https://github.com/vergissberlin/thinkport-ai

Training data to fine tune a custom model for Thinkport GmbH
https://github.com/vergissberlin/thinkport-ai

model openai

Last synced: 6 days ago
JSON representation

Training data to fine tune a custom model for Thinkport GmbH

• Host: GitHub
• URL: https://github.com/vergissberlin/thinkport-ai
• Owner: vergissberlin
• Created: 2023-11-20T22:19:13.000Z (12 months ago)
• Default Branch: main
• Last Pushed: 2023-12-15T02:18:45.000Z (11 months ago)
• Last Synced: 2024-10-10T19:41:58.842Z (27 days ago)
• Topics: model, openai
• Language: Python
• Homepage:
• Size: 28.3 KB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Fine tuning for Thinkport GmbH

This repository holds training data to finetune and OpenAI model. It also contains a script to convert the training data to the format that the OpenAI model expects.

## Install

To install the dependencies, run:

    ```bash

    pip install -r requirements.txt

    ```

## Usage

### Validation

It is importend to validate your data to prevent wasting money on training.

    ```bash

    phyton3 validate.py

    ```

### Training

When you are done with validating your data, you can start training.

    ```bash

    python3 finetune.py

    ```

Now visit the OpenAI website to see your training progress. It will take a couple of hours to train. And it wil cost you money. To test the model, visit the [playground](https://platform.openai.com/playground) and select the new model.