How to get started with machine learning on source code?

Use this list to explore foundational papers and datasets in areas like code summarization, but since it's outdated, supplement with recent tutorials from active communities or platforms like ml4code.github.io for current trends.

What are the best datasets for training code models in 2024?

The datasets section includes historical resources like Public Git Archive, but for up-to-date options, check newer repositories like CodeSearchNet or GitHub's own datasets, as this list hasn't been maintained since 2019.

Machine learning for code vs traditional static analysis: which should I use?

ML methods excel at handling complex patterns and natural language in code, but traditional analysis is often faster and more interpretable; the choice depends on your project's needs for accuracy versus speed.

How to implement code summarization with neural networks?

Refer to papers in the code summarization section for architectures like convolutional attention networks, but be prepared to adapt code from linked software projects or newer frameworks, as implementations may be outdated.

Is this awesome list still useful for current research?

It provides a valuable historical overview and foundational references, but due to lack of updates, rely on it for background while seeking recent publications from conferences like ICSE or ASE for cutting-edge work.

Awesome Machine Learning On Source Code

CC-BY-SA-4.0

A curated list of research papers, datasets, and software projects for machine learning applied to source code (MLonCode).

GitHub

What is Awesome Machine Learning On Source Code?

Awesome Machine Learning On Source Code is a comprehensive, community-driven collection of resources dedicated to the intersection of machine learning and source code. It serves as a central hub for researchers and practitioners exploring how ML techniques can analyze, understand, and generate source code. The project aims to lower the barrier to entry for MLonCode research by providing a well-organized, centralized repository of knowledge and tools.

Target Audience

Researchers and practitioners in machine learning and software engineering who are exploring applications of ML to source code, such as program synthesis, code summarization, bug detection, and code search. It is particularly useful for academics, data scientists, and engineers working on MLonCode projects.

Value Proposition

Developers choose this project because it offers a curated, structured collection of research papers, datasets, tools, and conferences specifically focused on ML applied to source code, which is not as readily available in general ML resource lists. Its community-driven nature ensures it stays updated with the latest advancements in the niche field of MLonCode.

Overview

Cool links & research papers related to Machine Learning applied to source code (MLonCode)

Use Cases

Best For

Related Projects

Community-curated · Updated weekly · 100% open source

Found a gem we're missing?

Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.

Submit a project Star on GitHub

6.6k stars837 forks0 contributors

Finding academic research papers on machine learning techniques for source code analysis and generation.
Discovering open-source software and frameworks for implementing MLonCode projects, such as code embedding models or program synthesis tools.
Locating publicly available datasets for training and evaluating ML models on source code, like GitHub commit histories or code snippet collections.
Identifying relevant academic conferences, workshops, and talks focused on ML and software engineering intersections.
Exploring curated lists of tools for tasks like code clone detection, sentiment analysis in code reviews, or automated code summarization.
Getting started with MLonCode research by accessing a centralized, organized repository that lowers the barrier to entry.

Not Ideal For

Developers seeking actively maintained, up-to-date resources with recent publications and tools beyond 2019
Teams needing hands-on implementation guides or step-by-step tutorials for building MLonCode applications
Projects looking for pre-trained models or plug-and-play APIs for code analysis without extensive background research
Researchers focused exclusively on very recent advancements post-2020, as updates have ceased

Pros & Cons

Pros

Comprehensive Resource Coverage

Organizes a vast array of research papers, datasets, and tools across subfields like program synthesis and bug detection, as detailed in the structured contents list from the README.

Community-Driven Curation

Actively maintained with a structured contribution process, fostering a collaborative hub for MLonCode knowledge, as highlighted in the philosophy and contributions sections.

Lowered Entry Barrier

Centralizes niche resources to help newcomers quickly access foundational MLonCode materials, reducing initial research overhead as stated in the project's goals.

Structured Navigation

Categorizes resources into digests, conferences, papers by topic, and software/datasets, making it easy to find specific information without sifting through scattered sources.

Cons

Discontinued Maintenance

The README explicitly states no further updates or issue handling, meaning resources may be outdated and miss recent breakthroughs in the fast-evolving MLonCode field.

No Quality Assessment

While curated, it lacks ratings or reviews for papers and tools, leaving users to independently evaluate the reliability and effectiveness of listed items.

Limited Practical Guidance

Focuses on listing resources rather than providing implementation advice or best practices, which can hinder hands-on application for practitioners.

Frequently Asked Questions

Home

Data Science

vinta

An opinionated list of Python frameworks, libraries, tools, and resources

Stars293,881

Forks27,742

Last commit2 days ago

awesome-machine-learning

A curated list of awesome Machine Learning frameworks, libraries and software.

Stars72,258

Forks15,417

Last commit2 days ago

awesome-awesomeness

A curated list of awesome awesomeness

Stars33,378

Forks3,592

Last commit1 year ago

data-science-ipython-notebooks

Data science Python notebooks: Deep learning (TensorFlow, Theano, Caffe, Keras), scikit-learn, Kaggle, big data (Spark, Hadoop MapReduce, HDFS), matplotlib, pandas, NumPy, SciPy, Python essentials, AWS, and various command lines.

Stars29,020

Forks8,030

Last commit2 years ago

#source-code-analysis

#awesome

#machine-learning

#software-engineering

Data Science28.8k