{"id":31774390,"url":"https://github.com/monatis/efficientnet-tf2","last_synced_at":"2025-10-10T04:51:59.407Z","repository":{"id":36396592,"uuid":"201539750","full_name":"monatis/efficientnet-tf2","owner":"monatis","description":"A reusable implementation of EfficientNet in TensorFlow 2.0 and Keras","archived":false,"fork":false,"pushed_at":"2022-02-11T02:55:56.000Z","size":18,"stargazers_count":16,"open_issues_count":8,"forks_count":12,"subscribers_count":5,"default_branch":"master","last_synced_at":"2023-03-04T08:10:13.573Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/monatis.png","metadata":{"files":{"readme":"readme.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2019-08-09T21:00:34.000Z","updated_at":"2023-03-04T06:54:00.000Z","dependencies_parsed_at":"2022-08-08T14:30:57.715Z","dependency_job_id":null,"html_url":"https://github.com/monatis/efficientnet-tf2","commit_stats":null,"previous_names":[],"tags_count":null,"template":null,"template_full_name":null,"purl":"pkg:github/monatis/efficientnet-tf2","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/monatis%2Fefficientnet-tf2","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/monatis%2Fefficientnet-tf2/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/monatis%2Fefficientnet-tf2/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/monatis%2Fefficientnet-tf2/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/monatis","download_url":"https://codeload.github.com/monatis/efficientnet-tf2/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/monatis%2Fefficientnet-tf2/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":279002653,"owners_count":26083442,"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":"2025-10-10T02:00:06.843Z","response_time":62,"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":[],"created_at":"2025-10-10T04:51:49.751Z","updated_at":"2025-10-10T04:51:59.403Z","avatar_url":"https://github.com/monatis.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# efficientnet-tf2\nA TensorFlow 2.0 implementation of [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://arxiv.org/abs/1905.11946), aka EfficientNet.\n\n## Motivation\n\nEfficientNet  is still one of the most efficient architectures for image classification. Considering that TensorFlow 2.0 has already hit version beta1, I think that a flexible and reusable implementation of EfficientNet in TF 2.0 might be useful for practitioners.\n\n## Implementation\n\nI  implemented a running mean and standard deviation calculation with [Welford algorithm](https://www.johndcook.com/blog/standard_deviation/), which eliminates the problem of loading the whole dataset into the memory. `Normalizer` class, calculating the mean and standard deviation, is also used as a `preprocessing_function` argument to `tf.keras.preprocessing.image.ImageDataGenerator`.\n\n## Install\n\n1. `conda create -n effnet python=3.6.8`\n2. `conda activate effnet`\n3. `git clone https://github.com/monatis/effnet-tf2.git`\n4. `cd efficientnet-tf2`\n5. `python -m pip install -r requirements.gpu.txt` # Change to `requirements.cpu.txt` if you're not using GPU.\n\n## Usage\n\n`train_dir` and `validation_dir` directories should contain a subdirectory for each class in the dataset. Then run:\n\n- `python train.py --train_dir /path/to/training/images --validation_dir /path/to/validation/images`\n- See `model/` directory for training output.\n\nrun `python train.py --help` to see all the options.\n\n## Roadmap\n\n- [x] Share model architecture and a training script.\n- [x] Implement export to saved model.\n- [x] Implement command line arguments to configure data augmentation.\n- [ ] Share an inference script.\n- [x] Implement mean and STD normalization.\n- [ ] Implement confusion matrix.\n- [ ] Implement export to TFLite for model inference.\n- [ ] Share an example Android app using the exported TFLite model.\n\n## License\n\nMIT","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmonatis%2Fefficientnet-tf2","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmonatis%2Fefficientnet-tf2","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmonatis%2Fefficientnet-tf2/lists"}