{"id":13788649,"url":"https://github.com/ShuangXieIrene/ssds.pytorch","last_synced_at":"2025-05-12T03:30:37.224Z","repository":{"id":49946139,"uuid":"119862040","full_name":"ShuangXieIrene/ssds.pytorch","owner":"ShuangXieIrene","description":"Repository for Single Shot MultiBox Detector and its variants, implemented with pytorch, python3.","archived":false,"fork":false,"pushed_at":"2020-08-18T00:49:25.000Z","size":4259,"stargazers_count":566,"open_issues_count":40,"forks_count":163,"subscribers_count":22,"default_branch":"master","last_synced_at":"2025-03-31T02:27:24.045Z","etag":null,"topics":["darknet","fssd","mobilenet","pytorch","rfb","ssd","vgg","yolo"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/ShuangXieIrene.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":"2018-02-01T16:27:31.000Z","updated_at":"2025-03-25T05:31:58.000Z","dependencies_parsed_at":"2022-08-12T20:50:44.772Z","dependency_job_id":null,"html_url":"https://github.com/ShuangXieIrene/ssds.pytorch","commit_stats":null,"previous_names":[],"tags_count":1,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ShuangXieIrene%2Fssds.pytorch","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ShuangXieIrene%2Fssds.pytorch/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ShuangXieIrene%2Fssds.pytorch/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ShuangXieIrene%2Fssds.pytorch/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/ShuangXieIrene","download_url":"https://codeload.github.com/ShuangXieIrene/ssds.pytorch/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":253667930,"owners_count":21944941,"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":["darknet","fssd","mobilenet","pytorch","rfb","ssd","vgg","yolo"],"created_at":"2024-08-03T21:00:51.458Z","updated_at":"2025-05-12T03:30:36.100Z","avatar_url":"https://github.com/ShuangXieIrene.png","language":"Python","funding_links":[],"categories":["2.) Lightweight Structures"],"sub_categories":["**[Papers]**"],"readme":"\u003cdiv align=\"center\"\u003e\n  \u003cimg src=\"doc/source/_static/img/ssds.pytorch.png\" width=\"400\"/\u003e\n\u003c/div\u003e\n\n# ssds.pytorch\nRepository for Single Shot MultiBox Detector and its variants, implemented with pytorch, python3. This repo is easy to setup and has plenty of visualization methods. We hope this repo can help people have a better understanding for ssd-like model and help people train and deploy the ssds model easily.\n\nCurrently, it contains these features:\n- **Multiple SSD Variants**: ssd, fpn, bifpn, yolo and etc.\n- **Multiple Base Network**: resnet, regnet, mobilenet and etc.\n- **Visualize** the features of the ssd-like models to help the user understand the model design and performance.\n- **Fast Training and Inference**: Utilize Nvidia Apex and Dali to fast training and support the user convert the model to ONNX or TensorRT for deployment.\n\nThis repo is depended on the work of [ODTK](https://github.com/NVIDIA/retinanet-examples), [Detectron](https://github.com/facebookresearch/Detectron) and [Tensorflow Object Detection API](https://github.com/tensorflow/models/tree/master/research/object_detection). Thanks for their works.\n\n**Notice** The pretrain model for the current version does not finished yet, please check the [previous version](https://github.com/ShuangXieIrene/ssds.pytorch/tree/v0.3.1) for enrich pretrain models.\n\n### Table of Contents\n- \u003ca href='#installation'\u003eInstallation\u003c/a\u003e\n- \u003ca href='#usage'\u003eUsage\u003c/a\u003e\n- \u003ca href='#performance'\u003ePerformance and Model Zoo\u003c/a\u003e\n- \u003ca href='#visualization'\u003eVisualization\u003c/a\u003e\n- [Documentation](https://foreveryounggithub.github.io/ssds.doc)\n\n## Installation\n### requirements\n* python\u003e=3.7\n* CUDA\u003e=10.0\n* pytorch\u003e=1.4\n### basic installation:\n```bash\nconda install pytorch torchvision cudatoolkit=10.2 -c pytorch\ngit clone https://github.com/ShuangXieIrene/ssds.pytorch.git\ncd ssds.pytorch\npython setup.py clean -a install\n```\n#### extra python libs for parallel training\nCurrently, nvidia DALI and apex is not include in the requirements.txt and need to install manually.\n\n* [DALI](https://docs.nvidia.com/deeplearning/sdk/dali-developer-guide/docs/installation.html)\n```bash\npip install --extra-index-url https://developer.download.nvidia.com/compute/redist/cuda/10.0 nvidia-dali\n```\n* [apex](https://github.com/NVIDIA/apex#linux)\n```bash\ngit clone https://github.com/NVIDIA/apex\ncd apex\npip install -v --no-cache-dir --global-option=\"--cpp_ext\" --global-option=\"--cuda_ext\" ./\n```\n\n## Docker\n```bash\ngit clone https://github.com/ShuangXieIrene/ssds.pytorch.git\ndocker build -t ssds:local ./ssds.pytorch/\ndocker run --gpus all -it --rm -v /data:/data ssds:local\n```\n\n## Usage\n### 0. Check the config file by Visualization\nDefined the network in a [config file](experiments/cfgs/tests/test.yml) and tweak the config file based on the visualized anchor boxes\n```bash\npython -m ssds.utils.visualize -cfg experiments/cfgs/tests/test.yml\n```\n\n### 1. Training\n```bash\n# basic training\npython -m ssds.utils.train -cfg experiments/cfgs/tests/test.yml\n# parallel training\npython -m torch.distributed.launch --nproc_per_node={num_gpus} -m ssds.utils.train_ddp -cfg experiments/cfgs/tests/test.yml\n```\n\n### 2. Evaluation\n```bash\npython -m ssds.utils.train -cfg experiments/cfgs/tests/test.yml -e\n```\n\n### 3. Export to ONNX or TRT model\n```bash\npython -m ssds.utils.export -cfg experiments/cfgs/tests/test.yml -c best_mAP.pth -h\n```\n\n## Performance\n\n\n## Visualization","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FShuangXieIrene%2Fssds.pytorch","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FShuangXieIrene%2Fssds.pytorch","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FShuangXieIrene%2Fssds.pytorch/lists"}