cwv-bench-v1
收藏资源简介:
该数据集包含与Web性能、代码复杂度和浏览器指标相关的详细数据。数据集结构复杂,包含多个嵌套字段,主要特征包括:ID、REPO_ID、FRAMEWORK、HOST_FILE_PATH等基础信息;IS_LIVE状态检查结果;cwv_mobile移动端性能指标(包括CLS、FCP、FID、INP、LCP、TTFB等核心Web指标的均值、中位数、评分和标准差);CODE_STATS代码统计信息(包括按语言分类的文件数、行数和大小,以及依赖项);COMPLEXITY_METRICS代码复杂度指标(如平均嵌套深度、最大嵌套深度、总函数数等);BROWSER_METRICS浏览器指标(如控制台错误、资源大小、请求统计等);以及THIRD_PARTY_EXTERNAL第三方集成信息(如Google Analytics、Segment等营销工具的使用情况)。数据集包含3,698个训练样本,总大小约为7.3MB。适用于Web性能分析、代码质量评估和第三方集成研究等任务。
This dataset contains detailed data related to Web performance, code complexity, and browser metrics. The dataset has a complex structure with multiple nested fields, and its main features include: basic information such as ID, REPO_ID, FRAMEWORK, HOST_FILE_PATH and other basic attributes; IS_LIVE status check results; cwv_mobile mobile-side performance metrics, including the mean, median, score and standard deviation of core Web metrics such as CLS, FCP, FID, INP, LCP, TTFB; CODE_STATS code statistics, including the number of files, lines of code and size categorized by programming language, as well as dependencies; COMPLEXITY_METRICS code complexity metrics such as average nesting depth, maximum nesting depth, total number of functions and other indicators; BROWSER_METRICS browser metrics such as console errors, resource sizes, request statistics and other related data; THIRD_PARTY_EXTERNAL third-party integration information such as the usage of marketing tools including Google Analytics, Segment and others. The dataset includes 3,698 training samples with a total size of approximately 7.3 MB. It is suitable for tasks such as Web performance analysis, code quality assessment and third-party integration research.




