{"full_name":"dmlc/gluon-cv","owner":"dmlc","name":"gluon-cv","shard":"1","url":"https://github.com/dmlc/gluon-cv","description":"Gluon CV Toolkit","homepage":"http://gluon-cv.mxnet.io","topics":["deep-learning","computer-vision","neural-network","gluon","mxnet","machine-learning","image-classification","object-detection","semantic-segmentation","gan","person-reid","action-recognition","pose-estimation"],"tag_pages":["computer-vision","deep-learning","gan","image-classification","machine-learning","mxnet","neural-network","object-detection","pose-estimation","semantic-segmentation"],"language":"Python","license":"Apache-2.0","avatar":"https://avatars.githubusercontent.com/u/11508361?v=4","created_at":"2018-02-26T01:33:21Z","pushed_at":"2024-11-25T15:30:52Z","metrics":{"stars":5914,"forks":1202,"watchers":146,"issues":834,"prs":956,"releases":10,"issues_open":40,"issues_closed":794,"prs_open":22,"prs_closed":101,"prs_merged":833,"discussions":0},"growth":{"day":{"stars":0,"forks":0,"base_day":"2026-11-15"}},"history":{"dates":["2026-11-10","2026-11-11","2026-11-12","2026-11-13","2026-11-14","2026-11-15","2026-11-16"],"stars":[5913,5915,5914,5914,5914,5914,5914],"forks":[1202,1202,1202,1202,1202,1202,1202],"watchers":[null,146,146,146,146,146,146],"issues_open":[null,40,40,40,40,40,40],"prs_open":[null,22,22,22,22,22,22]},"boards":{},"same_owner":[],"issues":[{"number":1694,"title":"20bn website is not available, where can I download data?","comments":19,"created_at":"2021-08-03T17:11:50Z"},{"number":1578,"title":"Can I know what is the size of the Kinetics 400 dataset used to reproduce the result in this repo?","comments":14,"created_at":"2020-12-31T11:50:22Z"},{"number":284,"title":"Is it possible training SSD or YOLOv3  in fp16?","comments":13,"created_at":"2018-08-29T13:16:40Z"},{"number":220,"title":"What can i do if i want to use focalloss in SSD model?","comments":10,"created_at":"2018-07-25T00:43:05Z"},{"number":229,"title":"How to use a  trained params to predict?","comments":9,"created_at":"2018-08-02T03:20:00Z"},{"number":800,"title":"Image augmentation: missing rotation","comments":6,"created_at":"2019-06-07T01:10:54Z"},{"number":757,"title":"[Discussion, Call for contribution] GluonCV 0.5 Roadmap","comments":5,"created_at":"2019-04-24T23:07:50Z"},{"number":513,"title":"Train SSD512 on COCO: Low validation accuracy and high smoothL1 loss compared with the official log","comments":5,"created_at":"2018-12-09T09:35:34Z"}],"discussions":[],"latest_release":{"name":"GluonCV 0.10.0 release","tag":"v0.10.0","published_at":"2021-03-09T00:20:06Z","body":"# Highlights \r\nGluonCV 0.10.0 release features a new `Auto Module` designed to bootstrap training tasks with less code and effort:\r\n\r\n\r\n- simpler and better custom dataset loading experience with pandas DataFrame visualization. Comparing with obsolete code based dataset composition, it allows you to load arbitrary datasets faster and more reliable.\r\n\r\n- one liner `fit` function with configuration file support(yaml configuration file)\r\n\r\n- built-in HPO support, for effortless tuning of hyper-parameters\r\n\r\n# gluoncv.auto\r\n\r\nThis release includes a new module called `gluoncv.auto`, with `gluoncv.auto` you can access many high-level APIs such as `data`, `estimators` and `tasks`.\r\n\r\n### gluoncv.auto.data\r\n\r\n`auto.data` module is designed to load arbitrary web datasets you find on the internet, such as Kaggle competition datasets.\r\nYou may refer to this [tutorial](https://cv.gluon.ai/build/examples_auto_module/demo_auto_data.html) or check out the fully compatible [d8 dataset](http://preview.d2l.ai/d8/main/image_classification/getting_started.html) for loading custom datasets.\r\n\r\n##### Loading data:\r\n<img src=\"https://user-images.githubusercontent.com/3307514/109356038-02c50c00-7835-11eb-83ba-2dfc070b4ef6.png\" height=\"150\">\r\n\r\nThe dataset has internal `DataFrame` storage for easier access and analysis\r\n\r\n<img src=\"https://user-images.githubusercontent.com/3307514/109357043-6bf94f00-7836-11eb-9db1-7628218fba43.png\" height=\"300\">\r\n\r\n\r\n##### Visualization:\r\n\r\n<img src=\"https://user-images.githubusercontent.com/3307514/109357126-8cc1a480-7836-11eb-88f5-071c9a817aac.png\" height=\"150\">\r\n\r\nsimilar for object detection:\r\n\r\n<img src=\"https://user-images.githubusercontent.com/3307514/109356385-7b2bcd00-7835-11eb-86e3-8fabc6557464.png\" height=\"200\">\r\n\r\n### gluoncv.auto.estimators\r\n\r\nIn this release, we packed the following high-level estimators for training and predicting images for image classification and object detection.\r\n\r\n- gluoncv.auto.estimators.ImageClassificationEstimator\r\n- gluoncv.auto.estimators.SSDEstimator\r\n- gluoncv.auto.estimators.CenterNetEstimator\r\n- gluoncv.auto.estimators.FasterRCNNEstimator\r\n- gluoncv.auto.estimators.YOLOv3Estimator\r\n\r\n#### Highlighted usages\r\n\r\n- `fit` function:\r\n\r\n![image](https://user-images.githubusercontent.com/3307514/109357783-a57e8a00-7837-11eb-9c99-f2a2fac99112.png)\r\n\r\n- `predict`, `predict_proba`(for image classification), `predict_feature`(for image classification)\r\n\r\n<img src=\"https://user-images.githubusercontent.com/3307514/109358005-ed9dac80-7837-11eb-811c-7690981066ab.png\" height=\"40\">\r\n<img src=\"https://user-images.githubusercontent.com/3307514/109358236-4a00cc00-7838-11eb-8256-2a6e6f51fab8.png\" height=\"150\">\r\n\r\n- `save` and `load`.\r\n\r\nYou may visit the tutorial website for more detailed [examples](https://cv.gluon.ai/build/examples_auto_module/train_image_classifier_basic.html).\r\n\r\n### gluoncv.auto.tasks\r\n\r\nIn this release, the following auto tasks are supported and have been massively tested on many datasets to ensure HPO performance:\r\n\r\n- gluoncv.auto.tasks.ImageClassification\r\n- gluoncv.auto.tasks.ObjectDetection\r\n\r\nComparing with pure algorithm-based estimators, the auto tasks provide identical APIs and functionalities but allow you to `fit` with hyper-parameter optimization(HPO) with specified `num_trials` and `time_limit`. For object detection, it allows multiple algorithms(e.g., SSDEstimator and FasterRCNNEstimator) to be tuned as a categorical search space.\r\n\r\nThe tutorial is available [here](https://cv.gluon.ai/build/examples_auto_module/demo_auto_detection.html)\r\n\r\n# Bug fixes and improvements\r\n\r\n- Improved training speed for mask-rcnn script (#1595, #1609)\r\n- Fix an issue in classification dataset (#1599)\r\n- Fix a batch-size issue for mask-rcnn validation during training (#1594)\r\n- Fix an os directory issue for model zoo folder (#1591)\r\n- Improved CI stability (#1581)","is_prerelease":false},"readme":true,"readme_zh":false,"news":[],"stale":false,"snapshot_at":"2026-11-16T09:23:14Z","history_start":"2026-09-06"}