https://github.com/jra333/automapper
A business specific use case leveraging a LLM (T5) to automate data mapping.
https://github.com/jra333/automapper
automation llms snowflake t5-model
Last synced: 3 months ago
JSON representation
A business specific use case leveraging a LLM (T5) to automate data mapping.
- Host: GitHub
- URL: https://github.com/jra333/automapper
- Owner: jra333
- Created: 2025-02-06T08:15:58.000Z (8 months ago)
- Default Branch: main
- Last Pushed: 2025-02-07T02:33:09.000Z (8 months ago)
- Last Synced: 2025-06-23T17:49:44.460Z (3 months ago)
- Topics: automation, llms, snowflake, t5-model
- Language: Jupyter Notebook
- Homepage:
- Size: 434 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# AUTOMAPPER
**Two folders include:**
- The actual app POC/DEMO
[Automapper](https://github.com/jra333/automapper/tree/main/automapper_app_demo) is a Streamlit-based web application designed to automate the processing, review, and archival of file submissions using AI-powered predictions. It integrates with Snowflake for data storage, staging, and user authentication, while leveraging a fine-tuned T5 model from Hugging Face Transformers for generating business-critical data mapping.
**SOME CODE AND PROCESSES HAVE BEEN REDACTED/ALTERED FOR PRIVACY AND WILL NOT BE UPDATED.**
***Actual finetuned model not included in repo.***
- T5 finetuning training & inference script [notebook](https://github.com/jra333/automapper/tree/main/automapper_notebooks_t5). The script is set up for a business specific project, it trains on custom dataset and processes outputs in applicable manner. The inference script weights rely on business context to reproduce.