JanSchTech/starcoderdata-python-edu-lang-score

`starcoderdata-python-edu-lang-score dataset`是`starcoderdata dataset`的Python子集，增加了评估代码教育质量和分类代码注释语言的功能。该数据集的主要目的是支持高质量的Python教育和基于语言的训练，特别关注教育评分和特定语言的代码注释。数据集适用于教育内容评估和多语言代码理解等应用。每个记录包含代码内容、清理后的内容、检测到的语言、语言分类的置信度分数、提取的注释、教育评分及其整数表示。数据集的创建目的是通过教育质量分类识别高质量代码，并过滤用于编写代码注释的自然语言。数据收集和处理步骤包括内容清理、语言分类和教育评分。

The `starcoderdata-python-edu-lang-score dataset` is a Python subset of the `starcoderdata dataset`, augmented with features that assess the educational quality of code and classify the language of code comments. The primary purpose of this dataset is to support high-quality Python education and language-based training, with a focus on educational scores and specific languages for code comments. The dataset is suitable for applications such as educational content evaluation and multilingual code understanding. Each record includes code content, cleaned content, detected language, language classification confidence score, extracted comments, educational score, and its integer representation. The creation of the dataset aims to identify high-quality code through educational quality classification and filter natural languages used for writing code comments. Data collection and processing steps include content cleaning, language classification, and educational scoring.