roytalman/Deep_contrastive_embedding

Deep supervised conistrastive learning for small datasets (few shot learning). This repository takes labeled embedding data ,that could be extracted from pre-trained NLP, vision, or any other algorithm that extract embedding, and use deep FFN to learn new embedding that is fine-tuned for the current data. Th algorithm can improve classification per

13
/ 100
Experimental

This tool helps data scientists and machine learning engineers improve the accuracy of classification models, especially when working with very small datasets. You provide existing labeled data embeddings (from sources like NLP or computer vision models), and it outputs refined embeddings that are better optimized for your specific classification task. This is ideal for fine-tuning models on limited data.

No commits in the last 6 months.

Use this if you need to boost the performance of a classification model when you only have a small number of labeled examples for your categories.

Not ideal if you are looking for a tool to extract initial embeddings from raw data (text, images, etc.) or if you have a very large, well-labeled dataset.

Machine Learning Data Science Few-shot Learning Classification Model Fine-tuning
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 8 / 25
Community 0 / 25

How are scores calculated?

Stars

10

Forks

Language

Jupyter Notebook

License

Last pushed

Aug 24, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/embeddings/roytalman/Deep_contrastive_embedding"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.