GithubHelp home page GithubHelp logo

marianqqq / neuro Goto Github PK

View Code? Open in Web Editor NEW

This project forked from shedou/neuro

0.0 0.0 0.0 44.61 MB

Useful assemblies of neural network software.

License: GNU Affero General Public License v3.0

Shell 37.24% Python 39.37% Batchfile 23.39%

neuro's Introduction

Внимание! Attention!

Пожалуйста, воздержитесь от использования любых файлов и информации из репозитория если вы не готовы взять ответственность за любые возможные и невозможные последствия!

В коде могут присутствовать ошибки и проблемы, которые могут привести к разнообразным последствиям, настоятельно рекомендую воздержаться от использования сборок в ответственных системах!

English translation

Please refrain from using any files and information from the repository unless you are ready to take responsibility for any possible and impossible consequences!

The code may contain errors and problems that can lead to various consequences, I strongly recommend that you refrain from using assemblies in critical systems!

Сборки Neuro Github Releases

Для получения подробной информации перейдите в раздел конкретной сборки:

English translation

For details, go to the specific build section:

- NVIDIA/CPU - CUDA - Linux Chimbalix 23.1 / 24.1 Alphachi

  • Linux вариант сборки Stable Diffusion WebUI.
English translation
  • Linux version of the Stable Diffusion WebUI build.

- NVIDIA/CPU - CUDA - Windows 7/10/11

  • Не умеете рисовать, но очень хочется? Позвольте это сделать приложению Stable Diffusion! Нейросеть нарисует всё что пожелаете, конечно же в пределах возможностей используемой модели... Для генерации изображения достаточно написать текст в поле ввода и нажать кнопку "Generate", но есть один нюанс, нейросеть понимает только слова на английском языке, впрочем, никто не запрещает использовать переводчик отдельно если с английским языком всё очень плохо...
English translation
  • Can't draw but really want to? Let Stable Diffusion do it! The neural network will draw whatever you want, of course, within the capabilities of the model used... To generate an image, just write text in the input field and click the "Generate" button, but there is one caveat, the neural network understands only words in English, however, no one forbids use a translator separately if everything is very bad with the English language...

- AMD/NVIDIA/CPU - DirectML - Windows 10/11

  • DirectML вариант приложения Stable Diffusion, подходит для работы с современными видеокартами AMD, хотя работает медленнее варианта для NVIDIA, но это лучше, чем ничего.
English translation
  • DirectML version of the Stable Diffusion application, suitable for modern AMD graphics cards, although it is slower than the NVIDIA version, but it's better than nothing.

- AMD/NVIDIA/CPU - DirectML - Windows 10/11

  • Предварительно скомпилированная, портативная сборка приложения roop, данная нейронная сеть позволяет заменять лица на видеозаписях таким образом создавая дипфейки (Deep Fakes), практической пользы я не вижу от данной нейронной сети, но повеселиться в кругу друзей вполне возможно.

  • Оригинальное описание:
    Take a video and replace the face in it with a face of your choice. You only need one image of the desired face. No dataset, no training.

English translation
  • A pre-compiled, portable build of the roop application, this neural network allows you to replace faces in videos, thus creating Deep Fakes, I don’t see any practical benefit from this neural network, but it’s quite possible to have fun with friends.

  • Original description:
    Take a video and replace the face in it with a face of your choice. You only need one image of the desired face. No dataset, no training.

Old Projects

Waifu2x and Real ESRGAN NCNN

- GPU/CPU - OpenCL API - Windows 7/10/11

  • Waifu2X поможет в обработке изображений, основное назначение это увеличение изображений и подавление шума разной степени при помощи нейронных сетей, наиболее полезно при чистке изображений от артефактов сжатия JPEG.
English translation
  • Waifu2X will help in image processing, the main purpose is to upscale images and suppress noise of varying strength using neural networks, it is most useful when cleaning images from JPEG compression artifacts.

- GPU - Vulkan API - Windows 7/10/11

  • Real ESRGAN по своей сути является свободным аналогом коммерческого тяжеловесного приложения Topaz Gigapixel AI, основное назначение это увеличение изображений при помощи нейронных сетей, а благодаря использованию Vulkan API производительность всегда на высшем уровне насколько позволяет видеокарта.
English translation
  • Real ESRGAN is essentially a free analogue of the commercial heavyweight Topaz Gigapixel AI application, the main purpose is to upscale images using neural networks, and thanks to the use of the Vulkan API, performance is always at the highest level as far as the graphics card allows.

Other licenses

Any third party are subject to their own licenses.

neuro's People

Contributors

shedou avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.