{"id":17695589,"url":"https://github.com/sisolieri/cnn-cifar10-classification","last_synced_at":"2026-05-03T01:33:17.233Z","repository":{"id":258270133,"uuid":"872664844","full_name":"sisolieri/CNN-CIFAR10-classification","owner":"sisolieri","description":" Classification of CIFAR-10 images using a CNN model as part of my Master's in Data Science. 12 experiments were conducted to improve accuracy, achieving 90.7% with data augmentation. Future work focuses on refining transfer learning techniques.","archived":false,"fork":false,"pushed_at":"2024-10-14T22:21:17.000Z","size":7390,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-07-23T16:49:29.514Z","etag":null,"topics":["cifar10","cnn-classification","convolutional-neural-networks","data-augmentation","deep-learning","image-classification","keras","machine-learning","python","tensorflow","transfer-learning"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","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/sisolieri.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":"2024-10-14T20:58:52.000Z","updated_at":"2025-03-31T09:18:51.000Z","dependencies_parsed_at":"2024-10-18T09:04:52.733Z","dependency_job_id":null,"html_url":"https://github.com/sisolieri/CNN-CIFAR10-classification","commit_stats":null,"previous_names":["sisolieri/cnn-cifar10-classification"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/sisolieri/CNN-CIFAR10-classification","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sisolieri%2FCNN-CIFAR10-classification","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sisolieri%2FCNN-CIFAR10-classification/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sisolieri%2FCNN-CIFAR10-classification/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sisolieri%2FCNN-CIFAR10-classification/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/sisolieri","download_url":"https://codeload.github.com/sisolieri/CNN-CIFAR10-classification/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sisolieri%2FCNN-CIFAR10-classification/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":274250724,"owners_count":25249440,"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":"2025-09-09T02:00:10.223Z","response_time":80,"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":["cifar10","cnn-classification","convolutional-neural-networks","data-augmentation","deep-learning","image-classification","keras","machine-learning","python","tensorflow","transfer-learning"],"created_at":"2024-10-24T14:06:08.466Z","updated_at":"2026-05-03T01:33:17.199Z","avatar_url":"https://github.com/sisolieri.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"![Python](https://img.shields.io/badge/Python-3776AB?style=for-the-badge\u0026logo=python\u0026logoColor=white)\n![Pandas](https://img.shields.io/badge/Pandas-150458?style=for-the-badge\u0026logo=pandas\u0026logoColor=white)\n![NumPy](https://img.shields.io/badge/NumPy-013243?style=for-the-badge\u0026logo=numpy\u0026logoColor=white)\n![Keras](https://img.shields.io/badge/Keras-D00000?style=for-the-badge\u0026logo=keras\u0026logoColor=white)\n![TensorFlow](https://img.shields.io/badge/TensorFlow-FF6F00?style=for-the-badge\u0026logo=tensorflow\u0026logoColor=white)\n![Scikit-learn](https://img.shields.io/badge/Scikit--Learn-F7931E?style=for-the-badge\u0026logo=scikit-learn\u0026logoColor=white)\n![Matplotlib](https://img.shields.io/badge/Matplotlib-3F4F75?style=for-the-badge\u0026logo=matplotlib\u0026logoColor=white)\n\n\n# CNN CIFAR-10 Image Classification\n\nThis project was developed as part of my Master's program in Data Science and Artificial Intelligence. It focuses on classifying images from the CIFAR-10 dataset using Convolutional Neural Networks (CNNs). The primary goal was to achieve at least 80% accuracy on the test set through architectural modifications alone. I conducted a total of 12 experiments, exceeding the minimum requirement of 10 different experiments.\n\n## Project Overview\n\nThe CIFAR-10 dataset consists of 60,000 32x32 color images across 10 classes. The goal of this project was to enhance the performance of a very basic CNN model provided by the professors, implementing 12 different experiments (in addition to the baseline model) to improve its accuracy.\n\n- **Baseline Accuracy**: 60.6% on the test set.\n- **Best Accuracy without Data Augmentation**: 87.7%.\n- **Best Overall Accuracy**: 90.7% using Data Augmentation.\n- **Transfer Learning**: The final experiment applied VGG-16, achieving lower accuracy than the augmented CNN. Future improvements will focus on better applying transfer learning.\n\n## Key Skills\n\n- **Deep Learning**: Implementation and optimization of CNN architectures.\n- **Model Optimization**: Techniques like dropout, batch normalization, and data augmentation.\n- **Data Augmentation**: Applied to improve model generalization and performance.\n- **Transfer Learning**: Initial exploration using VGG-16, with plans to improve this aspect in future work.\n- **Keras**: All models and experiments were implemented using the Keras library.\n\n## Repository Structure\n\n- `/notebooks`: Jupyter notebooks for each experiment conducted, detailing modifications and results.\n- `/docs`: Final project report (PDF) with detailed explanations of each experiment, results, and future developments.\n\n## Dataset\n\nThe CIFAR-10 dataset is available through [TensorFlow Datasets](https://www.tensorflow.org/datasets/catalog/cifar10) or can be downloaded from the [official page](https://www.cs.toronto.edu/~kriz/cifar.html).\n\n## Results and Future Work\n\nThe project successfully reached an accuracy of 90.7% using Data Augmentation. However, there is still room for improvement, especially in the area of **Transfer Learning**. The VGG-16 model used in the final experiment did not surpass the results obtained with Data Augmentation, and further study in this area could yield better results.\n\n---\n\n**Feel free to reach out for any questions or further discussions!**\n\n---\n\n## License\n\nThis project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsisolieri%2Fcnn-cifar10-classification","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsisolieri%2Fcnn-cifar10-classification","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsisolieri%2Fcnn-cifar10-classification/lists"}