nreHieW/DeepCoder-Partial-Edits-Synth-both

该数据集包含了编程问题及其解答的相关信息，具体包括问题描述(problem_spec)、错误答案(corrupted_answer)、完整答案(completion)、标准答案(canonical_solution)、提取的答案(extracted_completion)以及答案之间的莱文斯坦距离(levenshtein)。数据集分为训练集和测试集，训练集包含18071个示例，测试集包含99个示例，可用于编程问题解答相关的研究和模型训练。

The dataset includes information related to programming problems and their solutions, specifically problem description (problem_spec), incorrect answers (corrupted_answer), full solutions (completion), canonical solutions (canonical_solution), extracted solutions (extracted_completion), and the Levenshtein distance between solutions (levenshtein). The dataset is split into a training set with 18,071 examples and a test set with 99 examples, which can be used for research and model training related to programming problem solving.