{"id":13414939,"url":"https://github.com/commaai/research","last_synced_at":"2025-04-13T04:00:11.023Z","repository":{"id":64175123,"uuid":"64815511","full_name":"commaai/research","owner":"commaai","description":"dataset and code for 2016 paper \"Learning a Driving Simulator\"","archived":false,"fork":false,"pushed_at":"2022-08-16T09:22:12.000Z","size":48876,"stargazers_count":4128,"open_issues_count":37,"forks_count":1170,"subscribers_count":377,"default_branch":"master","last_synced_at":"2025-04-13T04:00:01.026Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"bsd-3-clause","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/commaai.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}},"created_at":"2016-08-03T04:38:57.000Z","updated_at":"2025-04-08T22:48:51.000Z","dependencies_parsed_at":"2023-01-15T02:31:10.844Z","dependency_job_id":null,"html_url":"https://github.com/commaai/research","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/commaai%2Fresearch","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/commaai%2Fresearch/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/commaai%2Fresearch/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/commaai%2Fresearch/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/commaai","download_url":"https://codeload.github.com/commaai/research/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248661706,"owners_count":21141450,"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":"2024-07-30T21:00:40.110Z","updated_at":"2025-04-13T04:00:10.974Z","avatar_url":"https://github.com/commaai.png","language":"Python","funding_links":[],"categories":["Uncategorized","Python","Miscellaneous","Datasets and Benchmarks","Data","Datasets"],"sub_categories":["Uncategorized","Sources"],"readme":"# the people's comma\n\n## the paper\n\n[Learning a Driving Simulator](http://arxiv.org/abs/1608.01230)\n\n## the comma.ai driving dataset\n\n7 and a quarter hours of largely highway driving. Enough to train what we had in [Bloomberg](http://www.bloomberg.com/features/2015-george-hotz-self-driving-car/).\n\n## Examples\n\nWe present two Machine Learning Experiments to show\npossible ways to use this dataset:\n\n\n\u003cimg src=\"./images/selfsteer.gif\"\u003e\n\n[Training a steering angle predictor](SelfSteering.md)\n\n\n\u003cimg src=\"./images/drive_simulator.gif\"\u003e\n\n[Training a generative image model](DriveSim.md)\n\n## Downloading the dataset\n\n```\n./get_data.sh\n```\n\nor get it at [archive.org comma dataset](https://archive.org/details/comma-dataset)\n\n45 GB compressed, 80 GB uncompressed\n\n```\ndog/2016-01-30--11-24-51 (7.7G)\ndog/2016-01-30--13-46-00 (8.5G)\ndog/2016-01-31--19-19-25 (3.0G)\ndog/2016-02-02--10-16-58 (8.1G)\ndog/2016-02-08--14-56-28 (3.9G)\ndog/2016-02-11--21-32-47 (13G)\ndog/2016-03-29--10-50-20 (12G)\nemily/2016-04-21--14-48-08 (4.4G)\nemily/2016-05-12--22-20-00 (7.5G)\nfrodo/2016-06-02--21-39-29 (6.5G)\nfrodo/2016-06-08--11-46-01 (2.7G)\n```\n\nDataset referenced on this page is copyrighted by comma.ai and published under the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 License. This means that you must attribute the work in the manner specified by the authors, you may not use this work for commercial purposes and if you alter, transform, or build upon this work, you may distribute the resulting work only under the same license.\n\n## Dataset structure\nThe dataset consists of 10 videos clips of variable size recorded at 20 Hz\nwith a camera mounted on the windshield of an Acura ILX 2016. In parallel to the videos\nwe also recorded some measurements such as car's speed, acceleration,\nsteering angle, GPS coordinates, gyroscope angles. See the full `log` list [here](Logs.md).\nThese measurements are transformed into a uniform 100 Hz time base.\n\nThe dataset folder structure is the following:\n```bash\n+-- dataset\n|   +-- camera\n|   |   +-- 2016-04-21--14-48-08\n|   |   ...\n|   +-- log\n|   |   +-- 2016-04-21--14-48-08\n|   |   ...\n```\n\nAll the files come in hdf5 format and are named with the time they were recorded.\nThe camera dataset has shape `number_frames x 3 x 160 x 320` and `uint8` type.\nOne of the `log` hdf5-datasets is called `cam1_ptr` and addresses the alignment\nbetween camera frames and the other measurements.\n\n## Requirements\n[anaconda](https://www.continuum.io/downloads)  \n[tensorflow-0.9](https://github.com/tensorflow/tensorflow)  \n[keras-1.0.6](https://github.com/fchollet/keras)  \n[cv2](https://anaconda.org/menpo/opencv3)\n\n## Hiring\n\nWant a job at [comma.ai](http://comma.ai)?\n\nShow us amazing stuff on this dataset\n\n## Credits\n\nRiccardo Biasini, George Hotz, Sam Khalandovsky, Eder Santana, and Niel van der Westhuizen\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcommaai%2Fresearch","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fcommaai%2Fresearch","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcommaai%2Fresearch/lists"}