{"id":19243195,"url":"https://github.com/acecoooool/lednet","last_synced_at":"2025-07-14T21:33:35.840Z","repository":{"id":201607385,"uuid":"186776570","full_name":"AceCoooool/LEDNet","owner":"AceCoooool","description":"This is an unofficial implemention of LEDNet https://arxiv.org/abs/1905.02423","archived":false,"fork":false,"pushed_at":"2019-05-25T02:28:24.000Z","size":13369,"stargazers_count":38,"open_issues_count":7,"forks_count":10,"subscribers_count":4,"default_branch":"master","last_synced_at":"2025-04-21T09:52:05.458Z","etag":null,"topics":["lightning-network","pytorch","segmentation"],"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/AceCoooool.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}},"created_at":"2019-05-15T07:50:01.000Z","updated_at":"2023-11-20T08:15:08.000Z","dependencies_parsed_at":null,"dependency_job_id":"b41eb9b1-d789-40e7-bcee-883cd6eb2943","html_url":"https://github.com/AceCoooool/LEDNet","commit_stats":null,"previous_names":["acecoooool/lednet"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/AceCoooool/LEDNet","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/AceCoooool%2FLEDNet","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/AceCoooool%2FLEDNet/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/AceCoooool%2FLEDNet/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/AceCoooool%2FLEDNet/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/AceCoooool","download_url":"https://codeload.github.com/AceCoooool/LEDNet/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/AceCoooool%2FLEDNet/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":265352229,"owners_count":23751883,"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":["lightning-network","pytorch","segmentation"],"created_at":"2024-11-09T17:16:57.446Z","updated_at":"2025-07-14T21:33:35.796Z","avatar_url":"https://github.com/AceCoooool.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# LEDNet\nThis is an unofficial implement of  [LEDNet](https://arxiv.org/abs/1905.02423). \n\n\u003e the official version：[LEDNet-official](https://github.com/xiaoyufenfei/LEDNet)\n\n\u003cdiv align=\"center\"\u003e\n\u003cimg src=\"png/demo.png\" width=\"250px\"\u003e\u003cimg src=\"png/gt.png\" width=\"250px\"\u003e\u003cimg src=\"png/output.png\" width=\"250px\"\u003e\n\u003c/div\u003e\n\n## Environment\n\n- Python 3.6\n- PyTorch 1.1\n\n## Performance\n\n- Base Size 1024, Crop Size 768, only fine. (new-version, with dropout)\n\n| Model  | Paper | OHEM | Drop-rate |   lr   | Epoch | val (crop)  |                             val                              |\n| :----: | :---: | :--: | :-------: | :----: | :---: | :---------: | :----------------------------------------------------------: |\n| LEDNet |   /   |  ✗   |    0.1    | 0.0005 |  800  | 60.32/94.51 |                         66.29/94.40                          |\n| LEDNet |   /   |  ✗   |    0.1    | 0.005  |  600  | 61.29/94.75 |                         66.56/94.72                          |\n| LEDNet |   /   |  ✗   |    0.3    |  0.01  |  800  | 63.84/94.83 | [69.09/94.75](https://drive.google.com/open?id=1oelPUKAnZYD75RruyBQU9HZKneMEMIAp) |\n\n\u003e Note:\n\u003e\n\u003e - The paper only provide the test results: 69.2/86.8 (class mIoU/category mIoU). \n\u003e - And the training setting is a little different with original paper (original paper use 1024x512)\n\nSome things you can use to improve the performance:\n\n1. use larger learning rate (like 0.01)\n2. use more epochs (like 1000)\n3. use larger training input size (like Base Size 1344, Crop Size 1024)\n\n## Demo\n\nPlease download [pretrained](https://drive.google.com/open?id=1oelPUKAnZYD75RruyBQU9HZKneMEMIAp) model first\n\n```shell\n$ python demo.py [--input-pic png/demo.png] [--pretrained your-root-of-pretrained] [--cuda true]\n```\n\n## Evaluation\n\nThe default data root is `~/.torch/datasets` (You can download dataset and build a soft-link to it)\n\n```shell\n$ python eval.py [--mode testval] [--pretrained root-of-pretrained-model] [--cuda true]\n```\n\n## Training\n\nRecommend to using distributed training.\n\n```shell\n$ export NGPUS=4\n$ python -m torch.distributed.launch --nproc_per_node=$NGPUS train.py [--dataset citys] [--batch-size 8] [--base-size 1024] [--crop-size 768] [--epochs 800] [--warmup-factor 0.1] [--warmup-iters 200] [--log-step 10] [--save-epoch 40] [--lr 0.005]\n```\n\n## Prepare data\n\nYour can reference [gluon-cv-cityspaces](https://gluon-cv.mxnet.io/build/examples_datasets/cityscapes.html#sphx-glr-build-examples-datasets-cityscapes-py) to prepare the dataset\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Facecoooool%2Flednet","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Facecoooool%2Flednet","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Facecoooool%2Flednet/lists"}