Features | Dependencies | SystemRequirements | Install | Usage | Models | Wiki | Acknowledgment | Licenses
- Work in progress! (ALPHA)
- English | Русский
A simple and convenient interface for using of various neural network models. You can create datasets, finetune, evaluate and generate with LLM and StableDiffusion, using various hyperparameters. You can also download the LLM and StableDiffusion models, change the application settings inside the interface and check system sensors
The goal of the project - to create the easiest possible application to finetune, evaluate and generate of neural network models
- Easy installation via install.bat(Windows) or install.sh(Linux)
- Flexible and optimized interface (By Gradio)
- Authentication via admin:admin (You can enter your login details in the GradioAuth.txt file)
- Support for Transformers: finetune, evaluate, quantize and generate (LLM)
- Support for Diffusers and Safetensors: finetune, evaluate, conversion and generate (StableDiffusion)
- Full and LORA types of finetune, evaluate and generate (For LLM and StableDiffusion)
- Ability to create a dataset (For LLM and StableDiffusion)
- ModelDownloader (For LLM and StableDiffusion)
- Application settings
- Ability to see system sensors
- C+ compiler
- Windows: VisualStudio
- Linux: GCC
- System: Windows or Linux
- GPU: 8GB+ or CPU: 16 core 3.6Ghz
- RAM: 24GB+
- Disk space: 10GB+
- Internet for installing
git clone https://github.com/Dartvauder/NeuroTrainerWebUI.git
to any location- Run the
Install.bat
and wait for installation - After installation, run
Start.bat
- Select the file version and wait for the application to launch
- Now you can start experiment with your models!
To get update, run Update.bat
To work with the virtual environment through the terminal, run Venv.bat
git clone https://github.com/Dartvauder/NeuroTrainerWebUI.git
to any location- In the terminal, run the
./Install.sh
and wait for installation of all dependencies - After installation, run
./Start.sh
- Wait for the application to launch
- Now you can start experiment with your models!
To get update, run ./Update.sh
To work with the virtual environment through the terminal, run ./Venv.sh
Interface has five tabs: LLM, StableDiffusion, ModelDownloader, Settings and System. Select the one you need and follow the instructions below
- Here you can create a new or expand an existing dataset
- Datasets are saved in a folder datasets/llm
- First upload your models to the folder: models/llm
- Upload your dataset to the folder: datasets/llm
- Select your model and dataset from the drop-down lists
- Select a finetune method
- Write a name for the model
- Set up the model hyper-parameters for finetuning
- Click the
Submit
button to receive the finetuned model
- First upload your models to the folder: finetuned-models/llm
- Upload your dataset to the folder: datasets/llm
- Select your models and dataset from the drop-down lists
- Set up the models parameters for evaluate
- Click the
Submit
button to receive the evaluate of model
- First upload your models to the folder: finetuned-models/llm
- Select a Model and Quantization Type
- Click the
Submit
button to receive the conversion of model
- Select your models from the drop-down list
- Set up the models according to the parameters you need
- Set up the models parameters to generate
- Click the
Submit
button to receive the generated text
- Here you can create a new or expand an existing dataset
- Datasets are saved in a folder datasets/sd
- First upload your models to the folder: models/sd
- Upload your dataset to the folder: datasets/sd
- Select your model and dataset from the drop-down lists
- Select a model type and finetune method
- Write a name for the model
- Set up the model hyper-parameters for finetuning
- Click the
Submit
button to receive the finetuned model
- First upload your models to the folder: finetuned-models/sd
- Upload your dataset to the folder: datasets/sd
- Select your models and dataset from the drop-down lists
- Select a model method and model type
- Enter your prompt
- Set up the models parameters for evaluate
- Click the
Submit
button to receive the evaluate of model
- First upload your models to the folder: finetuned-models/sd
- Select a model type
- Set up the models parameters for convert
- Click the
Submit
button to receive the conversion of model
- First upload your models to the folder: finetuned-models/sd
- Select your models from the drop-down list
- Select a model method and model type
- Enter your prompt
- Set up the models parameters to generate
- Click the
Submit
button to receive the generated image
- Here you can download
LLM
andStableDiffusion
models. Just choose the model from the drop-down list and click theSubmit
button
- Here you can change the application settings. For now you can only change
Share
mode toTrue
orFalse
- Here you can see the indicators of your computer's sensors by clicking on the
Submit
button
- All finetunes are saved in the finetuned-models folder
- You can press the
Clear
button to reset your selection - You can turn off the application using the
Close terminal
button - You can open the finetuned-models, datasets, and outputs folders by clicking on the folder name button
- LLM and StableDiffusion models can be taken from HuggingFace or from ModelDownloader inside interface
- LLM and StableDiffusion datasets can be taken from HuggingFace or you can create own datasets inside interface
Many thanks to these projects because thanks to their applications/libraries, i was able to create my application:
First of all, I want to thank the developers of PyCharm and GitHub. With the help of their applications, i was able to create and share my code
gradio
- https://github.com/gradio-app/gradiotransformers
- https://github.com/huggingface/transformersdiffusers
- https://github.com/huggingface/diffusers