rasa安装spaCy

拼搏现实的明天。 2022-09-05 01:40 164阅读 0赞

### 文章目录 ###

*   *  rasa nlp 组件
     *  Spacy安装说明
     *  rasa配置
     *  spacy实体抽取

## rasa nlp 组件 ##

Rasa官方提供了多种Language Models的组件引入,包括：

*  MitieNLP
 *  SpacyNLP
 *  HFTransformersNLP

组件说明见，地址：[https://rasa.com/docs/rasa/components][https_rasa.com_docs_rasa_components]

## Spacy安装说明 ##

Spacy提供的安装说明，如下：

pip install -U pip setuptools wheel
    pip install -U spacy
    python -m spacy download en_core_web_sm

详见官网地址：[https://spacy.io/usage\#quickstart][https_spacy.io_usage_quickstart]

下载模型将出现错误，可通过域名反查ip地址，把ip地址写入`hosts`解决。

域名反查ip地址：[https://www.ipaddress.com/][https_www.ipaddress.com]

**实际上**，模型从github地址下载：[https://github.com/explosion/spacy-models/tags][https_github.com_explosion_spacy-models_tags]

所以可以还可以通过以下方式安装：

# With external URL
    $ pip install https://github.com/explosion/spacy-models/releases/download/en_core_web_sm-3.0.0/en_core_web_sm-3.0.0-py3-none-any.whl
    $ pip install https://github.com/explosion/spacy-models/releases/download/en_core_web_sm-3.0.0/en_core_web_sm-3.0.0.tar.gz
    
    # With local file
    $ pip install /Users/you/en_core_web_sm-3.0.0-py3-none-any.whl
    $ pip install /Users/you/en_core_web_sm-3.0.0.tar.gz

## rasa配置 ##

rasa官方配置说明：[https://rasa.com/docs/rasa/tuning-your-model\#sensible-starting-pipelines][https_rasa.com_docs_rasa_tuning-your-model_sensible-starting-pipelines]

我的配置示例：

language: "zh"
    
    pipeline:
      - name: SpacyNLP
        model: "zh_core_web_sm"
        case_sensitive: TRUE
        intent_tokenization_flag: True
        intent_split_symbol: "_"
      - name: SpacyTokenizer
      - name: SpacyFeaturizer
      - name: RegexFeaturizer
      - name: LexicalSyntacticFeaturizer
      - name: CountVectorsFeaturizer
      - name: CountVectorsFeaturizer
        analyzer: "char_wb"
        min_ngram: 1
        max_ngram: 4
      - name: DIETClassifier
        epochs: 100
      - name: EntitySynonymMapper
      - name: ResponseSelector
        epochs: 100
    
    policies:
      - name: MemoizationPolicy
      - name: TEDPolicy
        max_history: 5
        epochs: 100
      - name: RulePolicy

## spacy实体抽取 ##

rasa官网描述：[spacyentityextractor][]

抽取维度见网址：[https://explosion.ai/demos/displacy-ent][https_explosion.ai_demos_displacy-ent]

我的实体抽取示例：

language: "zh"
    
    pipeline:
      - name: SpacyNLP
        model: "zh_core_web_sm"
        case_sensitive: TRUE
        intent_tokenization_flag: True
        intent_split_symbol: "_"
      - name: "SpacyEntityExtractor"
        dimensions: ["PERSON", "LOC", "ORG", "PRODUCT", "DATE", "TIME", "GPE"]
      - name: SpacyTokenizer
      - name: SpacyFeaturizer
      - name: RegexFeaturizer
      - name: LexicalSyntacticFeaturizer
      - name: CountVectorsFeaturizer
      - name: CountVectorsFeaturizer
        analyzer: "char_wb"
        min_ngram: 1
        max_ngram: 4
      - name: DIETClassifier
        epochs: 100
      - name: EntitySynonymMapper
      - name: ResponseSelector
        epochs: 100
    
    policies:
      - name: MemoizationPolicy
      - name: TEDPolicy
        max_history: 5
        epochs: 100
      - name: RulePolicy

![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM1Mzg1Njg3_size_16_color_FFFFFF_t_70_pic_center]

[https_rasa.com_docs_rasa_components]: https://rasa.com/docs/rasa/components
[https_spacy.io_usage_quickstart]: https://spacy.io/usage#quickstart
[https_www.ipaddress.com]: https://www.ipaddress.com/
[https_github.com_explosion_spacy-models_tags]: https://github.com/explosion/spacy-models/tags
[https_rasa.com_docs_rasa_tuning-your-model_sensible-starting-pipelines]: https://rasa.com/docs/rasa/tuning-your-model#sensible-starting-pipelines
[spacyentityextractor]: https://rasa.com/docs/rasa/components#spacyentityextractor
[https_explosion.ai_demos_displacy-ent]: https://explosion.ai/demos/displacy-ent
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM1Mzg1Njg3_size_16_color_FFFFFF_t_70_pic_center]: /images/20220829/6e36ff8eb171440b96a2f4307bc3ea9a.png