简体   繁体   English

BerTopic Model - 可视化忽略第 0 个索引

[英]BerTopic Model - Visualization ignores 0th index

The BerTopic model resulted the below Topics: BerTopic model 产生了以下主题:

在此处输入图像描述

As you can see from the above, the model is finetuned to generate lesser outliers '-1' which has the count of 3 and it appears in the last.从上面可以看出,model 被微调以生成较小的异常值“-1”,其计数为 3,并出现在最后。

While visualizing the Topics per class , 根据 class 可视化主题时

topic_model.visualize_topics_per_class(topics_per_class)

the below interactive visual is generated, and however it ignored the 0th index, to be precise the Topic 0. The Global Topic Representations are displayed from 1, 2, 3, 4, 5, 6, -1生成了以下交互式视觉效果,但是它忽略了0th个索引,准确地说是主题 0。全局主题表示从1, 2, 3, 4, 5, 6, -1显示

在此处输入图像描述

Is the BerTopic designed in a way that it always assumes the very first index will be an outlier ( -1 ), and eliminates it blindly? BerTopic 的设计方式是否总是假设第一个索引将是异常值 ( -1 ),并盲目地消除它?

Are the generated topics always accessed based on the count size, may be in descending order?生成的主题是否总是根据计数大小访问,可能是按降序排列的?

This issue is posted in the BerTopic github forum as well, and the response from the Author himself,这个问题也发布在BerTopic github论坛,以及作者本人的回复,

在此处输入图像描述

by setting top_n_topics=None , all the topics along with the 0th index can be viewed while visualizing,通过设置top_n_topics=None ,可以在可视化的同时查看所有主题以及0th个索引,

topic_model.visualize_topics_per_class(topics_per_class, top_n_topics=None)

声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

相关问题 Backward algorithm Hidden Markov Model, 0th index (termination step) 产生错误结果 - Backward algorithm Hidden Markov Model, 0th index (termination step) yields wrong result 有没有办法多次找到第 0 个索引? - Is there a way to find the 0th index a variable amount of times? 根据第 0 个索引处的项目对 2D 嵌套列表进行排序 - Sort 2D nested list based on item at 0th index 从字符串之间但从第0个索引和第-1个索引中删除引号 - Remove Quotation mark from in between of the String but from 0th index and -1th index python中如何删除未命名的索引行,将第0行替换为索引行进行数据分析 - How to delete Unnamed index row and replace 0th row as index row in python for data analysis Python - 将浮点数转换为 int 以用作数组索引返回数组的第 0 个元素 - Python - Casting a float as int to be used as array index returns 0th element of the array 无法加载 BERTopic model - Can't load BERTopic model 想要通过从第 0 个索引开始在其存在旁边插入每个元素直到满足条件来使列表等于目标长度 - Want to make a list equal to a target length by inserting every element next to its presence starting from 0th index until condition is satisfied 所有行都期望第0行 - All the rows expect 0th Row NLTK wordnet界面中的第0个synset - 0th synset in NLTK wordnet interface
 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM