{"id":13454937,"url":"https://github.com/tensorpack/tensorpack","last_synced_at":"2025-05-13T16:12:24.298Z","repository":{"id":42000797,"uuid":"48593532","full_name":"tensorpack/tensorpack","owner":"tensorpack","description":"A Neural Net Training Interface on TensorFlow, with focus on speed + flexibility","archived":false,"fork":false,"pushed_at":"2023-08-06T00:30:36.000Z","size":9096,"stargazers_count":6300,"open_issues_count":13,"forks_count":1811,"subscribers_count":196,"default_branch":"master","last_synced_at":"2024-10-29T14:50:34.126Z","etag":null,"topics":["deep-learning","machine-learning","neural-networks","reinforcement-learning","tensorflow"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"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":"CHANGES.md","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}},"created_at":"2015-12-25T23:08:44.000Z","updated_at":"2024-10-18T07:57:05.000Z","dependencies_parsed_at":"2023-01-31T06:01:39.074Z","dependency_job_id":null,"html_url":"https://github.com/tensorpack/tensorpack","commit_stats":{"total_commits":2923,"total_committers":59,"mean_commits":49.54237288135593,"dds":"0.16968867601778992","last_synced_commit":"1547a54e8546494614ca31c984a1bfd1d0e24b77"},"previous_names":["ppwwyyxx/tensorpack"],"tags_count":41,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tensorpack%2Ftensorpack","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tensorpack%2Ftensorpack/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tensorpack%2Ftensorpack/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tensorpack%2Ftensorpack/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/tensorpack","download_url":"https://codeload.github.com/tensorpack/tensorpack/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":250514767,"owners_count":21443208,"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":["deep-learning","machine-learning","neural-networks","reinforcement-learning","tensorflow"],"created_at":"2024-07-31T08:00:59.577Z","updated_at":"2025-04-23T21:02:10.761Z","avatar_url":"https://github.com/tensorpack.png","language":"Python","readme":"![Tensorpack](https://github.com/tensorpack/tensorpack/raw/master/.github/tensorpack.png)\n\nTensorpack is a neural network training interface based on graph-mode TensorFlow.\n\n[![ReadTheDoc](https://readthedocs.org/projects/tensorpack/badge/?version=latest)](http://tensorpack.readthedocs.io)\n[![Gitter chat](https://img.shields.io/badge/chat-on%20gitter-46bc99.svg)](https://gitter.im/tensorpack/users)\n[![model-zoo](https://img.shields.io/badge/model-zoo-brightgreen.svg)](http://models.tensorpack.com)\n## Features:\n\nIt's Yet Another TF high-level API, with the following highlights:\n\n1. Focus on __training speed__.\n  + Speed comes for free with Tensorpack -- it uses TensorFlow in the __efficient way__ with no extra overhead.\n    On common CNNs, it runs training [1.2~5x faster](https://github.com/tensorpack/benchmarks/tree/master/other-wrappers) than the equivalent Keras code.\n    Your training can probably gets faster if written with Tensorpack.\n\n  + Scalable data-parallel multi-GPU / distributed training strategy is off-the-shelf to use.\n    See [tensorpack/benchmarks](https://github.com/tensorpack/benchmarks) for more benchmarks.\n\n2. Squeeze the best data loading performance of Python with [`tensorpack.dataflow`](https://github.com/tensorpack/dataflow).\n  + Symbolic programming (e.g. `tf.data`) [does not](https://tensorpack.readthedocs.io/tutorial/philosophy/dataflow.html#alternative-data-loading-solutions)\n    offer the data processing flexibility needed in research.\n    Tensorpack squeezes the most performance out of __pure Python__ with various autoparallelization strategies.\n\n3. Focus on reproducible and flexible research:\n  + Built and used by researchers, we provide high-quality [reproducible implementation of papers](https://github.com/tensorpack/tensorpack#examples).\n\n4. It's not a model wrapper.\n  + There are too many symbolic function wrappers already. Tensorpack includes only a few common layers.\n    You can use any TF symbolic functions inside Tensorpack, including tf.layers/Keras/slim/tflearn/tensorlayer/....\n\nSee [tutorials and documentations](http://tensorpack.readthedocs.io/tutorial/index.html#user-tutorials) to know more about these features.\n\n## Examples:\n\nWe refuse toy examples.\nInstead of showing tiny CNNs trained on MNIST/Cifar10,\nwe provide training scripts that reproduce well-known papers.\n\nWe refuse low-quality implementations.\nUnlike most open source repos which only __implement__ papers,\n[Tensorpack examples](examples) faithfully __reproduce__ papers,\ndemonstrating its __flexibility__ for actual research.\n\n### Vision:\n+ [Train ResNet](examples/ResNet) and [other models](examples/ImageNetModels) on ImageNet\n+ [Train Mask/Faster R-CNN on COCO object detection](examples/FasterRCNN)\n+ [Unsupervised learning with Momentum Contrast](https://github.com/ppwwyyxx/moco.tensorflow) (MoCo)\n+ [Adversarial training with state-of-the-art robustness](https://github.com/facebookresearch/ImageNet-Adversarial-Training)\n+ [Generative Adversarial Network(GAN) variants](examples/GAN), including DCGAN, InfoGAN, Conditional GAN, WGAN, BEGAN, DiscoGAN, Image to Image, CycleGAN\n+ [DoReFa-Net: train binary / low-bitwidth CNN on ImageNet](examples/DoReFa-Net)\n+ [Fully-convolutional Network for Holistically-Nested Edge Detection(HED)](examples/HED)\n+ [Spatial Transformer Networks on MNIST addition](examples/SpatialTransformer)\n+ [Visualize CNN saliency maps](examples/Saliency)\n\n### Reinforcement Learning:\n+ [Deep Q-Network(DQN) variants on Atari games](examples/DeepQNetwork), including DQN, DoubleDQN, DuelingDQN.\n+ [Asynchronous Advantage Actor-Critic(A3C) with demos on OpenAI Gym](examples/A3C-Gym)\n\n### Speech / NLP:\n+ [LSTM-CTC for speech recognition](examples/CTC-TIMIT)\n+ [char-rnn for fun](examples/Char-RNN)\n+ [LSTM language model on PennTreebank](examples/PennTreebank)\n\n## Install:\n\nDependencies:\n\n+ Python 3.3+.\n+ Python bindings for OpenCV. (Optional, but required by a lot of features)\n+ TensorFlow ≥ 1.5\n  * TF is not not required if you only want to use `tensorpack.dataflow` alone as a data processing library\n  * When using TF2, tensorpack uses its TF1 compatibility mode. Note that a few examples in the repo are not yet migrated to support TF2.\n```\npip install --upgrade git+https://github.com/tensorpack/tensorpack.git\n# or add `--user` to install to user's local directories\n```\n\nPlease note that tensorpack is not yet stable.\nIf you use tensorpack in your code, remember to mark the exact version of tensorpack you use as your dependencies.\n\n## Citing Tensorpack:\n\nIf you use Tensorpack in your research or wish to refer to the examples, please cite with:\n```\n@misc{wu2016tensorpack,\n  title={Tensorpack},\n  author={Wu, Yuxin and others},\n  howpublished={\\url{https://github.com/tensorpack/}},\n  year={2016}\n}\n```\n","funding_links":[],"categories":["Python","Libraries","The Data Science Toolbox","Deep Learning Framework","其他_机器学习与深度学习","机器学习框架"],"sub_categories":["Deep Learning Packages","High-Level DL APIs"],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftensorpack%2Ftensorpack","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ftensorpack%2Ftensorpack","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftensorpack%2Ftensorpack/lists"}