Skip to main content

LeRobot-style dataset I/O:core / io / schemas(本地数据集与录制;不含 Web 服务)。

Project description

lsdataset

独立仓库中的 LeRobot 风格数据集库:coreLsRobotDatasetLsDatasetManager)、ioschemas。负责本地数据集的创建、写入、episode 生命周期管理与磁盘元信息;不包含 Web 框架或 HTTP 服务实现。

PyPI 项目名与 import 名均为 lsdataset

安装

pip install -e .

版本单一来源datasets / pyarrow / jsonlines / avpydanticnumpyscipyPillowpyproject.toml

异步图像写入并发度(默认 4):

  • LSDATASET_IMAGE_WRITER_PROCESSES
  • LSDATASET_IMAGE_WRITER_THREADS

测试

pip install -e ".[dev]"
make test
# 或:python -m pytest tests -q --tb=short

集成脚本在 scripts/(多为本地调试;parquet2json.py 需设置 LSDATASET_PARQUET2JSON_IN 等环境变量,见脚本内说明)。本仓库 tests/ 覆盖核心数据集与编码流程。

版本与依赖声明

在宿主项目的依赖文件中 pin lsdataset==x.y.z(与 PyPIeditable 安装一致即可)。

本地 editable 示例:

pip install -e /path/to/lsdataset

版本号与发布脚本

python scripts/release.py patch --sync-readme
make release PART=minor
make release-build PART=patch

--sync-readme 会替换本仓库 README.md 中出现的 lsdataset==旧版本,以及历史上可能存在的 lsdataset[…]==旧版本 写法。

发版后:更新依赖方声明的版本号,再执行 twine upload 等发布步骤。


发布到 PyPI(摘要)

  1. python scripts/release.py patch --sync-readme(或手改 version
  2. python -m buildtwine check dist/*twine upload dist/*
  3. git tag v<x.y.z>

近期主版本语义(摘要)0.2.0 起移除包内 HTTP 路由模块; 0.3.0 起移除与外部上传网关对接的实现子包; 0.3.1 起精简 schemas.types,删除仅用于该上传流程的请求模型(由集成方自行定义)。若曾依赖已删除的导入路径,请在集成侧替换实现。

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

lsdataset-0.3.3.tar.gz (56.2 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

lsdataset-0.3.3-py3-none-any.whl (67.1 kB view details)

Uploaded Python 3

File details

Details for the file lsdataset-0.3.3.tar.gz.

File metadata

  • Download URL: lsdataset-0.3.3.tar.gz
  • Upload date:
  • Size: 56.2 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.14

File hashes

Hashes for lsdataset-0.3.3.tar.gz
Algorithm Hash digest
SHA256 9bee13d4946597feeba070b84ae6a71f16e923b4bddc867f1ed611bbc71c7dce
MD5 9867c13b45b2e3efb075490de3306421
BLAKE2b-256 0b7c600096feedeadb0cf8854a341f74c63912971d149c84e0cd394a015c1179

See more details on using hashes here.

File details

Details for the file lsdataset-0.3.3-py3-none-any.whl.

File metadata

  • Download URL: lsdataset-0.3.3-py3-none-any.whl
  • Upload date:
  • Size: 67.1 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.14

File hashes

Hashes for lsdataset-0.3.3-py3-none-any.whl
Algorithm Hash digest
SHA256 92c75852ec86126ce417b326001707d1b677d28422cc29f146462189af4a158d
MD5 5a791b34eb49e7e38369273afe581423
BLAKE2b-256 c0dfe87b09682e2a6403566a98ac8bd084973c06e956e23b13d8751490437256

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page