跳到内容

rg.Query

要根据搜索条件收集记录，您可以使用 Query 和 Filter 类。Query 类用于定义搜索条件，而 Filter 类用于过滤搜索结果。Filter 传递给 Query 对象，因此您可以组合多个过滤器来创建复杂的搜索查询。Query 对象也可以传递给 Dataset.records，以根据搜索条件获取记录。

使用示例

搜索包含术语的记录

要搜索包含术语的记录，您可以将 Dataset.records 属性与查询字符串一起使用。搜索术语用于搜索在文本字段中包含这些术语的记录。

for record in dataset.records(query="paris"):
    print(record)

按条件过滤记录

Argilla 允许您根据条件过滤记录。您可以使用 Filter 类定义条件，并将它们传递给 Dataset.records 属性，以根据条件获取记录。条件包括 "==", ">=", "<=" 或 "in"。条件可以与点表示法结合使用，以根据元数据、建议或响应来过滤记录。

# create a range from 10 to 20
range_filter = rg.Filter(
    [
        ("metadata.count", ">=", 10),
        ("metadata.count", "<=", 20)
    ]
)

# query records with metadata count greater than 10 and less than 20
query = rg.Query(filters=range_filter, query="paris")

# iterate over the results
for record in dataset.records(query=query):
    print(record)

Query

此类用于将用户查询映射到内部查询模型

源代码在 src/argilla/records/_search.py 中 
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
class Query:
    """This class is used to map user queries to the internal query models"""

    def __init__(
        self,
        *,
        query: Union[str, None] = None,
        similar: Union[Similar, None] = None,
        filter: Union[Filter, Conditions, None] = None,
    ):
        """Create a query object for use in Argilla search requests.add()

        Parameters:
            query (Union[str, None], optional): The query string that will be used to search.
            similar (Union[Similar, None], optional): The similar object that will be used to search for similar records
            filter (Union[Filter, None], optional): The filter object that will be used to filter the search results.
        """

        if isinstance(filter, tuple):
            filter = [filter]

        if isinstance(filter, list):
            filter = Filter(conditions=filter)

        self.query = query
        self.filter = filter
        self.similar = similar

    def has_search(self) -> bool:
        return bool(self.query or self.has_similar() or self.filter)

    def has_similar(self) -> bool:
        return bool(self.similar)

    def api_model(self) -> SearchQueryModel:
        model = SearchQueryModel()

        if self.query or self.similar:
            query = QueryModel()

            if self.query is not None:
                query.text = TextQueryModel(q=self.query)

            if self.similar is not None:
                query.vector = self.similar.api_model()

            model.query = query

        if self.filter is not None:
            model.filters = self.filter.api_model()

        return model

__init__(*, query=None, similar=None, filter=None)

创建用于 Argilla 搜索请求的查询对象。add()

参数

名称 类型 描述 默认值
query Union[str, None]

将用于搜索的查询字符串。

 无
similar Union[Similar, None]

将用于搜索相似记录的 similar 对象

 无
filter Union[Filter, None]

将用于过滤搜索结果的 filter 对象。

 无
源代码在 src/argilla/records/_search.py 中 
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
def __init__(
    self,
    *,
    query: Union[str, None] = None,
    similar: Union[Similar, None] = None,
    filter: Union[Filter, Conditions, None] = None,
):
    """Create a query object for use in Argilla search requests.add()

    Parameters:
        query (Union[str, None], optional): The query string that will be used to search.
        similar (Union[Similar, None], optional): The similar object that will be used to search for similar records
        filter (Union[Filter, None], optional): The filter object that will be used to filter the search results.
    """

    if isinstance(filter, tuple):
        filter = [filter]

    if isinstance(filter, list):
        filter = Filter(conditions=filter)

    self.query = query
    self.filter = filter
    self.similar = similar

Filter

此类用于将用户过滤器映射到内部过滤器模型

源代码在 src/argilla/records/_search.py 中 
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
class Filter:
    """This class is used to map user filters to the internal filter models"""

    def __init__(self, conditions: Union[Conditions, None] = None):
        """ Create a filter object for use in Argilla search requests.

        Parameters:
            conditions (Union[List[Tuple[str, str, Any]], Tuple[str, str, Any], None], optional): \
                The conditions that will be used to filter the search results. \
                The conditions should be a list of tuples where each tuple contains \
                the field, operator, and value. For example `("label", "in", ["positive","happy"])`.\
        """

        if isinstance(conditions, tuple):
            conditions = [conditions]
        self.conditions = [Condition(condition) for condition in conditions]

    def api_model(self) -> AndFilterModel:
        return AndFilterModel.model_validate({"and": [condition.api_model() for condition in self.conditions]})

__init__(conditions=None)

创建用于 Argilla 搜索请求的过滤器对象。

参数

名称 类型 描述 默认值
conditions Union[List[Tuple[str, str, Any]], Tuple[str, str, Any], None]

将用于过滤搜索结果的条件。条件应为元组列表，其中每个元组包含字段、运算符和值。例如 ("label", "in", ["positive","happy"])。

 无
源代码在 src/argilla/records/_search.py 中 
131
132
133
134
135
136
137
138
139
140
141
142
143
def __init__(self, conditions: Union[Conditions, None] = None):
    """ Create a filter object for use in Argilla search requests.

    Parameters:
        conditions (Union[List[Tuple[str, str, Any]], Tuple[str, str, Any], None], optional): \
            The conditions that will be used to filter the search results. \
            The conditions should be a list of tuples where each tuple contains \
            the field, operator, and value. For example `("label", "in", ["positive","happy"])`.\
    """

    if isinstance(conditions, tuple):
        conditions = [conditions]
    self.conditions = [Condition(condition) for condition in conditions]

Similar

此类用于将用户相似性查询映射到内部查询模型

源代码在 src/argilla/records/_search.py 中 
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
class Similar:
    """This class is used to map user similar queries to the internal query models"""

    def __init__(self, name: str, value: Union[Iterable[float], "Record"], most_similar: bool = True):
        """
        Create a similar object for use in Argilla search requests.

        Parameters:
            name: The name of the vector field
            value: The vector value or the record to search for similar records
            most_similar: Whether to search for the most similar records or the least similar records
        """

        self.name = name
        self.value = value
        self.most_similar = most_similar if most_similar is not None else True

    def api_model(self) -> VectorQueryModel:
        from argilla.records import Record

        order = "most_similar" if self.most_similar else "least_similar"

        if isinstance(self.value, Record):
            return VectorQueryModel(name=self.name, record_id=self.value._server_id, order=order)

        return VectorQueryModel(name=self.name, value=self.value, order=order)

__init__(name, value, most_similar=True)

创建用于 Argilla 搜索请求的 similar 对象。

参数

名称 类型 描述 默认值
name str

向量字段的名称

 required
value Union[Iterable[float], Record]

向量值或用于搜索相似记录的记录

 required
most_similar bool

是否搜索最相似的记录或最不相似的记录

 True
源代码在 src/argilla/records/_search.py 中 
103
104
105
106
107
108
109
110
111
112
113
114
115
def __init__(self, name: str, value: Union[Iterable[float], "Record"], most_similar: bool = True):
    """
    Create a similar object for use in Argilla search requests.

    Parameters:
        name: The name of the vector field
        value: The vector value or the record to search for similar records
        most_similar: Whether to search for the most similar records or the least similar records
    """

    self.name = name
    self.value = value
    self.most_similar = most_similar if most_similar is not None else True