{"id":13443407,"url":"https://github.com/autowarefoundation/modelzoo","last_synced_at":"2025-08-19T02:42:38.274Z","repository":{"id":36955539,"uuid":"289128277","full_name":"autowarefoundation/modelzoo","owner":"autowarefoundation","description":"A collection of machine-learned models for use in autonomous driving applications.","archived":false,"fork":false,"pushed_at":"2023-02-06T23:09:38.000Z","size":187,"stargazers_count":64,"open_issues_count":3,"forks_count":23,"subscribers_count":6,"default_branch":"master","last_synced_at":"2025-07-25T00:13:24.444Z","etag":null,"topics":["autonomous-driving","autoware","machine-learning","model-zoo"],"latest_commit_sha":null,"homepage":"https://www.autoware.org/","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/autowarefoundation.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":"2020-08-20T23:03:49.000Z","updated_at":"2024-12-30T12:44:58.000Z","dependencies_parsed_at":"2023-02-19T13:01:09.093Z","dependency_job_id":null,"html_url":"https://github.com/autowarefoundation/modelzoo","commit_stats":null,"previous_names":[],"tags_count":11,"template":false,"template_full_name":null,"purl":"pkg:github/autowarefoundation/modelzoo","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/autowarefoundation%2Fmodelzoo","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/autowarefoundation%2Fmodelzoo/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/autowarefoundation%2Fmodelzoo/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/autowarefoundation%2Fmodelzoo/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/autowarefoundation","download_url":"https://codeload.github.com/autowarefoundation/modelzoo/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/autowarefoundation%2Fmodelzoo/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":271092128,"owners_count":24697903,"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","status":"online","status_checked_at":"2025-08-19T02:00:09.176Z","response_time":63,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"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":["autonomous-driving","autoware","machine-learning","model-zoo"],"created_at":"2024-07-31T03:02:00.285Z","updated_at":"2025-08-19T02:42:38.148Z","avatar_url":"https://github.com/autowarefoundation.png","language":"Python","funding_links":[],"categories":["Python","Sensor Processing"],"sub_categories":["Machine Learning"],"readme":"# Autoware Model Zoo\n\nA collection of machine-learned models for use in autonomous driving\napplications.\n\n## License\n\nAll source code published in this repository by default are licensed under\nApache 2.0. If contributions need to be made under a difference license, the\nLICENSE body need to be included in the sub-folder of the model. License for the\nmodel files can be specified using the license fields in\n[definition.yaml](#filling-out-the-definition-file).\n\n## Cloning The Model Zoo\n\nYou need to [install Git LFS](https://git-lfs.github.com/) for large file\nsupport. To clone the zoo:\n\n```sh\n$ GIT_LFS_SKIP_SMUDGE=1 git clone https://github.com/autowarefoundation/modelzoo.git\n$ cd modelzoo\n$ git lfs install\n$ git lfs pull\n```\n\n## Contributing Models\n\n### Folder Structure\n\nThe zoo is organized in terms of the following folder structure:\n\n```sh\nautoware-model-zoo/\u003ctask area\u003e/\u003csub task area\u003e/\u003cmodel name\u003e/\u003cmodel variant name\u003e\n```\n\n- **task area** The general autonomous driving area that the model is trying to\n  tackle. include perception, prediction and planning. The list will expand as\n  we develop further the autoware stack.\n- **sub task area** The specific sub task that the model is trained to do. E.g.\n  camera_traffic_light_detection.\n- **model name** indicates the name of the model architecture, e.g. yolo_v3,\n  ssd_mobilenet.\n- **model variant** this should be an unique name to identify a variant of the\n  model. Variants might differ in training data-set, quantization, input size,\n  etc. You should choose a model variant name that describe best your model and\n  avoid any potential conflict.\n\n### Setup LFS Tracking\n\n`git lfs` identify the files to track by their filename extension. To check what\nextensions are already tracked in this zoo:\n\n```sh\n$ git lfs track\n```\n\nIf you want to contribute something that is not on the list, add the filetype of\nlarge files. For tflite, you would use:\n\n```sh\n$ git lfs track \"*.tflite\"\n$ git add .gitattributes\n$ git commit -m \"tflite files are now tracked by LFS\"\n```\n\n### Adding Models\n\nThis is an example of adding a FP32 object recognition model `yolo_v2_tiny.pb`\nto the repository:\n\n1. Fork the repository and clone your fork follow the steps\n   [here](#cloning-the-model-zoo).\n2. Create a new branch `git checkout -b contrib/\u003cyour_name\u003e/\u003cfeature_name\u003e`.\n3. Create a folder to hold the model following the\n   [folder structure](#folder-structure). For this example, this would be\n   `perception/camera_obstacle_detection/yolo_v2_tiny/tensorflow_fp32_coco`.\n4. Put the model files into this folder:\n   `perception/camera_obstacle_detection/yolo_v2_tiny/tensorflow_fp32_coco/model_files`.\n5. Fill out the `definition.yaml` file and put it at the root of the folder\n   `perception/camera_obstacle_detection/yolo_v2_tiny/tensorflow_fp32_coco`. See\n   guidelines [here](#filling-out-the-definition-file).\n6. Stage and commit all new files\n   `git add perception/camera_obstacle_detection/yolo_v2_tiny/**/*`\n7. Push the changes `git push -u origin contrib/\u003cyour_name\u003e/\u003cfeature_name\u003e` and\n   create a\n   [merge request](https://docs.gitlab.com/ee/user/project/merge_requests/creating_merge_requests.html).\n\n### Filling out the definition file\n\nIn the folder of each of the model, there should be a file named\n`definition.yaml`. You can find a template of the file\n[here](definition_template.yaml). The template contains guide on how to fill out\neach field. Some important fields are discussed in the following paragraphs but\nall attempts should be made to fill out all the fields in the template file.\n\n#### Framework\n\nShould be one of the following: `TensorFlow`, `Caffe`, `TensorFlow Lite`,\n`ONNX`. If you would like to add any other format, follow\n[this step](#setup-lfs-tracking) and update documentation before making a merge\nrequest.\n\n#### Provenance\n\nThe provenance field should be an URL to the \"upstream\" location. If more\ninformation is required, they can be added in the [README](#additional-metadata)\nfile.\n\n#### Tensor size fields\n\nA shape is listed for each input/output node, and the intention is to capture\nthe input/output shape expected by the model. Entries having the number \"-1\"\nhere refer to components for which any size is accepted.\n\n#### Other fields\n\nPlease make every attempt at filling out all the fields in the\n[template](definition_template.yaml). Feel free to add any additional custom\nfields after the template for better documentation and provenance.\n\n### Additional Metadata\n\nIf you feel there are other information that would be relevant to other users,\nplease add a `README.md` to the root of the model's folder. You can share in the\ndocument:\n\n1. Link and description of any script for training\n1. Verbose description of how the training data is obtained and pre-processed\n1. Verbose discussion of bias and limitation in the training data\n1. Guide on how to do transfer learning using this model\n1. Share performance metrics on inference hardware\n1. Share accuracy metrics on well known datasets.\n\nDo commit any scripts or code you can share in the same folder as the models.\n\n### Example pipeline\n\nIt is a good idea to include some code to show the complete inference pipeline.\nThe code should include pre-processing and post-processing. These details are\noften difficult to convey in documentation and is hence best presented as code.\nSome example input data should also be provided to accompany the example\npipeline. E.g. an image, a point cloud file, etc.\n\n## Changing existing models\n\n**Do not overwrite or remove** an existing model without a proper deprecation\nprocess. Once someone has started analyzing, benchmarking or testing a model,\nthey want to be able to refer to the model, and know that that reference will\ncontinue to refer to the same model. Creating a new folder and deprecating the\nold one is preferred over modifying existing models.\n\n## TVM CLI\n\n[TVM](https://github.com/apache/incubator-tvm) is a machine learning compiler\nframework that compiles neutral network models to be executed on different\nhardware back-ends. TVM enables models specified in a wide range of ML\nframeworks to be run on a wide range of hardware devices.\n\nIn this model zoo, a CLI tool has been provided to compile models using TVM.\nThis forms part of the validation workflow of the zoo as well as model\ndeployment workflow for Autoware.\n\nSee [TVM CLI documentation](scripts/tvm_cli/README.md) on how to use the tool.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fautowarefoundation%2Fmodelzoo","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fautowarefoundation%2Fmodelzoo","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fautowarefoundation%2Fmodelzoo/lists"}