https://github.com/alexdavid22/react---node.js-and-huggingface-llm-inference.-full-stack-project---frontend-part

Science Score: 13.0%

This score indicates how likely this project is to be science-related based on various indicators:

○
CITATION.cff file
✓
codemeta.json file
Found codemeta.json file
○
.zenodo.json file
○
DOI references
○
Academic publication links
○
Academic email domains
○
Institutional organization owner
○
JOSS paper metadata
○
Scientific vocabulary similarity
Low similarity (10.0%) to scientific vocabulary

Last synced: 10 months ago · JSON representation

Repository

Basic Info

Host: GitHub
Owner: alexdavid22
Language: JavaScript
Default Branch: master
Size: 188 KB

Statistics

Stars: 0
Watchers: 2
Forks: 0
Open Issues: 0
Releases: 0

Created almost 2 years ago · Last pushed almost 2 years ago

Metadata Files

Readme

Full Stack AI Project

This is a full stack application that interacts with an AI model to generate responses based on predefined templates as specified by the company requirements. The project is fully operational and can be tested online.

Live Demo

The project is available online and can be tested at the following link: Live Demo

To use the application, simply visit the link, enter your request in the form, and click send. The AI will respond according to the predefined template.

Project Overview

The application is in English, and the AI model responds in English due to limitations in the open-source models, which are not as powerful in languages other than English.
The AI's responses are always formatted according to the specified template from the company examples.

How to Run the Project

No dependencies need to be installed, as the project is complete and ready for use. Simply visit the live demo link provided above.

AI Model Training Steps

Training Details

The AI model was initially trained using Llama3. During training, it was observed that the model cannot communicate effectively in Romanian.
For the web application, a new model, Mistral, was used. This model was context fine-tuned, meaning it was provided memory context to know how to formulate the answer.

Resources

Initial Hugging Face Model: Llama 3 8B Instruct BNB 4Bit Veziv Test
Current Model: Mistral (context fine-tuned for inference API)
Dataset: Finetuning Demo
Google Drive: Training Video (Jupyter Notebook code for initial training through 10 epochs)

Project Structure

Front-end: Developed with the modern JavaScript framework React.js for responsive and dynamic user interaction.
Back-end: Node.js used for server-side operations and API management.
AI Model: Integrated with the application to process user inputs and generate appropriate responses based on the training data and predefined templates.

Owner

Login: alexdavid22
Kind: user

Repositories: 1
Profile: https://github.com/alexdavid22

GitHub Events

Total

Last Year

Dependencies

package-lock.json npm

1275 dependencies

package.json npm

@testing-library/jest-dom ^5.17.0
@testing-library/react ^13.4.0
@testing-library/user-event ^13.5.0
react ^18.3.1
react-dom ^18.3.1
react-scripts 5.0.1
web-vitals ^2.1.4

ecosyste.ms

Data

Tools

Indexes

Applications

Experiments

Open Source Science