llm_dataset_generation

Introduction

This GitHub repository provides tools for generating training data using large language models (LLMs) specifically for logistics applications. The main objective is to assist a logistics company in creating an interactive Q&A bot that can classify user queries related to logistics.

Key Features

• Data Generation: Using DeepSeek-V3, it generates diverse datasets relevant to logistics and non-logistics queries.
• Data Review: Implements DeepSeek-R1 to verify and check the generated datasets for quality and classification accuracy.
• Python-Based: The tools are implemented in Python, making them accessible and modifiable for developers.

Benefits

• Automation: Reduces the hassle of manual data collection and labeling by automating the data generation process.
• Quality Assurance: Ensures that the generated datasets are relevant and accurate through robust data review mechanisms.

Highlights

• Specifically designed for logistics-related query classification.
• Simplifies the data preparation phase for building AI applications in logistics.
• Open-source project encouraging contributions and community support.