{"id":28364072,"url":"https://github.com/raphaeldenni/mninst-training-app","last_synced_at":"2025-06-24T00:30:56.558Z","repository":{"id":294548544,"uuid":"983347478","full_name":"raphaeldenni/mninst-training-app","owner":"raphaeldenni","description":"This is a simple digit classifier using the MNIST dataset. It uses a convolutional or dense neural network to classify the digits.","archived":false,"fork":false,"pushed_at":"2025-05-20T23:27:51.000Z","size":1441,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-06-04T22:52:37.908Z","etag":null,"topics":["ai","mnist","mnist-classification","python"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":false,"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/raphaeldenni.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,"zenodo":null}},"created_at":"2025-05-14T08:47:33.000Z","updated_at":"2025-05-20T23:28:23.000Z","dependencies_parsed_at":"2025-05-21T00:40:50.639Z","dependency_job_id":null,"html_url":"https://github.com/raphaeldenni/mninst-training-app","commit_stats":null,"previous_names":["raphaeldenni/mninst-training-app"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/raphaeldenni/mninst-training-app","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/raphaeldenni%2Fmninst-training-app","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/raphaeldenni%2Fmninst-training-app/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/raphaeldenni%2Fmninst-training-app/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/raphaeldenni%2Fmninst-training-app/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/raphaeldenni","download_url":"https://codeload.github.com/raphaeldenni/mninst-training-app/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/raphaeldenni%2Fmninst-training-app/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":261582493,"owners_count":23180602,"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":["ai","mnist","mnist-classification","python"],"created_at":"2025-05-28T20:14:16.237Z","updated_at":"2025-06-24T00:30:56.552Z","avatar_url":"https://github.com/raphaeldenni.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Digit MNIST Classifier\n\nThis is a simple digit classifier using the MNIST dataset. It uses a convolutional or dense neural network to classify the digits.\n\n## Installation\n\nIt is recommended to use UV package manager to run this project. You can install it by following the instructions [here](https://docs.astral.sh/uv/getting-started/installation/).\nAfter installing UV, you can go to the [Usage](#usage) section of this README to run the project.\n\nOr you can install the required packages manually by using pyenv and pip:\n```\npyenv install 3.12\npyenv local 3.12\npip install -r requirements.txt\n```\n\n## Usage\n\n### Training\n\nTo train the model, run the following command with UV (recommended):\n\n```\nuv run model_trainer.py\n```\n\nOr with Python 3:\n```\npython3 model_trainer.py\n```\n\nThis will train the model and save it to a file called `model.keras`.\nYou can also specify the following arguments:\n\n- `model`: The model to use. Can be either `conv` or `dense`. Default is `conv`.\n- `epochs`: The number of epochs to train for. Default is 20.\n- `learning rate`: The learning rate. Default is 0.0003/0.0005 for convolutional/dense models.\n\n### Testing\n\nTo test the model, run the following command with UV (recommended):\n\n```\nuv run model_usage.py\n```\n\nOr with Python 3:\n```\npython3 model_usage.py\n```\n\n## Best Hyperparameters\n\nBest values for convolutional model:\n- 20 epochs (stop at 19)\n- 0.0003 learning rate\n\nBest values for dense model:\n- 20 epochs (stop at 17)\n- 0.0005 learning rate\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fraphaeldenni%2Fmninst-training-app","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fraphaeldenni%2Fmninst-training-app","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fraphaeldenni%2Fmninst-training-app/lists"}