{"id":23864715,"url":"https://github.com/omidghadami95/efficientnetv2_quantization_ck","last_synced_at":"2026-04-30T08:41:10.866Z","repository":{"id":181271236,"uuid":"666505640","full_name":"OmidGhadami95/EfficientNetV2_Quantization_CK","owner":"OmidGhadami95","description":"EfficientNetV2 (Efficientnetv2-b2) and quantization int8 and fp32 (QAT and PTQ) on CK+ dataset . fine-tuning, augmentation, solving imbalanced dataset, etc.","archived":false,"fork":false,"pushed_at":"2024-05-04T12:15:14.000Z","size":352,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-01-03T08:33:29.323Z","etag":null,"topics":["ckplus","efficientnet","efficientnetv2","efficientnetv2-b2","emotion-recognition","facial-emotion-recognition","googlecolab","imbalanced-dataset","keras","post-training-quantization","ptq","python","qat","quantization","quantization-aware-training","real-time-emotion-classification","real-time-emotion-detection","scale-down","tensorflow"],"latest_commit_sha":null,"homepage":"https://omidghadami95.github.io/homepage/","language":"Jupyter Notebook","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/OmidGhadami95.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,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2023-07-14T17:37:13.000Z","updated_at":"2024-12-28T02:20:01.000Z","dependencies_parsed_at":null,"dependency_job_id":"ee9a45aa-d19b-4a6e-b83a-fc96d712f40c","html_url":"https://github.com/OmidGhadami95/EfficientNetV2_Quantization_CK","commit_stats":null,"previous_names":["omidghadami95/efficientnetv2_quantization_ck"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/OmidGhadami95%2FEfficientNetV2_Quantization_CK","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/OmidGhadami95%2FEfficientNetV2_Quantization_CK/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/OmidGhadami95%2FEfficientNetV2_Quantization_CK/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/OmidGhadami95%2FEfficientNetV2_Quantization_CK/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/OmidGhadami95","download_url":"https://codeload.github.com/OmidGhadami95/EfficientNetV2_Quantization_CK/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":240178366,"owners_count":19760565,"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":["ckplus","efficientnet","efficientnetv2","efficientnetv2-b2","emotion-recognition","facial-emotion-recognition","googlecolab","imbalanced-dataset","keras","post-training-quantization","ptq","python","qat","quantization","quantization-aware-training","real-time-emotion-classification","real-time-emotion-detection","scale-down","tensorflow"],"created_at":"2025-01-03T08:30:47.430Z","updated_at":"2026-04-30T08:41:05.846Z","avatar_url":"https://github.com/OmidGhadami95.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# EfficientNetV2_Quantization_CKplus tensorflow keras\n EfficientNetV2 (Efficientnetv2-b2) and quantization int8 and fp32 (QAT and PTQ) on CK+ dataset . fine-tuning, augmentation, solving imbalanced dataset and so on.\n\n\u003ca href=\"https://ibb.co/89CXnsW\"\u003e\u003cimg src=\"https://i.ibb.co/sqftpR0/Screenshot-3-8-2024-10-24-25-PM.png\" alt=\"Screenshot-3-8-2024-10-24-25-PM\" border=\"0\" /\u003e\u003c/a\u003e\n\nReal-time facial emotion recognition using EfficientNetV2 and quantization on CK+ dataset. This code includes:  \n1- data loading steps (download and split dataset).  \n2- preprocessing steps on CK+ dataset (normalization, resizing, augmentation and solving imbalanced dataset problem).  \n3- fine-tuning (using pre-trained weights from imagenet dataset as initial weights for training step).  \n4- quantization int8 and fp32 and fine-tuning after quantization ( Quantization-aware training integer8 (QAT) and Post-training quantization float32 (PTQ) ).  \n5- Macro, Micro, and Weighted for Precision, Recall, F1-score  \n6- Confusion Matrix\n\nNote that Quantization int8 has some benefits in reducing inference time and model size. But, Sometimes, it leads to a lower accuracy (PTQ). If we want to compensate for this loss, we need to use quantization-aware training approach. It means we need fine-tuning after quantization to compensate for lost accuracy. Finally, we compared int8 QAT and fp32 PTQ in terms of accuracy and model size, and inference time. \n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fomidghadami95%2Fefficientnetv2_quantization_ck","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fomidghadami95%2Fefficientnetv2_quantization_ck","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fomidghadami95%2Fefficientnetv2_quantization_ck/lists"}