| --- |
| license: cc-by-4.0 |
| tags: |
| - code |
| - source-code |
| - data-structures |
| - algorithms |
| - competitive-programming |
| - software-engineering |
| - computer-science |
| - code-generation |
| - code-completion |
| - code-understanding |
| - program-synthesis |
| - llm |
| - reasoning |
| - problem-solving |
| - coding-interviews |
| - leetcode |
| - sft |
| - rlhf |
| - self-supervised-learning |
| task_categories: |
| - text-generation |
| pretty_name: 'Codeing Dataset ' |
| size_categories: |
| - 1M<n<10M |
| --- |
| |
| ### Dataset Description |
|
|
| **This dataset is a large-scale collection of legacy codebase data, containing 17,554 files, 6.30 million lines of code, and 117.74 million lexical tokens, designed to support the development and training of advanced code generation models, software engineering AI systems, code modernization tools, and intelligent programming assistants.** |
|
|
| It consists of real-world software projects containing mature and long-lived codebases that reflect practical development patterns, architectural designs, maintenance workflows, refactoring requirements, and production software engineering practices. The dataset captures diverse coding styles, programming constructs, dependency structures, and software evolution patterns commonly encountered in enterprise and large-scale software systems. |
|
|
| This makes it highly valuable for building accurate, scalable, and production-ready AI systems for code generation, code completion, code understanding, software modernization, automated refactoring, bug detection, technical debt analysis, and software engineering research. Additionally, this dataset can be utilized in pipelines for **Supervised Fine-Tuning (SFT), Self-Supervised Learning (SSL), Reinforcement Learning with Human Feedback (RLHF), and code intelligence workflows.** |
|
|
| Dataset Specification |
|
|
| Total Files: 17,554 |
| Total Lines of Code: 6.30 Million |
| Total Lexical Tokens: 117.74 Million |
| Domain: Legacy Software Systems |
| Data Type: Source Code |
| Nature: Real-world and curated code repositories |
| |
| **Key Use Cases** |
|
|
| Code generation and completion |
| Automated debugging and error detection |
| Software development assistance |
| Program understanding and reasoning |
| Code summarization and documentation generation |
| |
|
|
| **Value of This Dataset** |
|
|
| Enables advanced AI training for coding and software engineering systems |
| Improves AI understanding of programming logic and software workflows |
| Supports AI systems for debugging, code reasoning, and developer assistance |
| Helps build intelligent coding copilots and automated software agents |
| Enhances performance in code-language and multimodal reasoning tasks |
| |
| **Basic JSON Schema** |
|
|
| ```json |
| { |
| "file_name": "string", |
| "file_path": "string", |
| "language": "string", |
| "code": "string" |
| } |
| ``` |
| **Data Creation** |
|
|
| Procured through formal agreements and generated in the ordinary course of business. |
|
|
|
|
| **Considerations** |
|
|
| This dataset is provided for research and educational purposes only. It contains only sample data. For access to the full dataset and enterprise licensing options, please visit our website [InfoBay.AI](https://infobay.ai/) or contact us directly. |
|
|
| -Ph: (91) 8303174762 |
| -Email: datareq@infobay.ai |
