{"id":18499134,"url":"https://github.com/tensorpack/dataflow","last_synced_at":"2025-06-14T14:06:43.162Z","repository":{"id":66049358,"uuid":"188529314","full_name":"tensorpack/dataflow","owner":"tensorpack","description":"Efficient Data Loading Pipeline in Pure Python","archived":false,"fork":false,"pushed_at":"2020-08-19T02:07:00.000Z","size":1001,"stargazers_count":211,"open_issues_count":0,"forks_count":16,"subscribers_count":8,"default_branch":"master","last_synced_at":"2025-04-08T09:51:18.541Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":false,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/tensorpack.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2019-05-25T06:28:32.000Z","updated_at":"2024-12-22T01:23:53.000Z","dependencies_parsed_at":"2023-04-11T14:35:12.071Z","dependency_job_id":null,"html_url":"https://github.com/tensorpack/dataflow","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/tensorpack/dataflow","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tensorpack%2Fdataflow","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tensorpack%2Fdataflow/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tensorpack%2Fdataflow/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tensorpack%2Fdataflow/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/tensorpack","download_url":"https://codeload.github.com/tensorpack/dataflow/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tensorpack%2Fdataflow/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":259827649,"owners_count":22917711,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-11-06T13:44:47.803Z","updated_at":"2025-06-14T14:06:43.112Z","avatar_url":"https://github.com/tensorpack.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Tensorpack DataFlow\n\nTensorpack DataFlow is an **efficient** and **flexible** data\nloading pipeline for deep learning, written in pure Python.\n\nIts main features are:\n\n1. **Highly-optimized for speed**.\n\t Parallelization in Python is hard and most libraries do it wrong.\n\t DataFlow implements highly-optimized\n\t parallel building blocks which gives you an easy interface to parallelize your workload.\n\n2. **Written in pure Python**.\n\t This allows it to be used together with any other Python-based library.\n\nDataFlow is originally part of the [tensorpack library](https://github.com/tensorpack/tensorpack/)\nand has been through many years of polishing.\nGiven its independence of the rest of the tensorpack library,\nit is now a separate library whose source code is synced with tensorpack.\nPlease use [tensorpack issues](https://github.com/tensorpack/tensorpack/issues/) for support.\n\nWhy would you want to use DataFlow instead of a platform-specific data loading solutions?\nWe recommend you to read\n[Why DataFlow?](https://tensorpack.readthedocs.io/tutorial/philosophy/dataflow.html).\n\n## Install:\n```\npip install --upgrade git+https://github.com/tensorpack/dataflow.git\n# or add `--user` to install to user's local directories\n```\nYou may also need to install opencv, which is used by many builtin DataFlows.\n\n## Examples:\n```python\nimport dataflow as D\nd = D.ILSVRC12('/path/to/imagenet')  # produce [img, label]\nd = D.MapDataComponent(d, lambda img: some_transform(img), index=0)\nd = D.MultiProcessMapData(d, num_proc=10, lambda img, label: other_transform(img, label))\nd = D.BatchData(d, 64)\nd.reset_state()\nfor img, label in d:\n  # ...\n```\n\n## Documentation:\n### Tutorials:\n1. [Basics](https://tensorpack.readthedocs.io/tutorial/dataflow.html)\n1. [Why DataFlow?](https://tensorpack.readthedocs.io/tutorial/philosophy/dataflow.html)\n1. [Write a DataFlow](https://tensorpack.readthedocs.io/tutorial/extend/dataflow.html)\n1. [Parallel DataFlow](https://tensorpack.readthedocs.io/tutorial/parallel-dataflow.html)\n1. [Efficient DataFlow](https://tensorpack.readthedocs.io/tutorial/efficient-dataflow.html)\n\n### APIs:\n1. [Built-in DataFlows](https://tensorpack.readthedocs.io/modules/dataflow.html)\n1. [Built-in Datasets](https://tensorpack.readthedocs.io/modules/dataflow.dataset.html)\n\n## Support \u0026 Contributing\n\nPlease send issues and pull requests (for the `dataflow/` directory) to the\n[tensorpack project](https://github.com/tensorpack/tensorpack/) where the source code is developed.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftensorpack%2Fdataflow","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ftensorpack%2Fdataflow","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftensorpack%2Fdataflow/lists"}