.add_noise(clean_images, noise, timesteps)
... with accelerator.accumulate(model):
... # Predict the noise residual
... noise_pred = model(noisy_images, timesteps, return_dict=False)[0]
... loss = F.mse_loss(noise_pred, noise)
... accelerator.backward(loss)
... if accelerator.sync_gradients:
... accelerator.clip_grad_norm_(model.parameters(), 1.0)
... optimizer.step()
... lr_scheduler.step()
... optimizer.zero_grad()
... progress_bar.update(1)
... logs = {"loss": loss.detach().item(), "lr": lr_scheduler.get_last_lr()[0], "step": global_step}
... progress_bar.set_postfix(**logs)
... accelerator.log(logs, step=global_step)
... global_step += 1
... # After each epoch you optionally sample some demo images with evaluate() and save the model
... if accelerator.is_main_process:
... pipeline = DDPMPipeline(unet=accelerator.unwrap_model(model), scheduler=noise_scheduler)
... if (epoch + 1) % config.save_image_epochs == 0 or epoch == config.num_epochs - 1:
... evaluate(config, epoch, pipeline)
... if (epoch + 1) % config.save_model_epochs == 0 or epoch == config.num_epochs - 1:
... if config.push_to_hub:
... upload_folder(
... repo_id=repo_id,
... folder_path=config.output_dir,
... commit_message=f"Epoch {epoch}",
... ignore_patterns=["step_*", "epoch_*"],
... )
... else:
... pipeline.save_pretrained(config.output_dir)呼，那可是一大段代码！但你终于准备好使用 🤗 Accelerate 的 [~accelerate.notebook_launcher] 函数来启动训练了。将训练循环、所有训练参数以及用于训练的进程数（你可以将此值更改为可用的 GPU 数量）传递给该函数：
py
>>> from accelerate import notebook_launcher
>>> args = (config, model, noise_scheduler, optimizer, train_dataloader, lr_scheduler)
>>> notebook_launcher(train_loop, args, num_processes=1)训练完成后，看看你的扩散模型生成的最终 🦋 图像 🦋 吧！
py
>>> import glob
>>> sample_images = sorted(glob.glob(f"{config.output_dir}/samples/*.png"))
>>> Image.open(sample_images[-1])
下一步
无条件图像生成是可训练的任务之一。你可以通过访问🧨 Diffusers 训练示例页面来探索其他任务和训练技术。以下是你可以学习的一些示例：
- 文本反转，一种算法，教模型一个特定的视觉概念并将其整合到生成的图像中。
- DreamBooth，一种技术，用于根据给定的几张输入图像生成主题的个性化图像。
- 指南，关于如何在自己的数据集上微调 Stable Diffusion 模型。
- 指南，关于使用 LoRA，一种内存高效的技术，用于更快地微调非常大的模型。