Ssebowa：用于文本、图像和视频创作的开源生成式AI库

引言

Ssebowa 是一个开源的 Python 库，提供多种生成式 AI（Generative AI）模型，包括：

• ssebowa-llm：用于文本生成的大型语言模型（LLM）
• ssebowa-vllm：用于视觉理解的视觉语言模型（VLLM）
• ssebowa-imagen：图像生成与定制微调（Fine-Tuning）模型
• Ssebowa-vigen：视频生成模型

With Ssebowa, you can easily generate text, translate languages, write different kinds of creative content, personalized image generation and answer your questions in an informative way.

For more detailed usage information, please refer to: Ssebowa's technical documentation

---

安装

Before running the script, ensure that the required libraries are installed. You can do this by executing the following commands:

git clone https://github.com/huggingface/diffusers
cd diffusers
pip install .

Then install Ssebowa:

pip install ssebowa

If you are running these commands in Colab or Jupyter Notebook, please use the following:

!git clone https://github.com/huggingface/diffusers
!cd diffusers
!pip install .

!pip install ssebowa

Now, you can access the different models by importing them from the library:

---

Ssebowa 图像生成

Ssebowa-Imagen is an open-source image synthesis model that utilizes a combination of diffusion modeling and generative adversarial networks (GANs) to generate high-quality images from text descriptions and allows also to turn your few photos into custom model that is capable of generating stunning images of your chosen subject. It leverages a 100 billion dataset of images and text descriptions, enabling it to accurately capture the nuances of real-world imagery and effectively translate text descriptions into compelling visual representations.

在自有数据上进行微调

• Prepare about 10-20 high-quality solo photos (jpg or png) like yours, friend, product or pets etc and put them in a specific directory.
• Please run on a machine with a GPU of 16GB or more. (If you're fine-tuning SDXL, you'll need 24GB of VRAM.)

from ssebowa.dataset import LocalDataset
from ssebowa.model import SdSsebowaModel
from ssebowa.trainer import LocalTrainer
from ssebowa.utils.image_helpers import display_images
from ssebowa.utils.prompt_helpers import make_prompt

DATA_DIR = "data"  # The directory where you put your prepared photos
OUTPUT_DIR = "models"

dataset = LocalDataset(DATA_DIR)
dataset = dataset.preprocess_images(detect_face=True)

SUBJECT_NAME = "<YOUR-NAME>"
CLASS_NAME = "person"

model = SdSsebowaModel(subject_name=SUBJECT_NAME, class_name=CLASS_NAME)
trainer = LocalTrainer(output_dir=OUTPUT_DIR)
predictor = trainer.fit(model, dataset)

# Use the prompt helper to create an awesome AI avatar!
prompt = next(make_prompt(SUBJECT_NAME, CLASS_NAME))
images = predictor.predict(
    prompt, height=768, width=512, num_images_per_prompt=2,
)

display_images(images, fig_size=10)

基础图像生成

from ssebowa import Ssebowa_imgen
model = Ssebowa_imgen()

Generate an image with the text description. For example, let's generate "A cat sitting on a bookshelf":

image = model.generate_image("A cat sitting on a bookshelf")

Save the image to a file:

image.save("cat_on_bookshelf.jpg")

---

Ssebowa 视觉语言模型

Ssebowa-vllm is an open-source visual large language model (VLLM) developed by Ssebowa AI. It is a powerful tool that can be used to understand images. Ssebowa-vllm has 11 billion visual parameters and 7 billion language parameters, supporting image understanding at a resolution of 1120*1120.

from ssebowa import ssebowa_vllm
model = ssebowa_vllm()

response = model.understand(image_path, prompt)
print(response)

---

模型对比

---

贡献

Ssebowa is open to contributions! Guidelines in progress...

许可证

Ssebowa is released under Apache License 2.0.

联系方式

If you have any questions or suggestions, please feel free to open an issue on GitHub or contact us at support@ssebowa.ai

常见问题（FAQ）

Ssebowa 需要什么硬件配置？

Ssebowa 需要配备至少 16GB 显存的 GPU 才能运行。如果微调 SDXL 模型，则需要 24GB 显存。

如何用 Ssebowa 微调自己的图像模型？

准备 10-20 张高质量单人照片，放入指定目录。使用 LocalDataset 加载并预处理，然后通过 LocalTrainer 微调 SdSsebowaModel 即可生成定制图像。

Ssebowa 支持哪些生成任务？

Ssebowa 支持文本生成（ssebowa-llm）、视觉理解（ssebowa-vllm）、图像生成与微调（ssebowa-imagen）以及视频生成（Ssebowa-vigen）。