{"id":13499145,"url":"https://github.com/leaderj1001/MobileNetV3-Pytorch","last_synced_at":"2025-03-29T04:30:52.045Z","repository":{"id":48173753,"uuid":"185542801","full_name":"leaderj1001/MobileNetV3-Pytorch","owner":"leaderj1001","description":"Implementing Searching for MobileNetV3 paper using Pytorch","archived":false,"fork":false,"pushed_at":"2019-07-04T00:41:50.000Z","size":70443,"stargazers_count":290,"open_issues_count":7,"forks_count":71,"subscribers_count":9,"default_branch":"master","last_synced_at":"2024-08-01T22:50:11.388Z","etag":null,"topics":["cifar-10","cifar-100","imagenet","mobilenetv3","pytorch","searching-for-mobilenetv3"],"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/leaderj1001.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":"2019-05-08T06:18:39.000Z","updated_at":"2024-07-05T20:39:44.000Z","dependencies_parsed_at":"2022-09-10T08:10:30.299Z","dependency_job_id":null,"html_url":"https://github.com/leaderj1001/MobileNetV3-Pytorch","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/leaderj1001%2FMobileNetV3-Pytorch","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/leaderj1001%2FMobileNetV3-Pytorch/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/leaderj1001%2FMobileNetV3-Pytorch/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/leaderj1001%2FMobileNetV3-Pytorch/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/leaderj1001","download_url":"https://codeload.github.com/leaderj1001/MobileNetV3-Pytorch/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":222456009,"owners_count":16987582,"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":["cifar-10","cifar-100","imagenet","mobilenetv3","pytorch","searching-for-mobilenetv3"],"created_at":"2024-07-31T22:00:29.900Z","updated_at":"2024-10-31T17:31:41.914Z","avatar_url":"https://github.com/leaderj1001.png","language":"Python","funding_links":[],"categories":["Papers\u0026Codes","DLA","2.) Lightweight Structures"],"sub_categories":["MobileNetV3","**[Papers]**"],"readme":"# Implementing Searching for MobileNetV3 paper using Pytorch\n- The current model is a very early model. I will modify it as a general model as soon as possible.\n## Paper\n- [Searching for MobileNetV3 paper](https://arxiv.org/abs/1905.02244)\n- Author: Andrew Howard(Google Research), Mark Sandler(Google Research, Grace Chu(Google Research), Liang-Chieh Chen(Google Research), Bo Chen(Google Research), Mingxing Tan(Google Brain), Weijun Wang(Google Research), Yukun Zhu(Google Research), Ruoming Pang(Google Brain), Vijay Vasudevan(Google Brain), Quoc V. Le(Google Brain), Hartwig Adam(Google Research)\n\n## Todo\n- Experimental need for ImageNet dataset.\n- Code refactoring\n\n## MobileNetV3 Block\n![캡처](https://user-images.githubusercontent.com/22078438/57360577-6f30d000-71b5-11e9-89a6-24034a3ecdde.PNG)\n\n## Experiments\n- For CIFAR-100 data, I experimented with resize (224, 224).\u003cbr\u003e\n\n| Datasets | Model | acc1 | acc5 | Epoch  | Parameters\n| :---: | :---: | :---: | :---: | :---: | :---: |\nCIFAR-100 | MobileNetV3(LARGE) | 70.44% | 91.34% | 80 | 3.99M\nCIFAR-100 | MobileNetV3(SMALL) | 67.04% | 89.41% | 55 | 1.7M\nIMAGENET | MobileNetV3(LARGE) WORK IN PROCESS | | | | 5.15M\nIMAGENET | MobileNetV3(SMALL) WORK IN PROCESS | | | | 2.94M\n\n## Usage\n\n### Train\n```\npython main.py\n```\n- If you want to change hyper-parameters, you can check \"python main.py --help\"\n\nOptions:\n- `--dataset-mode` (str) - which dataset you use, (example: CIFAR10, CIFAR100), (default: CIFAR100).\n- `--epochs` (int) - number of epochs, (default: 100).\n- `--batch-size` (int) - batch size, (default: 128).\n- `--learning-rate` (float) - learning rate, (default: 1e-1).\n- `--dropout` (float) - dropout rate, (default: 0.3).\n- `--model-mode` (str) - which network you use, (example: LARGE, SMALL), (default: LARGE).\n- `--load-pretrained` (bool) - (default: False).\n- `--evaluate` (bool) - Used when testing. (default: False).\n- `--multiplier` (float) - (default: 1.0).\n\n### Test\n```\npython main.py --evaluate True\n```\n- Put the saved model file in the checkpoint folder and saved graph file in the saved_graph folder and type \"python main.py --evaluate True\".\n- If you want to change hyper-parameters, you can check \"python test.py --help\"\n\nOptions:\n- `--dataset-mode` (str) - which dataset you use, (example: CIFAR10, CIFAR100), (default: CIFAR100).\n- `--epochs` (int) - number of epochs, (default: 100).\n- `--batch-size` (int) - batch size, (default: 128).\n- `--learning-rate` (float) - learning rate, (default: 1e-1).\n- `--dropout` (float) - dropout rate, (default: 0.3).\n- `--model-mode` (str) - which network you use, (example: LARGE, SMALL), (default: LARGE).\n- `--load-pretrained` (bool) - (default: False).\n- `--evaluate` (bool) - Used when testing. (default: False).\n- `--multiplier` (float) - (default: 1.0).\n\n### Number of Parameters\n```python\nimport torch\n\nfrom model import MobileNetV3\n\ndef get_model_parameters(model):\n    total_parameters = 0\n    for layer in list(model.parameters()):\n        layer_parameter = 1\n        for l in list(layer.size()):\n            layer_parameter *= l\n        total_parameters += layer_parameter\n    return total_parameters\n\ntmp = torch.randn((128, 3, 224, 224))\nmodel = MobileNetV3(model_mode=\"LARGE\", multiplier=1.0)\nprint(\"Number of model parameters: \", get_model_parameters(model))\n```\n\n## Requirements\n- torch==1.0.1\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fleaderj1001%2FMobileNetV3-Pytorch","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fleaderj1001%2FMobileNetV3-Pytorch","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fleaderj1001%2FMobileNetV3-Pytorch/lists"}