parasj/contracode
Contrastive Code Representation Learning: functionality-based JavaScript embeddings through self-supervised learning
This project helps machine learning engineers and researchers build more effective AI tools for code. It provides a way to train models that understand the core functionality of JavaScript programs, regardless of how they are written. By taking raw JavaScript code as input, it produces a learned representation that improves the accuracy of downstream tasks like code summarization and type prediction.
169 stars. No commits in the last 6 months.
Use this if you are developing AI-powered programming tools and need a robust, self-supervised method to learn deep representations of JavaScript code.
Not ideal if you are a JavaScript developer looking for a direct programming utility, as this is a research framework for building code AI, not an end-user tool.
Stars
169
Forks
27
Language
Jupyter Notebook
License
Apache-2.0
Category
Last pushed
Dec 26, 2021
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/parasj/contracode"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
AdaptiveMotorControlLab/CEBRA
Learnable latent embeddings for joint behavioral and neural analysis - Official implementation of CEBRA
theolepage/sslsv
Toolkit for training and evaluating Self-Supervised Learning (SSL) frameworks for Speaker...
PaddlePaddle/PASSL
PASSL包含 SimCLR，MoCo v1/v2，BYOL，CLIP，PixPro，simsiam, SwAV, BEiT，MAE 等图像自监督算法以及 Vision...
YGZWQZD/LAMDA-SSL
30 Semi-Supervised Learning Algorithms
ModSSC/ModSSC
ModSSC: A Modular Framework for Semi Supervised Classification