{"id":20490359,"url":"https://github.com/hanzopgp/deeplearninglib","last_synced_at":"2025-07-20T01:35:57.810Z","repository":{"id":109491308,"uuid":"468281575","full_name":"hanzopgp/DeepLearningLib","owner":"hanzopgp","description":"Deep learning library from scratch (MLP/CNN/Auto-Encoder etc...). Contains some experiences on MNIST/CIPHAR datasets + report.","archived":false,"fork":false,"pushed_at":"2022-05-25T23:24:36.000Z","size":8213,"stargazers_count":2,"open_issues_count":0,"forks_count":0,"subscribers_count":2,"default_branch":"main","last_synced_at":"2025-03-05T17:23:56.075Z","etag":null,"topics":["auto-encoder","convolutional-neural-networks","deep-learning","experiments","library","multi-layer-perceptron","university"],"latest_commit_sha":null,"homepage":"","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/hanzopgp.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":"2022-03-10T09:47:34.000Z","updated_at":"2023-02-16T10:34:46.000Z","dependencies_parsed_at":null,"dependency_job_id":"998905f2-095b-416b-9860-1cd1c47f2d50","html_url":"https://github.com/hanzopgp/DeepLearningLib","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/hanzopgp/DeepLearningLib","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hanzopgp%2FDeepLearningLib","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hanzopgp%2FDeepLearningLib/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hanzopgp%2FDeepLearningLib/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hanzopgp%2FDeepLearningLib/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/hanzopgp","download_url":"https://codeload.github.com/hanzopgp/DeepLearningLib/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hanzopgp%2FDeepLearningLib/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":266053885,"owners_count":23869499,"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":["auto-encoder","convolutional-neural-networks","deep-learning","experiments","library","multi-layer-perceptron","university"],"created_at":"2024-11-15T17:16:46.816Z","updated_at":"2025-07-20T01:35:57.747Z","avatar_url":"https://github.com/hanzopgp.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# DeepLearningLib\n\n## Presentation\n\nLibrairies for deep learning (MLP/CNN/Auto-Encoder etc...).\n\n## Project architecture\n\n\u003cpre\u003e\u003ccode\u003eDeepLearningLib/\n      ├── nndiy/                   \n      |    ├── __init__.py        (Contains the sequential object and global variables)\n      |    ├── activation.py      (Contains the activation functions such as ReLU/Tanh...)\n      |    ├── core.py            (Contains the abstract classes)\n      |    ├── early_stopping.py  (Contains the early stopping objects)\n      |    ├── layer.py           (Contains the layers object such as Linear/Dropout...)\n      |    ├── loss.py            (Contains the loss objects such as MSE/BCE...)\n      |    ├── optimizer.py       (Contains the optimizer objects such as SGD/ADAM...)\n      |    └── utils.py           (Contains the utility functions such as min_max_scale/one_hot...)\n      ├── cnn_demo.py             (Contains the demo for CNN)\n      ├── experiences.py          (Contains the MLP/AE/CNN experiences)\n      ├── mlp_unit_test.py        (Contains MLP unit tests on simple problems) \n      ├── report/                 (Folder containing the images and report)\n      |    ├── img_report/\n      |    └── report.pdf         \n      ├── README.md\t\t          \n      └── LICENSE  \n\u003c/pre\u003e\u003c/code\u003e\n\n## Features implemented\n\n- Linear/Convo1D/MaxPool1D/AvgPool1D/Flatten/Dropout layers\n- GD/MGD/SGD/ADAM optimizers\n- LearkyReLU/ReLU/Identity/Tanh/Sigmoid/Softmax activation functions\n- MAE/MSE/RMSE/BCE/SBCE/CCE/SCCE/SCCESoftmax loss functions\n- Uniform/Xavier initialization\n- L1/L2 regularisation\n- EarlyStopping callback\n\n## Experiences\n\nAll those experiments were done on MNIST digits and fashion datasets :\n- Multi layer perceptron image classification\n- Autoencoder image reconstruction (with different latent space dimensions)\n- Autoencoder removing noise (with different percentage of noise)\n- Multi layer perceptron image classification with latent space representation (using different dimension)\n- SGD/ADAM/Tanh/ReLU benchmarks on MNIST\n- 1D CNN on MNIST\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhanzopgp%2Fdeeplearninglib","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fhanzopgp%2Fdeeplearninglib","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhanzopgp%2Fdeeplearninglib/lists"}