StabilityAI的DeepFloyd Lab的 IF

我们介绍了 DeepFloyd IF，这是一种新颖的最先进的开源文本转图像模型，具有高度的照片级真实感和语言理解能力。DeepFloyd IF 是一个由冻结文本编码器和三个级联像素扩散模块组成的模块：一个基于文本提示生成 64x64 像素图像的基本模型和两个超分辨率模型，每个模型都旨在生成分辨率不断提高的图像：256x256 像素和 1024x1024 像素。该模型的所有阶段都使用基于 T5 变换器的冻结文本编码器来提取文本嵌入，然后将其输入到通过交叉注意力和注意力池增强的 UNet 架构中。结果是一个高效的模型，其表现优于目前最先进的模型，在 COCO 数据集上实现了 6.66 的零样本 FID 分数。我们的工作强调了更大的 UNet 架构在级联扩散模型第一阶段的潜力，并描绘了文本转图像合成的光明未来。

受到具有深度语言理解能力的逼真文本到图像扩散模型的启发

使用所有 IF 模型的最低要求：

IF-I-XL（4.3B 文本到 64x64 基本模块）和 IF-II-L（1.2B 到 256x256 升频模块）配备 16GB vRAM
IF-I-XL（4.3B 文本至 64x64 基本模块）和 IF-II-L（1.2B 至 256x256 升频器模块）和 Stable x4（至 1024x1024 升频器）均配备 24GB vRAM
xformers并设置环境变量FORCE_MEM_EFFICIENT_ATTN=1

快速开始

pip install deepfloyd_if==1.0.2rc0
pip install xformers==0.0.16
pip install git+https://github.com/openai/CLIP.git --no-deps

本地笔记本

这里的Jupyter Notebook 中提供了梦幻、风格转换、超分辨率或修复模式。

与🤗扩散器集成

IF 还与 🤗 Hugging Face Diffusers 库集成。

扩散器单独运行每个阶段，允许用户自定义图像生成过程并轻松检查中间结果。

例子

在使用 IF 之前，您需要接受其使用条件。具体操作如下：

确保拥有Hugging Face 账户并登录
在DeepFloyd/IF-I-XL-v1.0模型卡上接受许可证
确保本地登录。安装huggingface_hub

pip install huggingface_hub --upgrade

在 Python shell 中运行登录函数

from huggingface_hub import login
login()

<clipboard-copy aria-label="Copy" class="ClipboardButton btn btn-invisible js-clipboard-copy m-2 p-0 tooltipped-no-delay d-flex flex-justify-center flex-items-center" data-copy-feedback="Copied!" data-tooltip-direction="w" value="from huggingface_hub import login

login()" tabindex="0" role="button">

并输入您的Hugging Face Hub 访问令牌。

接下来我们安装diffusers依赖项：

pip install diffusers accelerate transformers safetensors

现在我们可以在本地运行该模型了。

默认情况下，diffusers利用模型 CPU 卸载来运行整个 IF 管道，且仅需 14 GB 的 VRAM。

如果您正在使用torch>=2.0.0，请确保删除所有 enable_xformers_memory_efficient_attention() 功能。

from diffusers import DiffusionPipeline
from diffusers.utils import pt_to_pil
import torch
# stage 1
stage_1 = DiffusionPipeline.from_pretrained("DeepFloyd/IF-I-XL-v1.0", variant="fp16", torch_dtype=torch.float16)
stage_1.enable_xformers_memory_efficient_attention()  # remove line if torch.version >= 2.0.0
stage_1.enable_model_cpu_offload()
# stage 2
stage_2 = DiffusionPipeline.from_pretrained(
"DeepFloyd/IF-II-L-v1.0", text_encoder=None, variant="fp16", torch_dtype=torch.float16
)
stage_2.enable_xformers_memory_efficient_attention()  # remove line if torch.version >= 2.0.0
stage_2.enable_model_cpu_offload()
# stage 3
safety_modules = {"feature_extractor": stage_1.feature_extractor, "safety_checker": stage_1.safety_checker, "watermarker": stage_1.watermarker}
stage_3 = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-x4-upscaler", **safety_modules, torch_dtype=torch.float16)
stage_3.enable_xformers_memory_efficient_attention()  # remove line if torch.version >= 2.0.0
stage_3.enable_model_cpu_offload()
prompt = 'a photo of a kangaroo wearing an orange hoodie and blue sunglasses standing in front of the eiffel tower holding a sign that says "very deep learning"'
# text embeds
prompt_embeds, negative_embeds = stage_1.encode_prompt(prompt)
generator = torch.manual_seed(0)
# stage 1
image = stage_1(prompt_embeds=prompt_embeds, negative_prompt_embeds=negative_embeds, generator=generator, output_type="pt").images
pt_to_pil(image)[0].save("./if_stage_I.png")
# stage 2
image = stage_2(
image=image, prompt_embeds=prompt_embeds, negative_prompt_embeds=negative_embeds, generator=generator, output_type="pt"
).images
pt_to_pil(image)[0].save("./if_stage_II.png")
# stage 3
image = stage_3(prompt=prompt, image=image, generator=generator, noise_level=100).images
image[0].save("./if_stage_III.png")

<clipboard-copy aria-label="Copy" class="ClipboardButton btn btn-invisible js-clipboard-copy m-2 p-0 tooltipped-no-delay d-flex flex-justify-center flex-items-center" data-copy-feedback="Copied!" data-tooltip-direction="w" value="from diffusers import DiffusionPipeline from diffusers.utils import pt_to_pil import torch

stage 1

stage_1 = DiffusionPipeline.from_pretrained("DeepFloyd/IF-I-XL-v1.0", variant="fp16", torch_dtype=torch.float16) stage_1.enable_xformers_memory_efficient_attention() # remove line if torch.version >= 2.0.0 stage_1.enable_model_cpu_offload()

stage 2

stage_2 = DiffusionPipeline.from_pretrained( "DeepFloyd/IF-II-L-v1.0", text_encoder=None, variant="fp16", torch_dtype=torch.float16 ) stage_2.enable_xformers_memory_efficient_attention() # remove line if torch.version >= 2.0.0 stage_2.enable_model_cpu_offload()

stage 3

safety_modules = {"feature_extractor": stage_1.feature_extractor, "safety_checker": stage_1.safety_checker, "watermarker": stage_1.watermarker} stage_3 = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-x4-upscaler", **safety_modules, torch_dtype=torch.float16) stage_3.enable_xformers_memory_efficient_attention() # remove line if torch.version >= 2.0.0 stage_3.enable_model_cpu_offload()

prompt = 'a photo of a kangaroo wearing an orange hoodie and blue sunglasses standing in front of the eiffel tower holding a sign that says "very deep learning"'

text embeds

prompt_embeds, negative_embeds = stage_1.encode_prompt(prompt)

generator = torch.manual_seed(0)

stage 1

image = stage_1(prompt_embeds=prompt_embeds, negative_prompt_embeds=negative_embeds, generator=generator, output_type="pt").images pt_to_pil(image)[0].save("./if_stage_I.png")

stage 2

image = stage_2( image=image, prompt_embeds=prompt_embeds, negative_prompt_embeds=negative_embeds, generator=generator, output_type="pt" ).images pt_to_pil(image)[0].save("./if_stage_II.png")

stage 3

image = stage_3(prompt=prompt, image=image, generator=generator, noise_level=100).images image[0].save("./if_stage_III.png")" tabindex="0" role="button">

有多种方法可以加快推理时间并进一步降低内存消耗diffusers。为此，请查看 Diffusers 文档：

🚀优化推理时间
⚙️推理期间优化低内存

有关如何使用 IF 的更多详细信息，请查看IF 博客文章和文档📖。

Diffusers dreambooth 脚本还支持微调 🎨 IF。通过参数高效微调，您可以使用单个 GPU 和 ~28 GB VRAM 向 IF 添加新概念。

在本地运行代码

将模型加载到 VRAM 中

from deepfloyd_if.modules import IFStageI, IFStageII, StableStageIII
from deepfloyd_if.modules.t5 import T5Embedder
device = 'cuda:0'
if_I = IFStageI('IF-I-XL-v1.0', device=device)
if_II = IFStageII('IF-II-L-v1.0', device=device)
if_III = StableStageIII('stable-diffusion-x4-upscaler', device=device)
t5 = T5Embedder(device="cpu")

<clipboard-copy aria-label="Copy" class="ClipboardButton btn btn-invisible js-clipboard-copy m-2 p-0 tooltipped-no-delay d-flex flex-justify-center flex-items-center" data-copy-feedback="Copied!" data-tooltip-direction="w" value="from deepfloyd_if.modules import IFStageI, IFStageII, StableStageIII from deepfloyd_if.modules.t5 import T5Embedder

device = 'cuda:0' if_I = IFStageI('IF-I-XL-v1.0', device=device) if_II = IFStageII('IF-II-L-v1.0', device=device) if_III = StableStageIII('stable-diffusion-x4-upscaler', device=device) t5 = T5Embedder(device="cpu")" tabindex="0" role="button">

我梦想

Dream 是 IF 模型的文本转图像模式

from deepfloyd_if.pipelines import dream
prompt = 'ultra close-up color photo portrait of rainbow owl with deer horns in the woods'
count = 4
result = dream(
t5=t5, if_I=if_I, if_II=if_II, if_III=if_III,
prompt=[prompt]*count,
seed=42,
if_I_kwargs={
"guidance_scale": 7.0,
"sample_timestep_respacing": "smart100",
},
if_II_kwargs={
"guidance_scale": 4.0,
"sample_timestep_respacing": "smart50",
},
if_III_kwargs={
"guidance_scale": 9.0,
"noise_level": 20,
"sample_timestep_respacing": "75",
},
)
if_III.show(result['III'], size=14)

<clipboard-copy aria-label="Copy" class="ClipboardButton btn btn-invisible js-clipboard-copy m-2 p-0 tooltipped-no-delay d-flex flex-justify-center flex-items-center" data-copy-feedback="Copied!" data-tooltip-direction="w" value="from deepfloyd_if.pipelines import dream

prompt = 'ultra close-up color photo portrait of rainbow owl with deer horns in the woods' count = 4

result = dream( t5=t5, if_I=if_I, if_II=if_II, if_III=if_III, prompt=[prompt]*count, seed=42, if_I_kwargs={ "guidance_scale": 7.0, "sample_timestep_respacing": "smart100", }, if_II_kwargs={ "guidance_scale": 4.0, "sample_timestep_respacing": "smart50", }, if_III_kwargs={ "guidance_scale": 9.0, "noise_level": 20, "sample_timestep_respacing": "75", }, )

if_III.show(result['III'], size=14)" tabindex="0" role="button">

II. 零样本图像到图像翻译

在风格转换模式下，你的提示输出的风格与support_pil_img

from deepfloyd_if.pipelines import style_transfer
result = style_transfer(
t5=t5, if_I=if_I, if_II=if_II,
support_pil_img=raw_pil_image,
style_prompt=[
'in style of professional origami',
'in style of oil art, Tate modern',
'in style of plastic building bricks',
'in style of classic anime from 1990',
],
seed=42,
if_I_kwargs={
"guidance_scale": 10.0,
"sample_timestep_respacing": "10,10,10,10,10,10,10,10,0,0",
'support_noise_less_qsample_steps': 5,
},
if_II_kwargs={
"guidance_scale": 4.0,
"sample_timestep_respacing": 'smart50',
"support_noise_less_qsample_steps": 5,
},
)
if_I.show(result['II'], 1, 20)

<clipboard-copy aria-label="Copy" class="ClipboardButton btn btn-invisible js-clipboard-copy m-2 p-0 tooltipped-no-delay d-flex flex-justify-center flex-items-center" data-copy-feedback="Copied!" data-tooltip-direction="w" value="from deepfloyd_if.pipelines import style_transfer

result = style_transfer( t5=t5, if_I=if_I, if_II=if_II, support_pil_img=raw_pil_image, style_prompt=[ 'in style of professional origami', 'in style of oil art, Tate modern', 'in style of plastic building bricks', 'in style of classic anime from 1990', ], seed=42, if_I_kwargs={ "guidance_scale": 10.0, "sample_timestep_respacing": "10,10,10,10,10,10,10,10,0,0", 'support_noise_less_qsample_steps': 5, }, if_II_kwargs={ "guidance_scale": 4.0, "sample_timestep_respacing": 'smart50', "support_noise_less_qsample_steps": 5, }, ) if_I.show(result['II'], 1, 20)" tabindex="0" role="button">

    <span data-target="animated-image.imageContainer">
        <img data-target="animated-image.replacedImage" alt="替代文本" class="AnimatedImagePlayer-animatedImage" src="https://github.com/deep-floyd/IF/raw/develop/pics/deep_floyd_if_image_2_image.gif" style="display: block; opacity: 1;">
      <canvas class="AnimatedImagePlayer-stillImage" aria-hidden="true" width="814" height="501"></canvas></span></a>
    <button data-target="animated-image.imageButton" class="AnimatedImagePlayer-images" tabindex="-1" aria-label="播放替代文本" hidden=""></button>
    <span class="AnimatedImagePlayer-controls" data-target="animated-image.controls" hidden="">
      <button data-target="animated-image.playButton" class="AnimatedImagePlayer-button" aria-label="播放替代文本">
        <svg aria-hidden="true" focusable="false" class="octicon icon-play" width="16" height="16" viewBox="0 0 16 16" fill="none" xmlns="http://www.w3.org/2000/svg">
          <path d="M4 13.5427V2.45734C4 1.82607 4.69692 1.4435 5.2295 1.78241L13.9394 7.32507C14.4334 7.63943 14.4334 8.36057 13.9394 8.67493L5.2295 14.2176C4.69692 14.5565 4 14.1739 4 13.5427Z">
        </path></svg>
        <svg aria-hidden="true" focusable="false" class="octicon icon-pause" width="16" height="16" viewBox="0 0 16 16" xmlns="http://www.w3.org/2000/svg">
          <rect x="4" y="2" width="3" height="12" rx="1"></rect>
          <rect x="9" y="2" width="3" height="12" rx="1"></rect>
        </svg>
      </button>
      <a data-target="animated-image.openButton" aria-label="在新窗口中打开替代文本" class="AnimatedImagePlayer-button" href="https://github.com/deep-floyd/IF/blob/develop/pics/deep_floyd_if_image_2_image.gif" target="_blank">
        <svg aria-hidden="true" class="octicon" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 16 16" width="16" height="16">
          <path fill-rule="evenodd" d="M10.604 1h4.146a.25.25 0 01.25.25v4.146a.25.25 0 01-.427.177L13.03 4.03 9.28 7.78a.75.75 0 01-1.06-1.06l3.75-3.75-1.543-1.543A.25.25 0 0110.604 1zM3.75 2A1.75 1.75 0 002 3.75v8.5c0 .966.784 1.75 1.75 1.75h8.5A1.75 1.75 0 0014 12.25v-3.5a.75.75 0 00-1.5 0v3.5a.25.25 0 01-.25.25h-8.5a.25.25 0 01-.25-.25v-8.5a.25.25 0 01.25-.25h3.5a.75.75 0 000-1.5h-3.5z"></path>
        </svg>
      </a>
    </span>
  </span></animated-image></p>

三、超分辨率

对于超分辨率，用户可以在不一定由 IF 生成的图像上运行IF-II和/或“稳定 x4”（两个级联）：IF-III

from deepfloyd_if.pipelines import super_resolution
middle_res = super_resolution(
t5,
if_III=if_II,
prompt=['woman with a blue headscarf and a blue sweaterp, detailed picture, 4k dslr, best quality'],
support_pil_img=raw_pil_image,
img_scale=4.,
img_size=64,
if_III_kwargs={
'sample_timestep_respacing': 'smart100',
'aug_level': 0.5,
'guidance_scale': 6.0,
},
)
high_res = super_resolution(
t5,
if_III=if_III,
prompt=[''],
support_pil_img=middle_res['III'][0],
img_scale=4.,
img_size=256,
if_III_kwargs={
"guidance_scale": 9.0,
"noise_level": 20,
"sample_timestep_respacing": "75",
},
)
show_superres(raw_pil_image, high_res['III'][0])

<clipboard-copy aria-label="Copy" class="ClipboardButton btn btn-invisible js-clipboard-copy m-2 p-0 tooltipped-no-delay d-flex flex-justify-center flex-items-center" data-copy-feedback="Copied!" data-tooltip-direction="w" value="from deepfloyd_if.pipelines import super_resolution

middle_res = super_resolution( t5, if_III=if_II, prompt=['woman with a blue headscarf and a blue sweaterp, detailed picture, 4k dslr, best quality'], support_pil_img=raw_pil_image, img_scale=4., img_size=64, if_III_kwargs={ 'sample_timestep_respacing': 'smart100', 'aug_level': 0.5, 'guidance_scale': 6.0, }, ) high_res = super_resolution( t5, if_III=if_III, prompt=[''], support_pil_img=middle_res['III'][0], img_scale=4., img_size=256, if_III_kwargs={ "guidance_scale": 9.0, "noise_level": 20, "sample_timestep_respacing": "75", }, ) show_superres(raw_pil_image, high_res['III'][0])" tabindex="0" role="button">

IV. 零样本修复

from deepfloyd_if.pipelines import inpainting
result = inpainting(
t5=t5, if_I=if_I,
if_II=if_II,
if_III=if_III,
support_pil_img=raw_pil_image,
inpainting_mask=inpainting_mask,
prompt=[
'oil art, a man in a hat',
],
seed=42,
if_I_kwargs={
"guidance_scale": 7.0,
"sample_timestep_respacing": "10,10,10,10,10,0,0,0,0,0",
'support_noise_less_qsample_steps': 0,
},
if_II_kwargs={
"guidance_scale": 4.0,
'aug_level': 0.0,
"sample_timestep_respacing": '100',
},
if_III_kwargs={
"guidance_scale": 9.0,
"noise_level": 20,
"sample_timestep_respacing": "75",
},
)
if_I.show(result['I'], 2, 3)
if_I.show(result['II'], 2, 6)
if_I.show(result['III'], 2, 14)

<clipboard-copy aria-label="Copy" class="ClipboardButton btn btn-invisible js-clipboard-copy m-2 p-0 tooltipped-no-delay d-flex flex-justify-center flex-items-center" data-copy-feedback="Copied!" data-tooltip-direction="w" value="from deepfloyd_if.pipelines import inpainting

result = inpainting( t5=t5, if_I=if_I, if_II=if_II, if_III=if_III, support_pil_img=raw_pil_image, inpainting_mask=inpainting_mask, prompt=[ 'oil art, a man in a hat', ], seed=42, if_I_kwargs={ "guidance_scale": 7.0, "sample_timestep_respacing": "10,10,10,10,10,0,0,0,0,0", 'support_noise_less_qsample_steps': 0, }, if_II_kwargs={ "guidance_scale": 4.0, 'aug_level': 0.0, "sample_timestep_respacing": '100', }, if_III_kwargs={ "guidance_scale": 9.0, "noise_level": 20, "sample_timestep_respacing": "75", }, ) if_I.show(result['I'], 2, 3) if_I.show(result['II'], 2, 6) if_I.show(result['III'], 2, 14)" tabindex="0" role="button">

    <span data-target="animated-image.imageContainer">
        <img data-target="animated-image.replacedImage" alt="deep_floyd_if_inpainting.gif" class="AnimatedImagePlayer-animatedImage" src="https://github.com/deep-floyd/IF/raw/develop/pics/deep_floyd_if_inpainting.gif" style="display: block; opacity: 1;">
      <canvas class="AnimatedImagePlayer-stillImage" aria-hidden="true" width="814" height="814"></canvas></span></a>
    <button data-target="animated-image.imageButton" class="AnimatedImagePlayer-images" tabindex="-1" aria-label="播放 deep_floyd_if_inpainting.gif" hidden=""></button>
    <span class="AnimatedImagePlayer-controls" data-target="animated-image.controls" hidden="">
      <button data-target="animated-image.playButton" class="AnimatedImagePlayer-button" aria-label="播放 deep_floyd_if_inpainting.gif">
        <svg aria-hidden="true" focusable="false" class="octicon icon-play" width="16" height="16" viewBox="0 0 16 16" fill="none" xmlns="http://www.w3.org/2000/svg">
          <path d="M4 13.5427V2.45734C4 1.82607 4.69692 1.4435 5.2295 1.78241L13.9394 7.32507C14.4334 7.63943 14.4334 8.36057 13.9394 8.67493L5.2295 14.2176C4.69692 14.5565 4 14.1739 4 13.5427Z">
        </path></svg>
        <svg aria-hidden="true" focusable="false" class="octicon icon-pause" width="16" height="16" viewBox="0 0 16 16" xmlns="http://www.w3.org/2000/svg">
          <rect x="4" y="2" width="3" height="12" rx="1"></rect>
          <rect x="9" y="2" width="3" height="12" rx="1"></rect>
        </svg>
      </button>
      <a data-target="animated-image.openButton" aria-label="在新窗口中打开 deep_floyd_if_inpainting.gif" class="AnimatedImagePlayer-button" href="https://github.com/deep-floyd/IF/blob/develop/pics/deep_floyd_if_inpainting.gif" target="_blank">
        <svg aria-hidden="true" class="octicon" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 16 16" width="16" height="16">
          <path fill-rule="evenodd" d="M10.604 1h4.146a.25.25 0 01.25.25v4.146a.25.25 0 01-.427.177L13.03 4.03 9.28 7.78a.75.75 0 01-1.06-1.06l3.75-3.75-1.543-1.543A.25.25 0 0110.604 1zM3.75 2A1.75 1.75 0 002 3.75v8.5c0 .966.784 1.75 1.75 1.75h8.5A1.75 1.75 0 0014 12.25v-3.5a.75.75 0 00-1.5 0v3.5a.25.25 0 01-.25.25h-8.5a.25.25 0 01-.25-.25v-8.5a.25.25 0 01.25-.25h3.5a.75.75 0 000-1.5h-3.5z"></path>
        </svg>
      </a>
    </span>
  </span></animated-image></p>

8000

🤗 模范动物园 🤗

下载权重以及模型卡的链接将很快在模型动物园的每个模型上提供

原来的

姓名	级联	参数	火焰离子化	批次大小	脚步
如果我是	我	4亿	8.86	3072	2.5米
中频	我	900米	8.06	3200	3.0米
如果-我-XL *	我	4.3B	6.66	3072	2.42万
IF-II-M	二	450米	-	1536	2.5米
如果-II-L *	二	1.2B	-	1536	2.5米
IF-III-L* （即将推出）	三	700米	-	3072	1.25米

*最佳模块

定量评估

FID = 6.66

执照

此存储库中的代码是根据定制许可证发布的（参见添加的第二点）。

这些重量将很快通过Hugging Face 的 DeepFloyd 组织提供，并拥有自己的许可证。

免责声明： IF 模型的初始版本暂时受到限制性研究目的许可，以收集反馈，之后我们打算发布与其他稳定性 AI 模型一致的完全开源模型。

局限性和偏见

此代码库中提供的模型具有已知的局限性和偏差。请参阅模型卡了解更多信息。

🎓 DeepFloyd IF 创作者：

Alex Shonenkov GitHub | Linktr
Misha Konstantinov GitHub | Twitter
Daria Bakshandaeva GitHub | Twitter
Christoph Schuhmann GitHub |推特
Ksenia Ivanova GitHub | Twitter
Nadiia Klokova GitHub |推特

📄 研究论文（即将发布）

致谢

特别感谢StabilityAI及其首席执行官Emad Mostaque的宝贵支持，提供 GPU 计算和基础设施来训练模型（我们感谢Richard Vencu）；特别感谢LAION和Christoph Schuhmann对项目的贡献和精心准备的数据集；感谢Huggingface团队在推理过程中优化模型的速度和内存消耗，创建演示并提供很棒的建议！

🚀 外部贡献者 🚀

最诚挚的感谢@Apolinário，感谢他们在各个阶段提供的想法、咨询、帮助和支持，使 IF 可以在开源中使用；感谢他们编写了大量文档和说明；感谢他们在困难时刻营造了友好的氛围🦉；
感谢@patrickvonplaten，将 unet 模型的加载时间提高了 80%；将 Stable-Diffusion-x4 集成为原生管道💪；
感谢@williamberman和@patrickvonplaten的扩散器集成🙌；
感谢@hysts和@Apolinário使用 IF 🚀创建了最佳 gradio 演示；
感谢@Dango233使用 xformers 内存高效注意力机制调整 IF 💪；

Name		Name	Last commit message	Last commit date
Latest commit History 52 Commits
deepfloyd_if		deepfloyd_if
pics		pics
.gitattributes		.gitattributes
.gitignore		.gitignore
.pre-commit-config.yaml		.pre-commit-config.yaml
CHANGELOG.md		CHANGELOG.md
LICENSE		LICENSE
LICENSE-MODEL		LICENSE-MODEL
README.md		README.md
requirements-dev.txt		requirements-dev.txt
requirements-test.txt		requirements-test.txt
requirements.txt		requirements.txt
setup.cfg		setup.cfg
setup.py		setup.py

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Repository files navigation

StabilityAI的DeepFloyd Lab的 IF

使用所有 IF 模型的最低要求：

快速开始

本地笔记本

与🤗扩散器集成

例子

stage 1

stage 2

stage 3

text embeds

stage 1

stage 2

stage 3

在本地运行代码

将模型加载到 VRAM 中

我梦想

II. 零样本图像到图像翻译

三、超分辨率

IV. 零样本修复

🤗 模范动物园 🤗

原来的

定量评估

执照

局限性和偏见

🎓 DeepFloyd IF 创作者：

📄 研究论文（即将发布）

致谢

🚀 外部贡献者 🚀

About

Uh oh!

Releases

Packages

Languages

License

yuanxiaoming8899/IF

Folders and files

Latest commit

History

Repository files navigation

StabilityAI的DeepFloyd Lab的 IF

使用所有 IF 模型的最低要求：

快速开始

本地笔记本

与🤗扩散器集成

例子

stage 1

stage 2

stage 3

text embeds

stage 1

stage 2

stage 3

在本地运行代码

将模型加载到 VRAM 中

我梦想

II. 零样本图像到图像翻译

三、超分辨率

IV. 零样本修复

🤗 模范动物园 🤗

原来的

定量评估

执照

局限性和偏见

🎓 DeepFloyd IF 创作者：

📄 研究论文（即将发布）

致谢

🚀 外部贡献者 🚀

About

Resources

License

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Languages

Packages