https://github.com/ezroot/gacc
GIACC - Generate Images, Art, Code and Conversations
https://github.com/ezroot/gacc
ai codegen cuda huggingface image imagegeneration python rust stablediffusion
Last synced: 2 months ago
JSON representation
GIACC - Generate Images, Art, Code and Conversations
- Host: GitHub
- URL: https://github.com/ezroot/gacc
- Owner: EZroot
- Created: 2023-05-27T21:40:27.000Z (about 3 years ago)
- Default Branch: master
- Last Pushed: 2023-07-04T02:44:09.000Z (almost 3 years ago)
- Last Synced: 2025-03-11T10:24:54.539Z (over 1 year ago)
- Topics: ai, codegen, cuda, huggingface, image, imagegeneration, python, rust, stablediffusion
- Language: Rust
- Homepage:
- Size: 17.1 MB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# GACC - Generate Art, Code, and Conversations
A collection of:
* Backend AI Server - Python AI backend server
* Windows App - 2D Image generator and editor
* Command Line Interface - Copilot Code generation
Below you will find the current modules I'm working on for this AI package.
## Picasso - 2D Image Generation AI
An image editing app with access to AI generation.
* Generate completely new art based on the latest AI models
* Draw images and quickly turn them professional with AI generation tools
* Regenerate, stylize, remove or add new objects to your images
## Copilot - Code Generation AI
* Automatically generate syntax passing files
* usage ```./copilot -pycode generate a file to do something```
> Supports python only for now.
## Converse - Text Generation AI
* TODO
requires mypy and some other setup.
use a python virtual environment or conda to deal with systemwide pips from windows interefereing from running your ai applications
python -m will use a particular pythons pip, so you can download linux one and isolate it to linux pythons pip
pre cuda installation
https://docs.nvidia.com/cuda/archive/11.8.0/cuda-installation-guide-linux/index.html#pre-installation-actions
cuda installation
https://docs.nvidia.com/cuda/archive/11.8.0/cuda-installation-guide-linux/index.html#wsl-installation
https://developer.nvidia.com/cuda-11-8-0-download-archive?target_os=Linux&target_arch=x86_64&Distribution=WSL-Ubuntu&target_version=2.0&target_type=deb_local
post cuda installation
https://docs.nvidia.com/cuda/archive/11.8.0/cuda-installation-guide-linux/index.html#post-installation-actions
Dont forget to create a config.json in the target/debug directory where the build is
```json
{
"api_key": "token",
"stable_diffusion_host_address": "localhost:6999",
"python_code_create_iterations": 5,
"python_use_conda_over_pip": true,
"stable_diffusion_config": {
"height": 512,
"width": 712,
"num_inference_steps": 800,
"guidance_scale": 7.5,
"img_count": 25,
"use_columns": true
}
}
```