使用各种模型预测文本类的简单工具。
text-classif的Python项目详细描述
#textclassive
model
*fasttext char
*fasttext word
*cnn char embedding
*cnn word embedding
*cnn char&word embedding
*cnn+bigru+char&;嵌入字词
*pyttp
*jieba
<嵌入
*fasttext(cbow/跳过gram)
*gensim
```python
>从文本分类导入文本分类
35;默认参数
t=textclassive()
text='
logtis=t.predict(文本,精准度<16')
t.indexx2label
t.indexx2label
t.get-top-label(text,k=5,精准度=16')
````
*模型:'fasttext'(默认值)、'cnn'cnn'cnn'mcnn'mcnn'mccnn'mgcnn'
t.t.get-top-top-label(text,k=5,k=16')
````/>*剪切:true,false(默认值)
*剪切模型:“pyltp”(默认值),“揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭.pth'
*cnn_word_型号:/home/home/keming/keming/github/custom-recom/cnn-word_-fulltext_best.pth'
*mcnn_模型:'/home/keming/github/custom-recom/mcnn-fulltext_-best.pth'
*mgcnn_模型:'/home/keming/github/custom-recom/mgcnn-fulltext_-best.pth'
*char嵌入模型:'/data-hdd/embedding/wiki/wiki-char-char_-char_-256.model'
*word/word/word/word嵌入模型:/data-hdd/embedding/wiki-word-256.model'
*words-index:'/data-hdd/zhhu/topic/topic/words.csv'
*chars-index:'/data-hdd/zhhud/zhhud/zhhud/topic/chars.csv'
*labels-index:'/data-hdd/zhhud/zhhuu/topic/topics.csv'
*delete-chard/zhhud/zhhud/zhhuu/topic/topictopics.csv'
*deletete-char char char:'/data/data-chard/data/chihud/chihud/zhhud/topics_dim:256
*num_filter:128
*char_句子长度:256
*字句长度:128
*字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字>
*文本
*精度:“16”(默认),“32”,“64”
`文本分类。获取Top_Label`
*文本
*k:5(默认值),要返回的标签数
*精度:'16'(默认值)、'32'、'64'
model
*fasttext char
*fasttext word
*cnn char embedding
*cnn word embedding
*cnn char&word embedding
*cnn+bigru+char&;嵌入字词
*pyttp
*jieba
<嵌入
*fasttext(cbow/跳过gram)
*gensim
```python
>从文本分类导入文本分类
35;默认参数
t=textclassive()
text='
logtis=t.predict(文本,精准度<16')
t.indexx2label
t.indexx2label
t.get-top-label(text,k=5,精准度=16')
````
*模型:'fasttext'(默认值)、'cnn'cnn'cnn'mcnn'mcnn'mccnn'mgcnn'
t.t.get-top-top-label(text,k=5,k=16')
````/>*剪切:true,false(默认值)
*剪切模型:“pyltp”(默认值),“揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭揭.pth'
*cnn_word_型号:/home/home/keming/keming/github/custom-recom/cnn-word_-fulltext_best.pth'
*mcnn_模型:'/home/keming/github/custom-recom/mcnn-fulltext_-best.pth'
*mgcnn_模型:'/home/keming/github/custom-recom/mgcnn-fulltext_-best.pth'
*char嵌入模型:'/data-hdd/embedding/wiki/wiki-char-char_-char_-256.model'
*word/word/word/word嵌入模型:/data-hdd/embedding/wiki-word-256.model'
*words-index:'/data-hdd/zhhu/topic/topic/words.csv'
*chars-index:'/data-hdd/zhhud/zhhud/zhhud/topic/chars.csv'
*labels-index:'/data-hdd/zhhud/zhhuu/topic/topics.csv'
*delete-chard/zhhud/zhhud/zhhuu/topic/topictopics.csv'
*deletete-char char char:'/data/data-chard/data/chihud/chihud/zhhud/topics_dim:256
*num_filter:128
*char_句子长度:256
*字句长度:128
*字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字字>
*文本
*精度:“16”(默认),“32”,“64”
`文本分类。获取Top_Label`
*文本
*k:5(默认值),要返回的标签数
*精度:'16'(默认值)、'32'、'64'