WrittenWithRust's picture
Update README.md
fb7d211 verified
|
Raw
History Blame Contribute Delete
3.6 kB
metadata
license: mit
language:
  - en
tags:
  - rust
  - code
  - instruction-tuning
  - magicoder
  - oss-instruct
pretty_name: Magicoder OSS Instruct Rust (3.9K Cleaned)
size_categories:
  - 1K<n<10K
task_categories:
  - text-generation

๐Ÿฆ€ Magicoder-OSS-Instruct-Rust (3.9K Cleaned)

Magicoder-OSS-Instruct-Rust is a high-quality, syntax-verified dataset of 3,909 Rust coding instructions derived from real-world open-source GitHub projects.

This dataset is extracted from ise-uiuc/Magicoder-OSS-Instruct-75K, filtered specifically for Rust, and validated via in-memory compiler checks. No language translation was applied; the dataset remains in its original English format.


โš™๏ธ Filtering and Verification Methodology

The dataset was processed using the following technical pipeline:

  • Language Filtering: Extracted entries strictly where lang: rust, removing all other programming languages.
  • In-Memory Syntax Validation (rustc RAM Check): All code snippets were evaluated on-the-fly via rustc --crate-type=lib through standard input (stdin). Entries with broken syntax, unclosed braces, or invalid AST structures were automatically discarded.
  • Format Standardization: Converted raw problem/solution pairs into standard ChatML (messages) format for direct compatibility with SFT and Fine-Tuning frameworks (LoRA / QLoRA).

โš ๏ธ Filtering Summary (75K -> 3.9K)

Out of the original 75,000 multi-language samples:

  1. Non-Rust programming languages were filtered out.
  2. Incomplete snippets or entries failing rustc syntax checks were removed.
  3. A clean, verified set of 3,909 high-quality Rust instructions was produced.

๐Ÿ“Š Dataset Structure

The dataset follows the ChatML (messages) format:

  • system: System prompt assigning the expert Rust developer role.
  • user: Rust programming task, module design requirement, or algorithmic problem.
  • assistant: Corresponding idiomatic Rust solution.

Schema Details

Field Type Description
id int64 Original dataset index
task_category string Task category identifier (oss_instruct)
messages list Conversation history containing role and content (system, user, assistant)

๐Ÿ’ก Example Entry

{
  "id": 1601,
  "task_category": "oss_instruct",
  "messages": [
    {
      "role": "system",
      "content": "You are an expert assistant who solves Rust programming problems and generates correct code."
    },
    {
      "role": "user",
      "content": "You are tasked with creating a Rust module system for a 2D game engine..."
    },
    {
      "role": "assistant",
      "content": "```rust\n// input_handler.rs\npub mod input_handler {\n    pub struct InputHandler {\n        // Implementation details\n    }\n}\n\n// input.rs\npub mod input {\n    pub use input_handler::InputHandler;\n    pub use keyboard::Keyboard;\n    pub use mouse::Mouse;\n    pub use gamepad::Gamepad;\n}\n```\n\nIn this solution, we have created the input modules..."
    }
  ]
}

๐Ÿ’ป Usage

from datasets import load_dataset

dataset = load_dataset("WrittenWithRust/Magicoder-OSS-Instruct-Rust-3.9K")
print(dataset["train"][0])

๐Ÿ“œ Citation

@article{wei2023magicoder,
  title={Magicoder: Source Code Is All You Need},
  author={Wei, Yuxiang and Wang, Zhe and Liu, Jiawei and Ding, Yuhang and Zhang, Lingming},
  journal={arXiv preprint arXiv:2312.02120},
  year={2023}
}