Functions — github.com/baidu/Familia

CodeHub MCPcopy Create free account

hub / github.com/baidu/Familia / functions

Functions145 in github.com/baidu/Familia

⨍ Functions 145◇ Types & classes 29

↓ 46 callersMethodsize

src/vocab.cpp:20

↓ 19 callersMethodsize

返回文档中词的数量

include/familia/document.h:65

↓ 12 callersMethodnum_topics

src/semantic_matching.cpp:32

↓ 12 callersFunctionsplit

src/util.cpp:11

↓ 12 callersMethodtopic_sum

src/model.cpp:25

↓ 11 callersFunctionrand

返回min~max之间的随机浮点数, 默认返回0~1的浮点数

include/familia/util.h:44

↓ 10 callersMethodinfer

src/inference_engine.cpp:37

↓ 8 callersMethodalpha

include/familia/model.h:75

↓ 7 callersMethodbeta

include/familia/model.h:83

↓ 7 callersMethodbeta_sum

include/familia/model.h:87

↓ 7 callersMethodtokenize

分词简易的FMM分词工具，只针对主题模型中出现的词表进行正向匹配 Args: input_text: 输入文本，编码需要跟词表编码保持一致 Returns: 返回一个list对象，其中每个元素为分

python/demo/familia_wrapper.py:47

↓ 7 callersMethodword_topic

返回模型中某个词在某个主题下的参数值，由于模型采用稀疏存储，若找不到则返回0

include/familia/model.h:42

↓ 6 callersMethodsparse_topic_dist

src/document.cpp:43

↓ 5 callersFunctionrand_k

返回[0, k - 1]之间的整型浮点数

include/familia/util.h:52

↓ 5 callersMethodtokenize

src/tokenizer.cpp:15

↓ 4 callersMethoddense_topic_dist

src/document.cpp:64

↓ 4 callersMethodnum_topics

src/demo/topic_word_demo.cpp:57

↓ 4 callersMethodset_topic

src/document.cpp:31

↓ 3 callersMethodcontains_word

src/semantic_matching.cpp:28

↓ 3 callersFunctionprint_result

python/demo/topic_word_demo.py:15

↓ 3 callersMethodterm_id

include/familia/model.h:34

↓ 2 callersMethodaccumulate_topic_sum

src/document.cpp:77

↓ 2 callersMethodalpha_sum

include/familia/model.h:79

↓ 2 callersFunctionfix_random_seed

固定随机种子并重置分布

include/familia/util.h:36

↓ 2 callersMethodgenerate

src/vose_alias.cpp:62

↓ 2 callersMethodget_model

返回模型指针以便获取模型参数

include/familia/inference_engine.h:53

↓ 2 callersMethodinit

-------------LDA Begin---------------

src/document.cpp:16

↓ 2 callersMethodinitialize

src/vose_alias.cpp:14

↓ 2 callersFunctionload_prototxt

include/familia/util.h:57

↓ 2 callersMethodmodel_type

返回模型类型, 指明为LDA还是SetennceLDA

include/familia/inference_engine.h:58

↓ 2 callersMethodnearest_words

寻求与目标词最相关的词对模型中的所有词语(不包哈目标词)进行检索，通过计算cosine相似度，返回最相关的k个词语 Args: word: 目标词 k: 设置返回最邻近词的个数，默认为10个 Retur

python/demo/familia_wrapper.py:164

↓ 2 callersMethodnearest_words_around_topic

寻求与目标主题最相关的词对模型中的所有词语进行检索，通过计算cosine相似度，返回最相关的k个词语 Args: topic_id: 目标主题ID k: 设置返回最邻近词的个数，默认为10个 Retur

python/demo/familia_wrapper.py:181

↓ 2 callersFunctionprint_doc_topic_dist

打印文档的主题分布

src/demo/inference_demo.cpp:25

↓ 2 callersMethodsample_doc

src/sampler.cpp:11

↓ 2 callersMethodset_alpha

配置文档先验参数alpha

include/familia/document.h:60

↓ 2 callersMethodtype

include/familia/model.h:91

↓ 1 callersMethodadd_sentence

src/document.cpp:93

↓ 1 callersMethodadd_token

src/document.cpp:24

↓ 1 callersMethodcal_doc_distance

计算长文本与长文本之间的距离计算两个长文本的主题分布之间的距离，包括jensen_shannon_divergence和hellinger_distance Args: doc1: 输入文档1，无需分词 doc2:

python/demo/familia_wrapper.py:101

↓ 1 callersMethodcal_doc_distance

计算长文本之间的相似度可选的指标包括常用的分布间距离Jensen-Shannon Diveregnce和Hellinger Distance

src/demo/doc_distance_demo.cpp:38

↓ 1 callersMethodcal_query_doc_similarity

计算短文本与长文本之间的相关性使用LDA模型和TWE模型分别衡量短文本跟长文本之间的相关性 Args: query: 输入短文本，无需分词 doc: 输入长文本，无需分词 Returns:

python/demo/familia_wrapper.py:120

↓ 1 callersMethodcal_query_doc_similarity

计算query (短文本) 与 document (长文本) 的相似度可选的指标包括: 1. document主题分布生成query的likelihood, 值越大相似度越高 2. 基于TWE模型的相似度计算

src/demo/query_doc_sim_demo.cpp:42

↓ 1 callersMethodfind_nearest_words

获取词距离最近的k个词，并打印出来

src/demo/word_distance_demo.cpp:33

↓ 1 callersMethodget_id

src/vocab.cpp:15

↓ 1 callersFunctionlda_infer

使用LDA模型对输入的分词文本进行infer

python/cpp/familia_wrapper.cpp:161

↓ 1 callersMethodlda_infer

LDA模型推断使用LDA模型对输入文本进行推断，得到其主题分布 Args: words: 分词后的list结果 Returns: 返回一个list对象，存放输入文本对应的稀疏主题分布，list中每

python/demo/familia_wrapper.py:60

↓ 1 callersMethodload

src/vocab.cpp:24

↓ 1 callersFunctionload_item_topic_table

加载模型信息

src/demo/show_topic_demo.cpp:66

↓ 1 callersFunctionload_vocabulary

加载词典信息

src/demo/show_topic_demo.cpp:45

↓ 1 callersMethodnearest_words

src/semantic_matching.cpp:78

↓ 1 callersMethodnearest_words_around_topic

src/semantic_matching.cpp:100

↓ 1 callersFunctionprint_result

python/demo/word_distance_demo.py:15

↓ 1 callersMethodshow_topics

打印指定主题下的前k个词

src/demo/show_topic_demo.cpp:114

↓ 1 callersMethodshow_topics

展示同个主题下不同召回方式的结果

src/demo/topic_word_demo.cpp:39

↓ 1 callersFunctionslda_infer

使用SentenceLDA模型对输入的分词文本进行infer

python/cpp/familia_wrapper.cpp:193

↓ 1 callersMethodslda_infer

SentenceLDA模型推断使用SentenceLDA模型对输入文本进行推断，得到其主题分布 Args: sentences: 其中每个元素为一个list对象，存放一个句子例如： [[A

python/demo/familia_wrapper.py:77

↓ 1 callersMethodtopic_sum

include/familia/document.h:69

↓ 1 callersMethodvocab_size

include/familia/model.h:71

MethodDocDistanceDemo

src/demo/doc_distance_demo.cpp:29

MethodGibbsSampler

include/familia/sampler.h:110

MethodInferenceEngine

src/inference_engine.cpp:16

include/familia/document.h:40

MethodMHSampler

include/familia/sampler.h:36

FunctionPyInit_familia

python/cpp/familia_wrapper.cpp:431

MethodQueryDocSimDemo

src/demo/query_doc_sim_demo.cpp:30

include/familia/document.h:102

MethodShowTopicDemo

src/demo/show_topic_demo.cpp:97

MethodSimpleTokenizer

include/familia/tokenizer.h:33

MethodTokenizer

include/familia/tokenizer.h:22

MethodTopicModel

src/model.cpp:13

MethodTopicModel

include/familia/model.h:30

MethodTopicWordDemo

src/demo/topic_word_demo.cpp:31

MethodTopicalWordEmbedding

include/familia/semantic_matching.h:35

include/familia/vocab.h:22

MethodVoseAlias

include/familia/vose_alias.h:19

MethodWordCount

src/demo/show_topic_demo.cpp:35

MethodWordDistanceDemo

src/demo/word_distance_demo.cpp:27

销毁各个对象指针

python/demo/familia_wrapper.py:40

销毁TWE对象

python/demo/familia_wrapper.py:159

初始化InferenceEngine,Tokenizer,TopicalWordEmbeddings对象指针 Args: model_dir: 模型目录 conf_file: 模型配置文件 e

python/demo/familia_wrapper.py:24

初始化TWE对象 Args: model_dir: 模型目录路径 emb_file: topical word embeddings模型文件

python/demo/familia_wrapper.py:151

Functioncal_doc_distance

计算长文本与长文本之间的距离

python/cpp/familia_wrapper.cpp:232

Functioncal_query_doc_similarity

计算短文本与长文本之间的相似度

python/cpp/familia_wrapper.cpp:280

Methodconstruct_alias_table

src/sampler.cpp:200

src/tokenizer.cpp:55

Methodcosine_similarity

计算两个embedding的余弦相似度

include/familia/semantic_matching.h:94

Functiondestroy_inference_engine

销毁InferenceEngine对象

python/cpp/familia_wrapper.cpp:60

Functiondestroy_tokenizer

销毁Tokenizer对象

python/cpp/familia_wrapper.cpp:92

Functiondestroy_twe

销毁Topical Word Embeddings对象

python/cpp/familia_wrapper.cpp:122

Methoddoc_proposal

src/sampler.cpp:61

Methoddoc_proposal_distribution

src/sampler.cpp:189

Methodengine_wrapper_t

include/familia/util.h:25

Methodhellinger_distance

Hellinger Distance REQUIRE: 传入的两个参数维度须一致

include/familia/semantic_matching.h:202

Functioninit_inference_engine

创建InferenceEngine对象

python/cpp/familia_wrapper.cpp:33

Functioninit_tokenizer

创建Tokenizer对象用来分词

python/cpp/familia_wrapper.cpp:75

Functioninit_twe

创建Topical Word Embeddings对象

python/cpp/familia_wrapper.cpp:105

Functioninitfamilia

模块初始化

python/cpp/familia_wrapper.cpp:437

Methodis_eng_char

检查字符是否为英文字符

include/familia/tokenizer.h:49

Methodjensen_shannon_divergence

Jensen-Shannon Divergence REQUIRE: 传入的两个参数维度须一致

include/familia/semantic_matching.h:181

Methodkullback_leibler_divergence

Kullback Leibler Divergence D(P||Q) = \sum_i {P(i) ln \frac {P(i)}{Q(i)} REQUIRE: 传入的两个参数维度须一致

include/familia/semantic_matching.h:168

next →1–100 of 145, ranked by callers