你当前正在访问 Microsoft Azure Global Edition 技术文档网站。 如果需要访问由世纪互联运营的 Microsoft Azure 中国技术文档网站,请访问 https://docs.azure.cn

TargetSentiment 类

TargetSentiment 包含预测的情绪、置信度分数和有关产品/服务的关键组件的其他信息。 例如,在“Foo酒店的食物是好的”,“食物”是“Foo酒店”的关键组成部分。

继承
azure.ai.textanalytics._dict_mixin.DictMixin
TargetSentiment

构造函数

TargetSentiment(**kwargs: Any)

方法

get
has_key
items
keys
update
values

get

get(key: str, default: Any | None = None) -> Any

参数

名称 说明
key
必需
default
默认值: None

has_key

has_key(k: str) -> bool

参数

名称 说明
k
必需

items

items() -> Iterable[Tuple[str, Any]]

keys

keys() -> Iterable[str]

update

update(*args: Any, **kwargs: Any) -> None

values

values() -> Iterable[Any]

属性

confidence_scores

“积极”和“负面”标签的目标的情绪置信度分数介于 0 和 1 之间。 “中性”的分数始终为 0

confidence_scores: SentimentConfidenceScores

length

目标文本长度。 此值取决于原始请求中设置的 string_index_type 参数的值,默认情况下为 UnicodeCodePoints。

length: int

offset

目标文本与文档开头的偏移量。 该值取决于原始请求中设置的 string_index_type 参数的值,默认情况下为 UnicodeCodePoints。

offset: int

sentiment

目标的预测情绪。 可能的值包括“positive”、“mixed”和“negative”。

sentiment: str

text

目标的文本值。

text: str