CodeInsight: A Curated Dataset of Practical Coding Solutions from Stack Overflow
This provides a resource for developers and researchers to improve code generation models, but it is incremental as it builds on existing datasets and methods.
The authors tackled the problem of code generation by introducing CodeInsight, a curated dataset of 3,409 practical coding examples from Stack Overflow with clarified intents, code snippets, and unit tests, designed for model finetuning and evaluation, and tested with models like Mistral 7B and GPT-4.
We introduce a novel dataset tailored for code generation, aimed at aiding developers in common tasks. Our dataset provides examples that include a clarified intent, code snippets associated, and an average of three related unit tests. It encompasses a range of libraries such as \texttt{Pandas}, \texttt{Numpy}, and \texttt{Regex}, along with more than 70 standard libraries in Python code derived from Stack Overflow. Comprising 3,409 crafted examples by Python experts, our dataset is designed for both model finetuning and standalone evaluation. To complete unit tests evaluation, we categorize examples in order to get more fine grained analysis, enhancing the understanding of models' strengths and weaknesses in specific coding tasks. The examples have been refined to reduce data contamination, a process confirmed by the performance of three leading models: Mistral 7B, CodeLLaMa 13B, and Starcoder 15B. We further investigate data-contamination testing GPT-4 performance on a part of our dataset. The benchmark can be accessed at \url{https://github.com/NathanaelBeau/CodeInsight}.