File size: 139,688 Bytes
a64918d |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 |
# Introduction to 🤗 Diffusers
![diffusers_library](https://github.com/huggingface/diffusers/raw/main/docs/source/imgs/diffusers_library.jpg)
在这个notebook里,你将训练你的第一个扩散模型来**生成美丽的蝴蝶的图片 🦋.** 在此过程中,您将了解🤗 Diffuers库,它将为我们稍后将在课程中介绍的更高级的应用程序提供良好的基础
让我们开始吧!
## 你将学习到
在这个notebook中,你将:
- 看到一个强大的自定义扩散模型管道 (并了解到如何制作一个自己的版本)
- 通过以下方式创建您自己的迷你管道:
- 回顾扩散模型背后的核心思想
- 从hub中加载数据进行培训
- 探索如何使用scheduler将噪声添加到数据中
- 创建和训练一个UNet模型
- 将各个组件拼装在一起来形成一个工作管道(working pipelines)
- 编辑并运行一个脚本,用于初始化一个较长的训练,该脚本将处理
- 使用Accelerate来进行多GPU加速训练
- 实验日志记录以跟踪关键统计数据
- 将最终的模型上传到 Hugging Face Hub
❓如果您有任何问题,请发布在Hugging Face的Discord服务器 `#diffusion-models-class`频道中. 你可以在这里完成注册: https://huggingface.co/join/discord
## 预备知识
在进入notebook之前,你需要:
* 📖 阅读第一单元的材料
* 🤗在Hugging Face Hub上创建一个账户. 你可以在这里完成注册: https://huggingface.co/join
## 步骤 1: 设置
运行以下单元以安装diffusers库以及一些其他要求:
```python
%pip install -qq -U diffusers datasets transformers accelerate ftfy pyarrow
```
接下来,请前往 https://huggingface.co/settings/tokens 创建具有写权限的访问令牌:
![Screenshot from 2022-11-10 12-23-34.png](data:image/png;base64,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)
您可以使用命令行来通过此令牌登录(`huggingface-cli login`)或者运行以下单元来登录:
```python
from huggingface_hub import notebook_login
notebook_login()
```
Login successful
Your token has been saved to /root/.huggingface/token
然后您需要安装Git LFS来上传模型检查点:
```python
%%capture
!sudo apt -qq install git-lfs
!git config --global credential.helper store
```
最后,让我们导入将要使用的库,并定义一些方便函数,稍后我们将会在notebook中使用这些函数:
```python
import numpy as np
import torch
import torch.nn.functional as F
from matplotlib import pyplot as plt
from PIL import Image
def show_images(x):
"""Given a batch of images x, make a grid and convert to PIL"""
x = x * 0.5 + 0.5 # Map from (-1, 1) back to (0, 1)
grid = torchvision.utils.make_grid(x)
grid_im = grid.detach().cpu().permute(1, 2, 0).clip(0, 1) * 255
grid_im = Image.fromarray(np.array(grid_im).astype(np.uint8))
return grid_im
def make_grid(images, size=64):
"""Given a list of PIL images, stack them together into a line for easy viewing"""
output_im = Image.new("RGB", (size * len(images), size))
for i, im in enumerate(images):
output_im.paste(im.resize((size, size)), (i * size, 0))
return output_im
# Mac users may need device = 'mps' (untested)
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
```
好了,我们都准备好了!
## Dreambooth:即将到来的巅峰
如果你在过去几个月里看过人工智能相关的社交媒体,你就会听说过Stable Diffusion模型。这是一个强大的文本条件隐式扩散模型(别担心,我们将会学习这些概念)。但它有一个缺点:它不知道你或我长什么样,除非我们足够出名以至于互联网上发布我们的照片。
Dreambooth允许我们创建自己的模型变体,并对特定的面部、对象或样式有一些额外的了解。Corridor Crew制作了一段出色的视频,用一致的任务形象来讲故事,这是一个很好的说明了这种技术的能力的例子:
```python
from IPython.display import YouTubeVideo
YouTubeVideo("W4Mcuh38wyM")
```
<iframe
width="400"
height="300"
src="https://www.youtube.com/embed/W4Mcuh38wyM"
frameborder="0"
allowfullscreen
></iframe>
这是一个使用了[这个模型](https://huggingface.co/sd-dreambooth-library/mr-potato-head)的例子。该模型的训练使用了5张著名的儿童玩具"Mr Potato Head"的照片。
首先,让我们来加载这个管道。这些代码会自动从Hub下载模型权重等需要的文件。由于这个只有一行的demo需要下载数GB的数据,因此您可以跳过此单元格,只需欣赏样例输出即可!
```python
from diffusers import StableDiffusionPipeline
# Check out https://huggingface.co/sd-dreambooth-library for loads of models from the community
model_id = "sd-dreambooth-library/mr-potato-head"
# Load the pipeline
pipe = StableDiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.float16).to(
device
)
```
Fetching 15 files: 0%| | 0/15 [00:00<?, ?it/s]
管道加载完成后,我们可以使用以下命令生成图像:
```python
prompt = "an abstract oil painting of sks mr potato head by picasso"
image = pipe(prompt, num_inference_steps=50, guidance_scale=7.5).images[0]
image
```
0%| | 0/51 [00:00<?, ?it/s]
![png](01_introduction_to_diffusers_CN_files/01_introduction_to_diffusers_CN_22_1.png)
**练习:** 您可以使用不同的提示(prompt)自行尝试。在这个demo中, `sks` 是一个新概念的唯一标识符(UID) - 那么如果把它留空的话会发生什么事呢? 您还可以尝试改变`num_inference_steps`和 `guidance_scale`。这两个参数分别代表了采样步骤的数量(试试最多可以设为多低?)和模型将花多大的努力来尝试匹配提示。
这条神奇的管道里有很多事情!课程结束时,你将知道这一切是如何运作的。现在,让我们看看如何从头开始训练扩散模型。
## MVP (最简可实行管道)
🤗 Diffusers 的核心API被分为三个主要部分:
1. **管道**: 从高层出发设计的多种类函数,旨在以易部署的方式,能够做到快速通过主流预训练好的扩散模型来生成样本。
2. **模型**: 训练新的扩散模型时用到的主流网络架构, *e.g.* [UNet](https://arxiv.org/abs/1505.04597).
3. **管理器(or调度器)**: 在*推理*中使用多种不同的技巧来从噪声中生成图像,同时也可以生成在*训练*中所需的带噪图像。
管道对于末端使用者来说已经非常棒,但你既然已经参加了这门课程,我们就索性认为你想了解更多其中的奥秘!在此篇笔记结束之后,我们会来构建属于你自己,能够生成小蝴蝶图片的管道。下面这里会是最终的结果:
```python
from diffusers import DDPMPipeline
# Load the butterfly pipeline
butterfly_pipeline = DDPMPipeline.from_pretrained(
"johnowhitaker/ddpm-butterflies-32px"
).to(device)
# Create 8 images
images = butterfly_pipeline(batch_size=8).images
# View the result
make_grid(images)
```
Fetching 4 files: 0%| | 0/4 [00:00<?, ?it/s]
0%| | 0/1000 [00:00<?, ?it/s]
![png](01_introduction_to_diffusers_CN_files/01_introduction_to_diffusers_CN_26_2.png)
```python
也许这看起来并不如DreamBooth所展示的样例那样惊艳,但要知道我们在训练这些图画时只用了不到训练稳定扩散模型用到数据的0.0001%。说到模型训练,从引入介绍直到本单元,训练一个扩散模型的流程看起来像是这样:
1. 从训练集中加载一些图像
2. 加入噪声,从不同程度上
3. 把带了不同版本噪声的数据送进模型
4. 评估模型在对这些数据做增强去噪时的表现
5. 使用这个信息来更新模型权重,然后重复此步骤
我们会在接下来几节中逐一实现这些步骤,直至训练循环可以完整的运行,在这之后我们会来探索如何使用训练好的模型来生成样本,还有如何封装模型到管道中来轻松的分享给别人。下面我们来从数据入手吧。。。
```
## 步骤 2:下载一个训练数据集
在这个例子中,我们会用到一个来自Hugging Face Hub的图像集。具体来说, [是个1000张蝴蝶图像收藏集](https://huggingface.co/datasets/huggan/smithsonian_butterflies_subset). 这是个非常小的数据集, 我们这里也同时包含了已被注释的内容指向一些规模更大的选择。如果你想使用你自己的图像收藏,你也可以使用这里被注释掉的示例代码,从一个指定的文件夹来装载图片。
```python
```
```python
import torchvision
from datasets import load_dataset
from torchvision import transforms
dataset = load_dataset("huggan/smithsonian_butterflies_subset", split="train")
# We'll train on 32-pixel square images, but you can try larger sizes too
image_size = 32
# You can lower your batch size if you're running out of GPU memory
batch_size = 64
# Define data augmentations
preprocess = transforms.Compose(
[
transforms.Resize((image_size, image_size)), # Resize
transforms.RandomHorizontalFlip(), # Randomly flip (data augmentation)
transforms.ToTensor(), # Convert to tensor (0, 1)
transforms.Normalize([0.5], [0.5]), # Map to (-1, 1)
]
)
def transform(examples):
images = [preprocess(image.convert("RGB")) for image in examples["image"]]
return {"images": images}
dataset.set_transform(transform)
# Create a dataloader from the dataset to serve up the transformed images in batches
train_dataloader = torch.utils.data.DataLoader(
dataset, batch_size=batch_size, shuffle=True
)
```
我们可以从中取出一批图像数据来看一看他们是什么样子:
```python
xb = next(iter(train_dataloader))["images"].to(device)[:8]
print("X shape:", xb.shape)
show_images(xb).resize((8 * 64, 64), resample=Image.NEAREST)
```
X shape: torch.Size([8, 3, 32, 32])
/tmp/ipykernel_4278/3975082613.py:3: DeprecationWarning: NEAREST is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.NEAREST or Dither.NONE instead.
show_images(xb).resize((8 * 64, 64), resample=Image.NEAREST)
![png](01_introduction_to_diffusers_CN_files/01_introduction_to_diffusers_CN_32_2.png)
我们在此篇笔记中使用一个只有32像素的小图片集来保证训练时长是可控的。
## 步骤 3:定义管理器
我们的训练计划是,取出这些输入图片然后对它们增添噪声,在这之后把带噪的图片送入模型。在推理阶段,我们将用模型的预测值来不断迭代去除这些噪点。在`diffusers`中,这两个步骤都是由**管理器(调度器)**来处理的。
噪声管理器决定在不同的迭代周期时分别加入多少噪声。我们可以这样创建一个管理器,是取自于训练并能取样 'DDPM'的默认配置。 (基于此篇论文 ["Denoising Diffusion Probabalistic Models"](https://arxiv.org/abs/2006.11239):
```python
from diffusers import DDPMScheduler
noise_scheduler = DDPMScheduler(num_train_timesteps=1000)
```
DDPM论文这样来描述一个损坏过程,为每一个 '迭代周期'(timestep)增添一点少量的噪声。设在某个迭代周期有 $x_{t-1}$, 我们可以得到它的下一个版本 $x_t$ (比之前更多一点点噪声):<br><br>
$q(\mathbf{x}_t \vert \mathbf{x}_{t-1}) = \mathcal{N}(\mathbf{x}_t; \sqrt{1 - \beta_t} \mathbf{x}_{t-1}, \beta_t\mathbf{I}) \quad
q(\mathbf{x}_{1:T} \vert \mathbf{x}_0) = \prod^T_{t=1} q(\mathbf{x}_t \vert \mathbf{x}_{t-1})$<br><br>
这就是说,我们取 $x_{t-1}$, 给他一个$\sqrt{1 - \beta_t}$ 的系数,然后加上带有 $\beta_t$系数的噪声。 这里 $\beta$ 是根据一些管理器来为每一个t设定的,来决定每一个迭代周期中添加多少噪声。 现在,我们不想把这个推演进行500次来得到 $x_{500}$,所以我们用另一个公式来根据给出的 $x_0$ 计算得到任意t时刻的 $x_t$: <br><br>
$\begin{aligned}
q(\mathbf{x}_t \vert \mathbf{x}_0) &= \mathcal{N}(\mathbf{x}_t; \sqrt{\bar{\alpha}_t} \mathbf{x}_0, {(1 - \bar{\alpha}_t)} \mathbf{I})
\end{aligned}$ where $\bar{\alpha}_t = \prod_{i=1}^T \alpha_i$ and $\alpha_i = 1-\beta_i$<br><br>
数学符号看起来总是很可怕!好在有管理器来为我们完成这些运算。我们可以画出 $\sqrt{\bar{\alpha}_t}$ (标记为 `sqrt_alpha_prod`) 和 $\sqrt{(1 - \bar{\alpha}_t)}$ (标记为 `sqrt_one_minus_alpha_prod`) 来看一下输入(x)与噪声是如何在不同迭代周期中量化和叠加的:
```python
plt.plot(noise_scheduler.alphas_cumprod.cpu() ** 0.5, label=r"${\sqrt{\bar{\alpha}_t}}$")
plt.plot((1 - noise_scheduler.alphas_cumprod.cpu()) ** 0.5, label=r"$\sqrt{(1 - \bar{\alpha}_t)}$")
plt.legend(fontsize="x-large");
```
**练习:** 你可以探索一下使用不同的beta_start时曲线是如何变化的,beta_end 与 beta_schedule可以通过以下注释内容来修改:
```python
# One with too little noise added:
# noise_scheduler = DDPMScheduler(num_train_timesteps=1000, beta_start=0.001, beta_end=0.004)
# The 'cosine' schedule, which may be better for small image sizes:
# noise_scheduler = DDPMScheduler(num_train_timesteps=1000, beta_schedule='squaredcos_cap_v2')
```
不论你选择了哪一个管理器(调度器),我们现在都可以使用 `noise_scheduler.add_noise`功能来添加不同程度的噪声,就像这样:
```python
timesteps = torch.linspace(0, 999, 8).long().to(device)
noise = torch.randn_like(xb)
noisy_xb = noise_scheduler.add_noise(xb, noise, timesteps)
print("Noisy X shape", noisy_xb.shape)
show_images(noisy_xb).resize((8 * 64, 64), resample=Image.NEAREST)
```
Noisy X shape torch.Size([8, 3, 32, 32])
![png](01_introduction_to_diffusers_CN_files/01_introduction_to_diffusers_CN_41_1.png)
再来,在这里探索使用这里不同噪声管理器和预设参数带来的效果。 [This video](https://www.youtube.com/watch?v=fbLgFrlTnGU) 这个视频很好的解释了一些上述数学运算的细节,同时也是对此类概念的一个很好引入介绍。
## 步骤 4:定义模型
现在我们来到了核心部分:模型本身。
大多数扩散模型使用的模型结构都是一些[U-net]的变形(https://arxiv.org/abs/1505.04597) 也是我们在这里会用到的结构。
![](https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/unet-model.png)
概括来说:
- 输入模型中的图片经过几个由ResNetLayer构成的层,其中每层都使图片尺寸减半。
- 之后在经过同样数量的层把图片升采样。
- 其中还有对特征在相同位置的上、下采样层残差连接模块。
模型一个关键特征既是,输出图片尺寸与输入图片相同,这正是我们这里需要的。
Diffusers 为我们提供了一个易用的 `UNet2DModel` 类,用来在PyTorch创建所需要的结构。
我们来使用U-net为我们生成目标大小的图片吧。
注意这里 `down_block_types` 对应下采样模块 (上图中绿色部分), 而 `up_block_types` 对应上采样模块 (上图中红色部分):
```python
from diffusers import UNet2DModel
# Create a model
model = UNet2DModel(
sample_size=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=(64, 128, 128, 256), # More channels -> more parameters
down_block_types=(
"DownBlock2D", # a regular ResNet downsampling block
"DownBlock2D",
"AttnDownBlock2D", # a ResNet downsampling block with spatial self-attention
"AttnDownBlock2D",
),
up_block_types=(
"AttnUpBlock2D",
"AttnUpBlock2D", # a ResNet upsampling block with spatial self-attention
"UpBlock2D",
"UpBlock2D", # a regular ResNet upsampling block
),
)
model.to(device);
```
当在处理更高分辨率的输入时,你可能想用更多层的下、上采样模块,让注意力层只聚焦在最低分辨率(最底)层来减少内存消耗。我们在之后会讨论该如何实验来找到最适用与你手头场景的配置方法。
我们可以通过输入一批数据和随机的迭代周期数来看输出是否与输入尺寸相同:
```python
with torch.no_grad():
model_prediction = model(noisy_xb, timesteps).sample
model_prediction.shape
```
torch.Size([8, 3, 32, 32])
在下一步中,我们来看如何训练这个模型。
## 步骤 5:创建训练循环
终于可以训练了!下面这是PyTorch中经典的优化迭代循环,在这里一批一批的送入数据然后通过优化器来一步步更新模型参数 - 在这个样例中我们使用学习率为0.0004的AdamW优化器。
对于每一批的数据,我们要
- 随机取样几个迭代周期
- 根据预设为数据加入噪声
- 把带噪数据送入模型
- 使用MSE作为损失函数来比较目标结果与模型预测结果(在这里是加入噪声的场景)
- 通过 `loss.backward()` 与 `optimizer.step()`来更新模型参数
在这个过程中我们记录Loss值用来后续的绘图。
NB: 这段代码大概需10分钟来运行 - 你也可以跳过以下两块操作直接使用预训练好的模型。供你选择,你可以探索下通过缩小模型层中的通道数会对运行速度有多少提升。
官方扩散模型示例 [official diffusers training example](https://colab.research.google.com/github/huggingface/notebooks/blob/main/diffusers/training_example.ipynb) 训练了在更高分辨率数据集上的一个更大的模型,这也是一个极为精简训练循环的优秀示例:
```python
# Set the noise scheduler
noise_scheduler = DDPMScheduler(
num_train_timesteps=1000, beta_schedule="squaredcos_cap_v2"
)
# Training loop
optimizer = torch.optim.AdamW(model.parameters(), lr=4e-4)
losses = []
for epoch in range(30):
for step, batch in enumerate(train_dataloader):
clean_images = batch["images"].to(device)
# Sample noise to add to the images
noise = torch.randn(clean_images.shape).to(clean_images.device)
bs = clean_images.shape[0]
# Sample a random timestep for each image
timesteps = torch.randint(
0, noise_scheduler.num_train_timesteps, (bs,), device=clean_images.device
).long()
# Add noise to the clean images according to the noise magnitude at each timestep
noisy_images = noise_scheduler.add_noise(clean_images, noise, timesteps)
# Get the model prediction
noise_pred = model(noisy_images, timesteps, return_dict=False)[0]
# Calculate the loss
loss = F.mse_loss(noise_pred, noise)
loss.backward(loss)
losses.append(loss.item())
# Update the model parameters with the optimizer
optimizer.step()
optimizer.zero_grad()
if (epoch + 1) % 5 == 0:
loss_last_epoch = sum(losses[-len(train_dataloader) :]) / len(train_dataloader)
print(f"Epoch:{epoch+1}, loss: {loss_last_epoch}")
```
Epoch:5, loss: 0.16273280512541533
Epoch:10, loss: 0.11161588924005628
Epoch:15, loss: 0.10206522420048714
Epoch:20, loss: 0.08302505919709802
Epoch:25, loss: 0.07805309211835265
Epoch:30, loss: 0.07474562455900013
绘制loss曲线,我们能看到模型在一开始快速的收敛,接下来以一个较慢的速度持续优化(我们用右边log坐标轴的视图可以看的更清楚):
```python
fig, axs = plt.subplots(1, 2, figsize=(12, 4))
axs[0].plot(losses)
axs[1].plot(np.log(losses))
plt.show()
```
[<matplotlib.lines.Line2D at 0x7f40fc40b7c0>]
![png](01_introduction_to_diffusers_CN_files/01_introduction_to_diffusers_CN_55_1.png)
你可以选择运行上面的代码,也可以这样通过管道来调用模型:
```python
# Uncomment to instead load the model I trained earlier:
# model = butterfly_pipeline.unet
```
## 步骤 6:生成图像
我们怎么从这个模型中得到图像呢?
### 方法 1:建立一个管道:
```python
from diffusers import DDPMPipeline
image_pipe = DDPMPipeline(unet=model, scheduler=noise_scheduler)
```
```python
pipeline_output = image_pipe()
pipeline_output.images[0]
```
0%| | 0/1000 [00:00<?, ?it/s]
![png](01_introduction_to_diffusers_CN_files/01_introduction_to_diffusers_CN_61_1.png)
我们可以在本地文件夹这样保存一个管道:
```python
image_pipe.save_pretrained("my_pipeline")
```
检查文件夹的内容:
```python
!ls my_pipeline/
```
model_index.json scheduler unet
这里 `scheduler` 与 `unet` 子文件夹中包含了生成图像所需的全部组件。比如,在 `unet` 文件中能看到模型参数(`diffusion_pytorch_model.bin`) 与描述模型结构的配置文件。
```python
!ls my_pipeline/unet/
```
config.json diffusion_pytorch_model.bin
以上,这些文件包含了重新建立一个管道的全部内容。你可以手动把它们上传到hub来与他人分享你制作的管道,或使用下一节的API方法来记载。
### 方法 2:写一个取样循环
如果你去查看了管道中的forward方法,你可以看到在运行 `image_pipe()`时发生了什么:
```python
# ??image_pipe.forward
```
从随机噪声开始,遍历管理器的迭代周期来看从最嘈杂直到最微小的噪声变化,基于模型的预测一步步减少一些噪声:
```python
# Random starting point (8 random images):
sample = torch.randn(8, 3, 32, 32).to(device)
for i, t in enumerate(noise_scheduler.timesteps):
# Get model pred
with torch.no_grad():
residual = model(sample, t).sample
# Update sample with step
sample = noise_scheduler.step(residual, t, sample).prev_sample
show_images(sample)
```
![png](01_introduction_to_diffusers_CN_files/01_introduction_to_diffusers_CN_72_0.png)
`noise_scheduler.step()` 函数相应做了 `sample`(取样)时的数学运算。其实有很多取样的方法 - 在下一个单元我们将看到在已有模型的基础上如何换一个不同的取样器来加速图片生成,也会讲到更多从扩散模型中取样的背后原理。
## 步骤 7:把你的模型Push到Hub
在上面的例子中我们把管道保存在了本地。把模型push到hub上,我们会需要建立模型和相应文件的仓库名。我们根据你的选择(模型ID)来决定仓库的名字(大胆的去替换掉 `model_name`吧;需要包含你的用户名,`get_full_repo_name()`会帮你做到):
```python
from huggingface_hub import get_full_repo_name
model_name = "sd-class-butterflies-32"
hub_model_id = get_full_repo_name(model_name)
hub_model_id
```
'lewtun/sd-class-butterflies-32'
然后,在 🤗 Hub上创建模型仓库并push它吧:
```python
from huggingface_hub import HfApi, create_repo
create_repo(hub_model_id)
api = HfApi()
api.upload_folder(
folder_path="my_pipeline/scheduler", path_in_repo="", repo_id=hub_model_id
)
api.upload_folder(folder_path="my_pipeline/unet", path_in_repo="", repo_id=hub_model_id)
api.upload_file(
path_or_fileobj="my_pipeline/model_index.json",
path_in_repo="model_index.json",
repo_id=hub_model_id,
)
```
'https://huggingface.co/lewtun/sd-class-butterflies-32/blob/main/model_index.json'
最后一件事是创建一个超棒的模型卡,如此,我们的蝴蝶生成器可以轻松的在Hub上被找到(请在描述中随意发挥!):
```python
from huggingface_hub import ModelCard
content = f"""
---
license: mit
tags:
- pytorch
- diffusers
- unconditional-image-generation
- diffusion-models-class
---
# Model Card for Unit 1 of the [Diffusion Models Class 🧨](https://github.com/huggingface/diffusion-models-class)
This model is a diffusion model for unconditional image generation of cute 🦋.
## Usage
```python
from diffusers import DDPMPipeline
pipeline = DDPMPipeline.from_pretrained('{hub_model_id}')
image = pipeline().images[0]
image
```
"""
card = ModelCard(content)
card.push_to_hub(hub_model_id)
```
现在模型已经在Hub上了,你可以这样从任何地方使用 `DDPMPipeline`的 `from_pretrained()`方法来下来它:
```python
from diffusers import DDPMPipeline
image_pipe = DDPMPipeline.from_pretrained(hub_model_id)
pipeline_output = image_pipe()
pipeline_output.images[0]
```
Fetching 4 files: 0%| | 0/4 [00:00<?, ?it/s]
0%| | 0/1000 [00:00<?, ?it/s]
![png](01_introduction_to_diffusers_CN_files/01_introduction_to_diffusers_CN_81_2.png)
太棒了,成功了!
# 使用🤗 Accelerate来扩大规模
这篇笔记是用来教学,为此我尽力保证代码的简洁与轻量化。但也因为这样,我们也略去了一些内容你也许在使用更多数据训练一个更大的模式时,可能所需要用到的内容,如多块GPU支持,进度记录和样例图片,用于支持更大batchsize的导数记录功能,自动上传模型等等。好在这些功能大多数在这个示例代码中包含 [here](https://github.com/huggingface/diffusers/raw/main/examples/unconditional_image_generation/train_unconditional.py).
你可以这样下载该文件:
```python
!wget https://github.com/huggingface/diffusers/raw/main/examples/unconditional_image_generation/train_unconditional.py
```
打开文件,你就可以看到模型是怎么定义的,以及有哪些可选的配置参数。我使用如下命令运行了该代码:
```python
# Let's give our new model a name for the Hub
model_name = "sd-class-butterflies-64"
hub_model_id = get_full_repo_name(model_name)
hub_model_id
```
'lewtun/sd-class-butterflies-64'
```python
!accelerate launch train_unconditional.py \
--dataset_name="huggan/smithsonian_butterflies_subset" \
--resolution=64 \
--output_dir={model_name} \
--train_batch_size=32 \
--num_epochs=50 \
--gradient_accumulation_steps=1 \
--learning_rate=1e-4 \
--lr_warmup_steps=500 \
--mixed_precision="no"
```
如之前一样,把模型push到hub,并且创建一个超酷的模型卡(按你的想法随意填写!):
```python
create_repo(hub_model_id)
api = HfApi()
api.upload_folder(
folder_path=f"{model_name}/scheduler", path_in_repo="", repo_id=hub_model_id
)
api.upload_folder(
folder_path=f"{model_name}/unet", path_in_repo="", repo_id=hub_model_id
)
api.upload_file(
path_or_fileobj=f"{model_name}/model_index.json",
path_in_repo="model_index.json",
repo_id=hub_model_id,
)
content = f"""
---
license: mit
tags:
- pytorch
- diffusers
- unconditional-image-generation
- diffusion-models-class
---
# Model Card for Unit 1 of the [Diffusion Models Class 🧨](https://github.com/huggingface/diffusion-models-class)
This model is a diffusion model for unconditional image generation of cute 🦋.
## Usage
```python
from diffusers import DDPMPipeline
pipeline = DDPMPipeline.from_pretrained('{hub_model_id}')
image = pipeline().images[0]
image
```
"""
card = ModelCard(content)
card.push_to_hub(hub_model_id)
```
'https://huggingface.co/lewtun/sd-class-butterflies-64/blob/main/README.md'
大概45分钟之后,得到这样的结果:
```python
pipeline = DDPMPipeline.from_pretrained(hub_model_id).to(device)
images = pipeline(batch_size=8).images
make_grid(images)
```
0%| | 0/1000 [00:00<?, ?it/s]
![png](01_introduction_to_diffusers_CN_files/01_introduction_to_diffusers_CN_91_1.png)
**练习:** 看看你能不能找到训练出在短时内能得到满意结果的模型训练设置参数,并与社群分享你的发现。阅读这些脚本看看你能不能理解它们,如果遇到了一些看上去令人迷惑的地方,请向大家提问来寻求解释。
# 更高阶的探索之路
希望这些能够让你初步了解可以使用 🤗 Diffusers library来做什么!可能一些后续的步骤是这样:
- 尝试在新数据集上训练一个无限制的扩散模型 - 如果你能直接自己完成那就太好了 [create one yourself](https://huggingface.co/docs/datasets/image_dataset). 你可以在Hub这里找到一些能完成这个任务的超棒图像数据集 [HugGan organization](https://huggingface.co/huggan). 如果你不想等待模型训练太久的话,一定记得对图片做下采样!
- 试试用DreamBooth来创建你自己定制的扩散模型管道,看看这里 [this Space](https://huggingface.co/spaces/multimodalart/dreambooth-training) 或者 [this notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/diffusers/sd_dreambooth_training.ipynb)
- 修改训练脚本来探索使用不同的UNet超参数(层数深度,通道数等等),不同的噪声管理器等等。
- 来瞧瞧 [Diffusion Models from Scratch](https://github.com/huggingface/diffusion-models-class/blob/main/unit1/02_diffusion_models_from_scratch.ipynb) 在本单元的核心思想之上的一些不同看法。
祝好,敬请关注第2单元!
```python
```
|