← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 6490 · 生成模型:GPT-6

映射归约聚合同类词频

理解等级:中级

映射阶段把文档内容转换为中间键值对,归约阶段按相同键聚合计数。键的选择决定统计属于哪个作者、文档或词项;并行处理依赖这种可拆分与再聚合结构,不能把每段都并行就当成结果已经正确合并。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      Mapper breaks the title's sentence in word-by-word format and assigns a frequency to these words. One Master copy is produced out of the numerous copies. In the framework, there are N Map and Reduce tasks to allocate. The master selects an idle worker, and Map/Reduce task is assigned. The mapper returns an intermediate <key, value> pair for each word in a document.

    参考来源

    - 位置:《Artificial Intelligence and Data Science in Recommendation System: Current Trends, Technologies, and Applications》第05章《主题建模与作者专长推荐》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Artificial Intelligence and Data Science in Recommendation System: Current Trends, Technologies, and Applications/_epub-src/OEBPS/Text/9789815136746-C5.html) - 结论依据:摘录中关于映射归约聚合同类词频的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:Mapper breaks the title's sentence in word-by-word format and assigns a frequency to these words. One Master copy is produced out of the numerous copies. In the framework, there are N Map and Reduce tasks to allocate. The master selects an idle worker, and Map/Reduce task is assigned. The mapper returns an intermediate <key, value> pair for each word in a document.