https://github.com/torchstack-ai/bert-finetuning
This repository contains code that fine-tunes BERT for text classification on financial tweets.
https://github.com/torchstack-ai/bert-finetuning
bert bert-fine-tuning fine-tuning huggingface llm mlflow text-classification twitter
Last synced: 8 months ago
JSON representation
This repository contains code that fine-tunes BERT for text classification on financial tweets.
- Host: GitHub
- URL: https://github.com/torchstack-ai/bert-finetuning
- Owner: torchstack-ai
- License: mit
- Created: 2024-04-28T23:53:12.000Z (about 2 years ago)
- Default Branch: main
- Last Pushed: 2024-05-05T06:06:12.000Z (about 2 years ago)
- Last Synced: 2025-05-05T09:28:50.985Z (about 1 year ago)
- Topics: bert, bert-fine-tuning, fine-tuning, huggingface, llm, mlflow, text-classification, twitter
- Language: Jupyter Notebook
- Homepage: https://huggingface.co/torchstack/hf-bert-finetuning
- Size: 72.3 KB
- Stars: 0
- Watchers: 1
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
## Summary
This tutorial covers fine-tuning a BERT model to perform sentiment analysis from the Financial Tweet Sentiment Dataset.
We applied the following tools in this notebook:
* We used the HuggingFace library to perform data processing, fine-tune the BERT model, and evaluate the accuracy between the predicted class and reference.
* We benchmarked the naive BERT model and the fine-tuned BERT model, and we were able to boost performance by 17%.
* We used MLFlow for model training logging, which is a useful MLOps tool.
Follow the our blog, [The Stack](https://torchstack.ai/blogs/torchstack) for more hands-on tutorials, reviews and summaries of essential AI concepts, and what we're seeing go on in the AI space.
You can also check out our [GitHub](https://github.com/torchstack-ai) page for more projects and code samples.