如何解决Duckling 不提取时间和有序实体
我在我的管道中使用 DuckingEntityExtractor 来提取时间和序数实体,但它没有提取一些以前用于 DucklingHTTPExtractor 的实体。但是它适用于小鸭解析器。
这是我的管道和实体示例的片段:
管道:
language: en
pipeline:
- name: Whitespacetokenizer
- name: RegexFeaturizer
- name: LexicalSyntacticFeaturizer
- name: CountVectorsFeaturizer
- name: CountVectorsFeaturizer
analyzer: char_wb
min_ngram: 1
max_ngram: 4
- name: DIETClassifier
epochs: 100
- name: EntitySynonymMapper
- name: "DucklingHTTPExtractor"
# url of the running duckling server
url: "http://xxxx.xxx.com"
# dimensions to extract
dimensions: ["time","number","email","ordinal"]
# allows you to configure the locale,by default the language is
# used
locale: "de_DE"
- name: FallbackClassifier
threshold: 0.7
ambiguity_threshold: 0.1
示例:
正文:9 月 22 日
Next Message:
22 sep
{
"text": "22 sep","intent": {
"id": -1711287076799534405,"name": "book_slot","confidence": 0.9992443919181824
},"entities": [
{
"start": 0,"end": 6,"text": "22 sep","value": "2021-09-22T00:00:00.000-07:00","confidence": 1.0,"additional_info": {
"values": [
{
"value": "2021-09-22T00:00:00.000-07:00","grain": "day","type": "value"
},{
"value": "2022-09-22T00:00:00.000-07:00",{
"value": "2023-09-22T00:00:00.000-07:00","type": "value"
}
],"type": "value"
},"entity": "time","extractor": "DucklingEntityExtractor"
}
]
正文:下周五
next friday
{
"text": "next friday","intent": {
"id": 1386275381868498380,"name": "goodbye","confidence": 0.9995949268341064
},"entities": []
}
文字:第一
Next message:
First
{
"text": "First","intent": {
"name": "nlu_fallback","confidence": 0.5972589552402496
},"entities": []
}
小鸭解析器输出:
cmd: curl -d "text=next friday" -X POST http://xxxx.xxx.com/parse
[{"body":"next friday","start":0,"value":{"values":[{"value":"2021-01-15T00:00:00.000-08:00","grain":"day","type":"value"}],"value":"2021-01-15T00:00:00.000-08:00","type":"value"},"end":11,"dim":"time","latent":false}]
有人可以帮忙吗?
版权声明:本文内容由互联网用户自发贡献,该文观点与技术仅代表作者本人。本站仅提供信息存储空间服务,不拥有所有权,不承担相关法律责任。如发现本站有涉嫌侵权/违法违规的内容, 请发送邮件至 dio@foxmail.com 举报,一经查实,本站将立刻删除。