crumb/Functions-53K

Functions-53K数据集包含53,000个代码样本，这些样本经过注释，包括输入和输出类型以及使用所需的依赖项或库。所有变量名都经过匿名化处理，以减少高频特征的影响，使模型空间更平滑。数据集来源于550,000个文件，经过清理、匿名化和去重处理，最终得到53,000个样本。fifty-three thousand samples of code, annotated with input and output types as well as any dependencies or libraries required for their use. completely anonymized variables. i anonymized all the variables to sort of low-pass a bunch of high frequency features... a model of the space should be smoother to navigate. what ever you think im doing with this its cooler than that.