知识卡片
词频神经输入仍要保持坐标契约
词袋、TF–IDF或词元序列均可作为不同网络的输入,必须为每个位置固定对应含义。输入数值中的房屋面积若仅作为字符串词项,不自动保留其数值大小;是否抽取结构化数值仍是任务设计选择。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
- · 神经网络基础
原书出处 · 1 处
- · 查看原书依据
What were your inputs when you did a logistic regression on natural language sentences in earlier chapters? You first processed the text with a keyword detector, CountVectorizer, or TfidfVectorizer. These models use a tokenizer, like the ones you learned about in chapter 2, to split the text into individual words and then count them up. So for NLP, it’s common to use the bag-of-words (BOW)
参考来源
- 位置:《Natural Language Processing in Action, Second Edition》第05章《神经网络基础》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Natural Language Processing in Action, Second Edition/_epub-src/OEBPS/Text/chapter-5.html)
- 结论依据:摘录中关于词频神经输入仍要保持坐标契约的定义、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:What were your inputs when you did a logistic regression on natural language sentences in earlier chapters? You first processed the text with a keyword detector, CountVectorizer, or TfidfVectorizer. These models use a tokenizer, like the ones you learned about in chapter 2, to split the text into individual words and then count them up. So for NLP, it’s common to use the bag-of-words (BOW)