{"id":22429846,"url":"https://github.com/techn0man1ac/machinelearningcv","last_synced_at":"2026-04-19T17:12:34.116Z","repository":{"id":266011101,"uuid":"897108693","full_name":"techn0man1ac/MachineLearningCV","owner":"techn0man1ac","description":"This project is a web-based machine learning application that allows training and testing CNN and VGG16 models based on the Fashion MNIST dataset. The application was created using Streamlit.","archived":false,"fork":false,"pushed_at":"2024-12-11T08:10:33.000Z","size":49211,"stargazers_count":0,"open_issues_count":4,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-03-27T07:13:42.262Z","etag":null,"topics":["cnn","fashion-mnist","machine-learning","mit-license","streamlit","vgg16","web"],"latest_commit_sha":null,"homepage":"https://mlcvzero.streamlit.app/","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/techn0man1ac.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":"2024-12-02T03:24:38.000Z","updated_at":"2024-12-11T08:10:37.000Z","dependencies_parsed_at":"2024-12-02T17:50:59.292Z","dependency_job_id":null,"html_url":"https://github.com/techn0man1ac/MachineLearningCV","commit_stats":null,"previous_names":["techn0man1ac/machinelearningcv"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/techn0man1ac%2FMachineLearningCV","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/techn0man1ac%2FMachineLearningCV/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/techn0man1ac%2FMachineLearningCV/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/techn0man1ac%2FMachineLearningCV/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/techn0man1ac","download_url":"https://codeload.github.com/techn0man1ac/MachineLearningCV/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":245798358,"owners_count":20673902,"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":["cnn","fashion-mnist","machine-learning","mit-license","streamlit","vgg16","web"],"created_at":"2024-12-05T21:05:56.916Z","updated_at":"2025-10-15T02:19:47.374Z","avatar_url":"https://github.com/techn0man1ac.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# ML Application with Streamlit\r\n\r\n![Model training screenshot](https://raw.githubusercontent.com/techn0man1ac/MachineLearningCV/refs/heads/main/screenshots/ModelTraining.png)\r\n\r\nThis project is a web-based machine learning application that allows training and testing CNN and VGG16 models based on the Fashion MNIST dataset. The application was created using [Streamlit](https://github.com/streamlit/streamlit).\r\n\r\n# 📋 Description\r\n\r\nThe application has the following functions:\r\n1. Training the CNN model on black and white images (Fashion MNIST).\r\n2. Training the VGG16 model on color images (converted from Fashion MNIST).\r\n3. Visualization of training results, including loss and accuracy graphs.\r\n4. Testing of images uploaded by the user based on the selected model.\r\n5. Use pre-trained models in the `Model test` tab(without training). Pre-trained models should be in the `saveModels` directory. Here my prepared models:\r\n\r\nhttps://drive.google.com/drive/folders/11ptCnpoiAlmtYGRsQOJaR54Vhexvcw34?usp=sharing\r\n\r\n# 🛠️ File structure\r\n\r\n- `main.py` The script to run the Streamlit application.\r\n- `app.py` The main module containing the application functionality, including model definition, training, visualization, and testing.\r\n\r\n# Running the application\r\n\r\n```bash\r\npython main.py\r\n```\r\n\r\nOr directly:\r\n\r\n```bash\r\nstreamlit run main.py\r\n```\r\n\r\n# 📊 Models\r\n\r\n## CNN:\r\n\r\nUses 2D convolutional layers.\r\nThe architecture includes Dropout to prevent overfitting.\r\nTrained on 28x28 grayscale images.\r\n\r\n## VGG16:\r\n\r\nPre-trained on ImageNet.\r\nInput images are scaled to 32x32 with three channels (RGB).\r\nOnly the VGG16 main unit (frozen) is used.\r\n\r\n# 🖼️ How to work with the application\r\n\r\nRunn app, click the `Model Test` tab and select the image you want to recognize. After selecting the image, it will be converted to the correct format and size for submitting the model for recognition. Two models(CNN/VGG16) are trained to recognize type of product(clothing item) such as bags, T-shirts, dresses... Based on a dataset [Fashion-MNIST](https://github.com/zalandoresearch/fashion-mnist).\r\n\r\n![Recognize image screenshot](https://raw.githubusercontent.com/techn0man1ac/MachineLearningCV/refs/heads/main/screenshots/RecognizeImage.png)\r\n\r\n1. Training models\r\nSelect `CNN` or `VGG16` in the menu.\r\nEnter the number of epochs and click “Train model”.\r\n2. Testing\r\nUpload an image in .jpg, .jpeg, .png, or .gif format.\r\nSelect a model (CNN or VGG16).\r\nClick `Recognize image`.\r\n3. Saving models\r\nAfter completing the training of the models app saved traine model in the `saveModels` folder in the `.keras` format.\r\n\r\n# 📂 Example result.\r\n\r\n## Training graphs\r\n\r\n![Traine statistics screenshot](https://raw.githubusercontent.com/techn0man1ac/MachineLearningCV/refs/heads/main/screenshots/TraineStatistics.png)\r\n\r\nGraph of loss and accuracy.\r\n\r\n## Image recognition\r\n\r\n![Shirt image recognition screenshot](https://raw.githubusercontent.com/techn0man1ac/MachineLearningCV/refs/heads/main/screenshots/Shirt.png)\r\n\r\nImage with class prediction and probability histogram.\r\n\r\n![Lable histogram](https://raw.githubusercontent.com/techn0man1ac/MachineLearningCV/refs/heads/main/screenshots/LableHistogram.png)\r\n\r\n## 📑 License\r\n\r\nThis project is licensed under the MIT License - see the [LICENSE](https://github.com/techn0man1ac/MachineLearningCV/blob/main/LICENSE) file for details. \r\n\r\nStreamlit software is also distributed under the [Apache-2.0 license](https://github.com/streamlit/streamlit?tab=Apache-2.0-1-ov-file).\r\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftechn0man1ac%2Fmachinelearningcv","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ftechn0man1ac%2Fmachinelearningcv","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftechn0man1ac%2Fmachinelearningcv/lists"}