feat(annotation): 添加文本数据集段落切片功能

- 在前端组件中新增 segmentationEnabled 字段控制切片开关
- 为文本数据集添加段落切片配置选项,默认启用切片功能
- 在后端接口中新增 segmentation_enabled 参数传递给标注项目
- 实现切片逻辑控制,支持文本数据的自动段落分割
- 添加数据集类型判断,仅文本数据集支持切片配置
- 更新标注任务创建和编辑表单中的切片相关字段处理
This commit is contained in:
2026-01-26 12:05:21 +08:00
parent fa160164d2
commit 371df12a96
7 changed files with 218 additions and 32 deletions

View File

@@ -23,6 +23,11 @@ class DatasetMappingCreateRequest(BaseModel):
description: Optional[str] = Field(None, alias="description", description="标注项目描述")
template_id: Optional[str] = Field(None, alias="templateId", description="标注模板ID")
label_config: Optional[str] = Field(None, alias="labelConfig", description="Label Studio XML配置")
segmentation_enabled: Optional[bool] = Field(
None,
alias="segmentationEnabled",
description="是否启用文本分段",
)
class Config:
# allow population by field name when constructing model programmatically
@@ -49,6 +54,11 @@ class DatasetMappingResponse(BaseModel):
template_id: Optional[str] = Field(None, alias="templateId", description="关联的模板ID")
template: Optional['AnnotationTemplateResponse'] = Field(None, description="关联的标注模板详情")
label_config: Optional[str] = Field(None, alias="labelConfig", description="实际使用的 Label Studio XML 配置")
segmentation_enabled: Optional[bool] = Field(
None,
alias="segmentationEnabled",
description="是否启用文本分段",
)
total_count: int = Field(0, alias="totalCount", description="数据集总数据量")
annotated_count: int = Field(0, alias="annotatedCount", description="已标注数据量")
created_at: datetime = Field(..., alias="createdAt", description="创建时间")
@@ -62,4 +72,4 @@ class DatasetMappingResponse(BaseModel):
class DeleteDatasetResponse(BaseResponseModel):
"""删除数据集响应模型"""
id: str = Field(..., description="映射UUID")
status: str = Field(..., description="删除状态")
status: str = Field(..., description="删除状态")