intellif-aihub 0.1.1__tar.gz → 0.1.3__tar.gz

intellif_aihub-0.1.3/PKG-INFO

@@ -0,0 +1,110 @@
+Metadata-Version: 2.4
+Name: intellif-aihub
+Version: 0.1.3
+Summary: Intellif AI-hub SDK.
+Author-email: Platform Team <aihub@example.com>
+License-Expression: Apache-2.0
+Keywords: AI-hub,sdk,intellif
+Classifier: Programming Language :: Python :: 3
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.8
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: httpx<0.28,>=0.27
+Requires-Dist: pydantic<3.0,>=2.5.3
+Requires-Dist: typing-extensions<5.0,>=4.13.2
+Requires-Dist: pyarrow<16.0,>=15.0.2
+Requires-Dist: tqdm<5.0,>=4.66
+Requires-Dist: loguru>=0.7.3
+Requires-Dist: minio>=7.2.7
+Dynamic: license-file
+
+# Intellif AI-Hub SDK
+
+**Intellif AI-Hub** 官方 Python 开发包。  
+一个 `Client` 对象即可完成数据集管理、标注统计、任务中心等常见操作，无需手写 HTTP 请求。
+
+```
+aihub_sdk/
+├─ pyproject.toml
+├─ requirements.txt
+├─ src/aihub/
+│   ├─ client.py
+│   ├─ exceptions.py
+│   ├─ models/…
+│   ├─ services/…
+│   └─ utils/…
+└─ tests/
+```
+
+---
+
+## 💻 安装
+
+```bash
+# PyPI 安装
+pip install intellif-aihub
+# 运行环境：Python ≥ 3.8
+```
+
+---
+
+## 🚀 快速上手
+
+```python
+from aihub import Client
+
+BASE  = "http://192.168.13.160:30021"
+TOKEN = "eyJhb..."   # 或设置环境变量：export AI_HUB_TOKEN=...
+
+with Client(base_url=BASE, token=TOKEN) as cli:
+    # 1. 同时创建数据集 + 版本（上传本地 ZIP）
+    ds_id, ver_id, tag = cli.dataset_management.create_dataset_and_version(
+        dataset_name="cats",
+        is_local_upload=True,
+        local_file_path="/data/cats.zip",
+        version_description="first release",
+    )
+    print("数据集标识:", tag)  # 输出：cats/V1
+
+    # 2. 下载数据集
+    cli.dataset_management.run_download(
+        dataset_version_name=tag,
+        local_dir="/tmp/cats",
+        worker=8,
+    )
+
+    # 3. 获取标注平台全局统计
+    stats = cli.labelfree.get_project_global_stats("cat-project")
+    print("总标注数:", stats.global_stats.total_annotations)
+```
+
+---
+
+## 🌍 环境变量
+
+| 变量                       | 作用                                      | 默认值                           |
+|----------------------------|-------------------------------------------|----------------------------------|
+| `AI_HUB_TOKEN`             | API 鉴权 Token（可不在 `Client` 中显式传入） | –                                |
+
+---
+
+## 📦 打包 & 发布
+
+项目采用 PEP 517 / `pyproject.toml` 构建规范。
+
+```bash
+# 1️⃣ 构建 wheel / sdist
+python -m pip install --upgrade build
+python -m build                 # 生成 dist/*.whl dist/*.tar.gz
+
+# 2️⃣ 本地验证
+pip install --force-reinstall dist/*.whl
+python -c "import aihub, sys; print('SDK 版本:', aihub.__version__)"
+
+# 3️⃣ 发布到 PyPI 或私有仓库
+python -m pip install --upgrade twine
+twine upload dist/*
+```
+
+---

intellif_aihub-0.1.3/README.md

@@ -0,0 +1,89 @@
+# Intellif AI-Hub SDK
+
+**Intellif AI-Hub** 官方 Python 开发包。  
+一个 `Client` 对象即可完成数据集管理、标注统计、任务中心等常见操作，无需手写 HTTP 请求。
+
+```
+aihub_sdk/
+├─ pyproject.toml
+├─ requirements.txt
+├─ src/aihub/
+│   ├─ client.py
+│   ├─ exceptions.py
+│   ├─ models/…
+│   ├─ services/…
+│   └─ utils/…
+└─ tests/
+```
+
+---
+
+## 💻 安装
+
+```bash
+# PyPI 安装
+pip install intellif-aihub
+# 运行环境：Python ≥ 3.8
+```
+
+---
+
+## 🚀 快速上手
+
+```python
+from aihub import Client
+
+BASE  = "http://192.168.13.160:30021"
+TOKEN = "eyJhb..."   # 或设置环境变量：export AI_HUB_TOKEN=...
+
+with Client(base_url=BASE, token=TOKEN) as cli:
+    # 1. 同时创建数据集 + 版本（上传本地 ZIP）
+    ds_id, ver_id, tag = cli.dataset_management.create_dataset_and_version(
+        dataset_name="cats",
+        is_local_upload=True,
+        local_file_path="/data/cats.zip",
+        version_description="first release",
+    )
+    print("数据集标识:", tag)  # 输出：cats/V1
+
+    # 2. 下载数据集
+    cli.dataset_management.run_download(
+        dataset_version_name=tag,
+        local_dir="/tmp/cats",
+        worker=8,
+    )
+
+    # 3. 获取标注平台全局统计
+    stats = cli.labelfree.get_project_global_stats("cat-project")
+    print("总标注数:", stats.global_stats.total_annotations)
+```
+
+---
+
+## 🌍 环境变量
+
+| 变量                       | 作用                                      | 默认值                           |
+|----------------------------|-------------------------------------------|----------------------------------|
+| `AI_HUB_TOKEN`             | API 鉴权 Token（可不在 `Client` 中显式传入） | –                                |
+
+---
+
+## 📦 打包 & 发布
+
+项目采用 PEP 517 / `pyproject.toml` 构建规范。
+
+```bash
+# 1️⃣ 构建 wheel / sdist
+python -m pip install --upgrade build
+python -m build                 # 生成 dist/*.whl dist/*.tar.gz
+
+# 2️⃣ 本地验证
+pip install --force-reinstall dist/*.whl
+python -c "import aihub, sys; print('SDK 版本:', aihub.__version__)"
+
+# 3️⃣ 发布到 PyPI 或私有仓库
+python -m pip install --upgrade twine
+twine upload dist/*
+```
+
+---

intellif_aihub-0.1.3/pyproject.toml

@@ -0,0 +1,48 @@
+[project]
+name            = "intellif-aihub"
+version         = "0.1.3"
+description     = "Intellif AI-hub SDK."
+readme          = {file = "README.md", content-type = "text/markdown"}
+requires-python = ">=3.8"
+license         = "Apache-2.0"
+authors         = [{name="Platform Team", email="aihub@example.com"}]
+keywords        = ["AI-hub", "sdk", "intellif"]
+classifiers     = [
+  "Programming Language :: Python :: 3",
+  "Operating System :: OS Independent",
+]
+dependencies    = [
+  "httpx>=0.27,<0.28",
+  "pydantic>=2.5.3,<3.0",
+  "typing-extensions>=4.13.2,<5.0",
+    "pyarrow>=15.0.2,<16.0",
+  "tqdm>=4.66,<5.0",
+  "loguru>=0.7.3",
+  "minio>=7.2.7",
+]
+
+
+[build-system]
+requires = ["setuptools>=61", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[tool.setuptools]
+package-dir = {"" = "src"}
+include-package-data = true
+
+[tool.setuptools.packages.find]
+where = ["src"]
+
+[[tool.uv.index]]
+name = "tsinghua"
+url = "https://pypi.tuna.tsinghua.edu.cn/simple"
+default = true
+
+[dependency-groups]
+dev = [
+    "black>=24.8.0",
+    "griffe-pydantic; python_version>='3.9'",
+    "mkdocs-glightbox>=0.4.0",
+    "mkdocs-material>=9.6.15",
+    "mkdocstrings-python>=1.11.1",
+]

intellif_aihub-0.1.3/src/aihub/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.1.3"

intellif_aihub-0.1.3/src/aihub/client.py

@@ -0,0 +1,91 @@
+from __future__ import annotations
+
+import os
+
+import httpx
+
+from .exceptions import APIError
+from .services import artifact
+from .services import dataset_management
+from .services import document_center
+from .services import labelfree
+from .services import model_training_platform
+from .services import tag_resource_management
+from .services import task_center
+from .services import user_system
+from .services.artifact import ArtifactService
+from .services.dataset_management import DatasetManagementService
+from .services.document_center import DocumentCenterService
+from .services.labelfree import LabelfreeService
+from .services.model_training_platform import ModelTrainingPlatformService
+from .services.tag_resource_management import TagResourceManagementService
+from .services.task_center import TaskCenterService
+from .services.user_system import UserSystemService
+
+
+class Client:
+    """AI-HUB python SDK 客户端
+
+    Attributes:
+        dataset_management (DatasetManagementService): 数据集管理服务
+        labelfree (LabelfreeService): 标注服务
+        task_center (TaskCenterService): 任务中心
+        artifact (ArtifactService): 制品管理
+
+    """
+
+    labelfree: LabelfreeService = None
+    tag_resource_management: TagResourceManagementService = None
+    document_center: DocumentCenterService = None
+    task_center: TaskCenterService = None
+    dataset_management: DatasetManagementService = None
+    artifact: ArtifactService = None
+    user_system: UserSystemService = None
+    model_training_platform: ModelTrainingPlatformService = None
+
+    def __init__(self, *, base_url: str, token: str | None = None, timeout: float = 60.0):
+        """AI-HUB python SDK 客户端
+
+        Args:
+            base_url (str): 服务地址
+            token (str): 密钥，显式传入，或在环境变量AI_HUB_TOKEN中设置
+
+        Examples:
+            >>> from aihub.client import Client
+            >>> client = Client(base_url="xxx", token="xxxx")
+
+        """
+        if not base_url:
+            raise ValueError("base_url必须填写")
+
+        token = os.getenv("AI_HUB_TOKEN") or token
+        if not token:
+            raise ValueError("缺少token：请显式传入，或在环境变量AI_HUB_TOKEN中设置")
+
+        self._http = httpx.Client(
+            base_url=base_url.rstrip("/"),
+            timeout=timeout,
+            headers={"Authorization": f"Bearer {token}"},
+            # event_hooks={"response": [self._raise_for_status]},
+        )
+        self.dataset_management = dataset_management.DatasetManagementService(self._http)
+        self.labelfree = labelfree.LabelfreeService(self._http)
+        self.tag_resource_management = tag_resource_management.TagResourceManagementService(self._http)
+        self.document_center = document_center.DocumentCenterService(self._http)
+        self.task_center = task_center.TaskCenterService(self._http)
+        self.artifact = artifact.ArtifactService(self._http)
+        self.user_system = user_system.UserSystemService(self._http)
+        self.model_training_platform = model_training_platform.ModelTrainingPlatformService(self._http)
+
+    @staticmethod
+    def _raise_for_status(r: httpx.Response):
+        try:
+            r.raise_for_status()
+        except httpx.HTTPStatusError as e:
+            raise APIError(f"{e.response.status_code}: {e.response.text}") from e
+
+    def __enter__(self):
+        return self
+
+    def __exit__(self, *a):
+        self._http.close()

intellif_aihub-0.1.3/src/aihub/exceptions.py

@@ -0,0 +1,18 @@
+from __future__ import annotations
+
+from typing import Any, Optional
+
+
+class SDKError(Exception):
+    pass
+
+
+class APIError(SDKError):
+    def __init__(self, message: str, *, status: Optional[int] = None, detail: Any = None) -> None:
+        super().__init__(message)
+        self.message: str = message
+        self.status: Optional[int] = status
+        self.detail: Any = detail
+
+    def __str__(self) -> str:
+        return f"[HTTP {self.status}] {self.message}" if self.status else self.message

intellif_aihub-0.1.3/src/aihub/models/artifact.py

@@ -0,0 +1,137 @@
+# !/usr/bin/env python
+# -*-coding:utf-8 -*-
+"""制品管理模型模块
+
+该模块定义了制品管理相关的数据模型，包括制品类型、创建制品请求、制品响应等。
+"""
+from __future__ import annotations
+
+from enum import Enum
+from typing import List, Optional
+
+from pydantic import BaseModel
+
+
+class ArtifactType(str, Enum):
+    """制品类型枚举
+
+    定义了系统支持的各种制品类型。
+    """
+
+    dataset = "dataset"  # 数据集类型
+    model = "model"  # 模型类型
+    metrics = "metrics"  # 指标类型
+    log = "log"  # 日志类型
+    checkpoint = "checkpoint"  # 检查点类型
+    image = "image"  # 图像类型
+    prediction = "prediction"  # 预测结果类型
+    other = "other"  # 其他类型
+
+
+class CreateArtifactsReq(BaseModel):
+    """创建制品请求模型
+
+    用于向服务器发送创建制品的请求。
+    """
+
+    entity_id: str
+    """实体ID，通常是运行ID，用于关联制品与特定运行"""
+
+    entity_type: ArtifactType = ArtifactType.other
+    """制品类型，指定制品的类型，默认为other"""
+
+    src_path: str
+    """源路径，制品在系统中的路径标识"""
+
+    is_dir: bool = False
+    """是否为目录，True表示制品是一个目录，False表示是单个文件"""
+
+
+class CreateArtifactsResponseData(BaseModel):
+    """创建制品响应数据模型
+
+    服务器创建制品后返回的数据。
+    """
+
+    id: int  # 制品ID
+    s3_path: str  # S3存储路径
+
+
+class CreateArtifactsResponseModel(BaseModel):
+    """创建制品响应模型
+
+    服务器对创建制品请求的完整响应。
+    """
+
+    code: int  # 响应码，0表示成功
+    msg: str = ""  # 响应消息
+    data: CreateArtifactsResponseData | None = None  # 响应数据
+
+
+class CreatEvalReq(BaseModel):
+    """创建评估请求模型
+
+    用于创建模型评估的请求。
+    """
+
+    dataset_id: int  # 数据集ID
+    dataset_version_id: int  # 数据集版本ID
+    prediction_artifact_path: str  # 预测结果制品路径
+    evaled_artifact_path: str  # 评估结果制品路径
+    run_id: str  # 运行ID
+    user_id: int  # 用户ID
+    report: dict = {}  # 评估报告
+
+
+class ArtifactResp(BaseModel):
+    """制品响应模型
+
+    表示一个制品的详细信息。
+    """
+
+    id: int  # 制品ID
+    entity_type: str  # 实体类型
+    entity_id: str  # 实体ID
+    src_path: str  # 源路径
+    s3_path: str  # S3存储路径
+    is_dir: bool  # 是否为目录
+
+
+class ArtifactRespData(BaseModel):
+    """制品响应数据模型
+
+    包含分页信息和制品列表的响应数据。
+    """
+
+    total: int  # 总记录数
+    page_size: int  # 每页大小
+    page_num: int  # 页码
+    data: List[ArtifactResp]  # 制品列表
+
+
+class ArtifactRespModel(BaseModel):
+    """制品响应模型
+
+    服务器对获取制品请求的完整响应。
+    """
+
+    code: int  # 响应码，0表示成功
+    msg: str = ""  # 响应消息
+    data: ArtifactRespData  # 响应数据
+
+
+# 无限大的页面大小，用于一次性获取所有制品
+InfinityPageSize = 10000 * 100
+
+
+class StsResp(BaseModel):
+    """STS临时凭证响应模型
+
+    包含访问S3存储所需的临时凭证信息。
+    """
+
+    access_key_id: Optional[str] = None  # 访问密钥ID
+    secret_access_key: Optional[str] = None  # 秘密访问密钥
+    session_token: Optional[str] = None  # 会话令牌
+    expiration: Optional[int] = None  # 过期时间
+    endpoint: Optional[str] = None  # 端点URL

intellif_aihub-0.1.3/src/aihub/models/common.py

@@ -0,0 +1,13 @@
+from __future__ import annotations
+
+from typing import Generic, Optional, TypeVar
+
+from pydantic import BaseModel
+
+T = TypeVar("T")
+
+
+class APIWrapper(BaseModel, Generic[T]):
+    code: int
+    msg: Optional[str] = None
+    data: Optional[T]

intellif_aihub-0.1.3/src/aihub/models/dataset_management.py

@@ -0,0 +1,99 @@
+from __future__ import annotations
+
+from typing import List, Optional
+
+from pydantic import BaseModel, Field
+
+
+class CreateDatasetRequest(BaseModel):
+    name: str
+    description: str
+    tags: List[int]
+    cover_img: Optional[str] = Field(None, alias="cover_img")
+    create_by: Optional[int] = Field(None, alias="create_by")
+    is_private: Optional[bool] = Field(None, alias="is_private")
+    access_user_ids: Optional[List[int]] = Field(None, alias="access_user_ids")
+
+
+class CreateDatasetResponse(BaseModel):
+    id: int = Field(alias="id")
+
+
+class DatasetVersionBase(BaseModel):
+    id: int
+    version: int
+    status: int
+    parquet_index_path: Optional[str] = Field(None, alias="parquet_index_path")
+    data_count: int = Field(alias="data_count")
+
+
+class DatasetDetail(BaseModel):
+    id: int
+    name: str
+    description: str
+    cover_img: Optional[str] = Field(None, alias="cover_img")
+    created_at: int = Field(alias="created_at")
+    updated_at: int = Field(alias="update_at")
+    user_id: int = Field(alias="user_id")
+    username: str
+    tags: List[int]
+    access_user_ids: Optional[List[int]] = Field(None, alias="access_user_ids")
+    is_private: Optional[bool] = Field(None, alias="is_private")
+    versions: List[DatasetVersionBase]
+
+
+class ExtInfo(BaseModel):
+    rec_file_path: Optional[str] = Field(None, alias="rec_file_path")
+    idx_file_path: Optional[str] = Field(None, alias="idx_file_path")
+    json_file_path: Optional[str] = Field(None, alias="json_file_path")
+    image_dir_path: Optional[str] = Field(None, alias="image_dir_path")
+
+
+class CreateDatasetVersionRequest(BaseModel):
+    upload_path: str = Field(alias="upload_path")
+    description: Optional[str] = None
+    dataset_id: int = Field(alias="dataset_id")
+    object_cnt: Optional[int] = Field(None, alias="object_cnt")
+    data_size: Optional[int] = Field(None, alias="data_size")
+    create_by: Optional[int] = Field(None, alias="create_by")
+    upload_type: Optional[int] = Field(4, alias="upload_type")
+    ext_info: Optional[ExtInfo] = Field(None, alias="ext_info")
+
+
+class CreateDatasetVersionResponse(BaseModel):
+    id: int = Field(alias="id")
+
+
+class UploadDatasetVersionRequest(BaseModel):
+    upload_path: str = Field(alias="upload_path")
+    upload_type: int = Field(alias="upload_type")
+    dataset_id: int = Field(alias="dataset_id")
+    parent_version_id: Optional[int] = Field(None, alias="parent_version_id")
+    description: Optional[str] = Field(None, alias="description")
+
+
+class UploadDatasetVersionResponse(BaseModel):
+    id: int = Field(alias="id")
+
+
+class DatasetVersionDetail(BaseModel):
+    id: int
+    version: int
+    dataset_id: int = Field(alias="dataset_id")
+    upload_path: str = Field(alias="upload_path")
+    upload_type: int = Field(alias="upload_type")
+    parent_version_id: Optional[int] = Field(None, alias="parent_version_id")
+    description: Optional[str] = None
+    status: int
+    message: Optional[str] = None
+    created_at: int = Field(alias="created_at")
+    user_id: int = Field(alias="user_id")
+    data_size: Optional[int] = Field(None, alias="data_size")
+    data_count: Optional[int] = Field(None, alias="data_count")
+    parquet_index_path: Optional[str] = Field(None, alias="parquet_index_path")
+    ext_info: Optional[ExtInfo] = Field(None, alias="ext_info")
+
+
+class FileUploadData(BaseModel):
+    path: str
+    url: str

intellif_aihub-0.1.3/src/aihub/models/document_center.py

@@ -0,0 +1,28 @@
+# !/usr/bin/env python
+# -*-coding:utf-8 -*-
+
+from __future__ import annotations
+
+from typing import List, Optional, Any
+
+from pydantic import BaseModel
+
+
+class Document(BaseModel):
+    id: int
+    title: str
+    type: int
+    edit_time: int
+    need_update: bool
+    content: str
+    username: str
+    user_id: int
+    created_at: int
+    comments: Optional[Any] = None
+
+
+class GetDocumentsResponse(BaseModel):
+    total: int
+    page_size: int
+    page_num: int
+    data: List[Document]

intellif_aihub-0.1.3/src/aihub/models/labelfree.py

@@ -0,0 +1,31 @@
+from __future__ import annotations
+
+from typing import Optional
+
+from pydantic import BaseModel, Field
+
+
+class Stats(BaseModel):
+    """标注统计信息"""
+
+    total_annotations: int = Field(alias="total_annotations")
+    labeled_annotations: int = Field(alias="labeled_annotations")
+    total_labels: int = Field(alias="total_labels")
+    total_reviews: Optional[int] = Field(None, alias="total_reviews")
+    unlabeled_reviews: Optional[int] = Field(None, alias="unlabeled_reviews")
+    labeled_reviews: Optional[int] = Field(None, alias="labeled_reviews")
+    accepted_count: Optional[int] = Field(None, alias="accepted_count")
+    rejected_count: Optional[int] = Field(None, alias="rejected_count")
+
+
+class GetGlobalStatsResponse(BaseModel):
+    """
+    标注统计概况
+    """
+
+    global_stats: Stats = Field(alias="global_stats")
+    valid_ten_percent: bool = Field(alias="valid_ten_percent")
+    valid_fifty_percent: bool = Field(alias="valid_fifty_percent")
+    valid_hundred_percent: bool = Field(alias="valid_hundred_percent")
+    data_exported_count: int = Field(alias="data_exported_count")
+    exported_dataset_name: str = Field(alias="exported_dataset_name")