{"id":13443476,"url":"https://github.com/jfzhang95/pytorch-deeplab-xception","last_synced_at":"2025-04-11T06:21:35.558Z","repository":{"id":38203506,"uuid":"137429940","full_name":"jfzhang95/pytorch-deeplab-xception","owner":"jfzhang95","description":"DeepLab v3+ model in PyTorch. Support different backbones.","archived":false,"fork":false,"pushed_at":"2024-08-04T04:22:21.000Z","size":939,"stargazers_count":2951,"open_issues_count":134,"forks_count":781,"subscribers_count":43,"default_branch":"master","last_synced_at":"2025-04-03T07:08:22.947Z","etag":null,"topics":["deeplab-v3-plus","drn","mobilenetv2","pytorch","resnet","xception"],"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/jfzhang95.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,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2018-06-15T02:07:37.000Z","updated_at":"2025-04-01T12:08:38.000Z","dependencies_parsed_at":"2025-01-15T20:14:54.048Z","dependency_job_id":"91bba6ba-736d-47a4-be92-3401d7aa5314","html_url":"https://github.com/jfzhang95/pytorch-deeplab-xception","commit_stats":{"total_commits":63,"total_committers":3,"mean_commits":21.0,"dds":0.09523809523809523,"last_synced_commit":"9135e104a7a51ea9effa9c6676a2fcffe6a6a2e6"},"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jfzhang95%2Fpytorch-deeplab-xception","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jfzhang95%2Fpytorch-deeplab-xception/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jfzhang95%2Fpytorch-deeplab-xception/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jfzhang95%2Fpytorch-deeplab-xception/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/jfzhang95","download_url":"https://codeload.github.com/jfzhang95/pytorch-deeplab-xception/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248352458,"owners_count":21089452,"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":["deeplab-v3-plus","drn","mobilenetv2","pytorch","resnet","xception"],"created_at":"2024-07-31T03:02:01.449Z","updated_at":"2025-04-11T06:21:35.536Z","avatar_url":"https://github.com/jfzhang95.png","language":"Python","funding_links":[],"categories":["Python","Background subtraction"],"sub_categories":[],"readme":"# pytorch-deeplab-xception\n\n**Update on 2018/12/06. Provide model trained on VOC and SBD datasets.**  \n\n**Update on 2018/11/24. Release newest version code, which fix some previous issues and also add support for new backbones and multi-gpu training. For previous code, please see in `previous` branch**  \n\n### TODO\n- [x] Support different backbones\n- [x] Support VOC, SBD, Cityscapes and COCO datasets\n- [x] Multi-GPU training\n\n\n\n| Backbone  | train/eval os  |mIoU in val |Pretrained Model|\n| :-------- | :------------: |:---------: |:--------------:|\n| ResNet    | 16/16          | 78.43%     | [google drive](https://drive.google.com/open?id=1NwcwlWqA-0HqAPk3dSNNPipGMF0iS0Zu) |\n| MobileNet | 16/16          | 70.81%     | [google drive](https://drive.google.com/open?id=1G9mWafUAj09P4KvGSRVzIsV_U5OqFLdt) |\n| DRN       | 16/16          | 78.87%     | [google drive](https://drive.google.com/open?id=131gZN_dKEXO79NknIQazPJ-4UmRrZAfI) |\n\n\n\n### Introduction\nThis is a PyTorch(0.4.1) implementation of [DeepLab-V3-Plus](https://arxiv.org/pdf/1802.02611). It\ncan use Modified Aligned Xception and ResNet as backbone. Currently, we train DeepLab V3 Plus\nusing Pascal VOC 2012, SBD and Cityscapes datasets.\n\n![Results](doc/results.png)\n\n\n### Installation\nThe code was tested with Anaconda and Python 3.6. After installing the Anaconda environment:\n\n0. Clone the repo:\n    ```Shell\n    git clone https://github.com/jfzhang95/pytorch-deeplab-xception.git\n    cd pytorch-deeplab-xception\n    ```\n\n1. Install dependencies:\n\n    For PyTorch dependency, see [pytorch.org](https://pytorch.org/) for more details.\n\n    For custom dependencies:\n    ```Shell\n    pip install matplotlib pillow tensorboardX tqdm\n    ```\n### Training\nFollow steps below to train your model:\n\n0. Configure your dataset path in [mypath.py](https://github.com/jfzhang95/pytorch-deeplab-xception/blob/master/mypath.py).\n\n1. Input arguments: (see full input arguments via python train.py --help):\n    ```Shell\n    usage: train.py [-h] [--backbone {resnet,xception,drn,mobilenet}]\n                [--out-stride OUT_STRIDE] [--dataset {pascal,coco,cityscapes}]\n                [--use-sbd] [--workers N] [--base-size BASE_SIZE]\n                [--crop-size CROP_SIZE] [--sync-bn SYNC_BN]\n                [--freeze-bn FREEZE_BN] [--loss-type {ce,focal}] [--epochs N]\n                [--start_epoch N] [--batch-size N] [--test-batch-size N]\n                [--use-balanced-weights] [--lr LR]\n                [--lr-scheduler {poly,step,cos}] [--momentum M]\n                [--weight-decay M] [--nesterov] [--no-cuda]\n                [--gpu-ids GPU_IDS] [--seed S] [--resume RESUME]\n                [--checkname CHECKNAME] [--ft] [--eval-interval EVAL_INTERVAL]\n                [--no-val]\n\n    ```\n\n2. To train deeplabv3+ using Pascal VOC dataset and ResNet as backbone:\n    ```Shell\n    bash train_voc.sh\n    ```\n3. To train deeplabv3+ using COCO dataset and ResNet as backbone:\n    ```Shell\n    bash train_coco.sh\n    ```    \n\n### Acknowledgement\n[PyTorch-Encoding](https://github.com/zhanghang1989/PyTorch-Encoding)\n\n[Synchronized-BatchNorm-PyTorch](https://github.com/vacancy/Synchronized-BatchNorm-PyTorch)\n\n[drn](https://github.com/fyu/drn)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjfzhang95%2Fpytorch-deeplab-xception","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fjfzhang95%2Fpytorch-deeplab-xception","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjfzhang95%2Fpytorch-deeplab-xception/lists"}