ES6.1.2 +LTR 常用DSL

ES6.1.2 +LTR 常用DSL

https://elasticsearch-learning-to-rank.readthedocs.io/
LTR插件中用的DSL是老的版本，新版本很多内容都改变了，新版本特征集构建部分应改为以下形式，在es6.1.2上测试成功。

function 返回某一列

GET tmdb/movie/_search
{"query": {"function_score": {"query": {"match_all": {}},"functions": [{"script_score": {"script": "_score +doc['vote_average'].value"}}]}},"_source": ["original_title","id","vote_average"]
}

LTR构建特征集

#more movie features
PUT _ltr/_featureset/more_movie_features
{"featureset": {"features": [{"name": "body_query","params": ["keywords"],"template": {"match": {"overview": "{{keywords}}"}}},{"name": "title_query","params": ["keywords"],"template": {"match": {"title": "{{keywords}}"}}}]}
}

增加一个Function_score的特征

POST /_ltr/_featureset/more_movie_features/_addfeatures
{"features": [{"name": "user_rating","params": [],"template_language": "mustache","template": {"function_score": {"query": {"match_all": {}},"functions": [{"script_score": {"script": "doc['vote_average'].value"}}]}}}]
}

对以上特征进行查询验证

GET tmdb/movie/_search
{"query": {"bool": {"filter": [{"terms": {"_id": ["7555","1370","1369"]}},{"sltr": { "_name": "logged_featureset","featureset": "more_movie_features","params": {"keywords": "rambo"}}}]}},"ext": {"ltr_log": {"log_specs": {"name": "log_entry1","named_query": "logged_featureset"}}},"_source": ["original_title","id","vote_average"]
}

创建模型

POST _ltr/_featureset/more_movie_features/_createmodel
{"model": {"name": "my_ranklib_model","model": {"type": "model/ranklib","definition": """
## LambdaMART
## No. of trees = 1000
## No. of leaves = 10
## No. of threshold candidates = 256
## Learning rate = 0.1
## Stop early = 100 1  10.357836  2  11.950331  1  6.815881  2  8.725015  1  0.0 
 -2.0 
 -2.0 
 -1.990678310394287  2  8.814796  2  7.3007236 
 -1.9988385438919067 
 -1.9554523229599  1  7.0104666 
 -1.739653468132019 
 -1.516905426979065 
 2.0  1  10.686367 
 2.0 
 2.0 

"""}}
}

查看当前model

GET _ltr/_model/my_ranklib_model

search use model

POST tmdb/_search
{"query": {"query_string": {"query": "rambo"}},"rescore":{"query":{"rescore_query":{"sltr":{"params":{"keywords":"rambo"},"model":"my_ranklib_model"}}}},"ext":{"ltr_log":{"log_specs":{"name":"log_entry1","rescore_index":0}}},"_source": ["original_title","id","vote_average"]
}

取列的值

GET tmdb/movie/71508
POST tmdb/_search
{"query": {"function_score": {"field_value_factor": {"field": "popularity","missing": 0}}},"_source": ["original_title","id","vote_average","popularity"]
}

本文来自互联网用户投稿，文章观点仅代表作者本人，不代表本站立场，不承担相关法律责任。如若转载，请注明出处。 如若内容造成侵权/违法违规/事实不符，请点击【内容举报】进行投诉反馈！