{"id":25604722,"url":"https://github.com/hyouteki/malpractice","last_synced_at":"2025-11-11T09:31:11.640Z","repository":{"id":277285199,"uuid":"879655372","full_name":"hyouteki/malpractice","owner":"hyouteki","description":"A lightweight C library for creating, training, and evaluating simple neural networks","archived":false,"fork":false,"pushed_at":"2024-12-10T08:11:13.000Z","size":45,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"master","last_synced_at":"2025-02-13T06:01:43.267Z","etag":null,"topics":["c","library","neural-network"],"latest_commit_sha":null,"homepage":"","language":"C","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/hyouteki.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":"2024-10-28T10:10:46.000Z","updated_at":"2024-12-19T09:49:26.000Z","dependencies_parsed_at":"2025-02-13T06:01:46.080Z","dependency_job_id":"72d94dfe-54c4-42be-954b-7adaf6f28258","html_url":"https://github.com/hyouteki/malpractice","commit_stats":null,"previous_names":["hyouteki/malpractice"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hyouteki%2Fmalpractice","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hyouteki%2Fmalpractice/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hyouteki%2Fmalpractice/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hyouteki%2Fmalpractice/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/hyouteki","download_url":"https://codeload.github.com/hyouteki/malpractice/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":240063610,"owners_count":19742223,"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":["c","library","neural-network"],"created_at":"2025-02-21T17:50:03.661Z","updated_at":"2025-11-11T09:31:11.634Z","avatar_url":"https://github.com/hyouteki.png","language":"C","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003e A lightweight C library for creating, training, and evaluating simple neural networks with customizable initialization methods.\r\n\u003e This library supports basic forward and backward propagation, sigmoid activation, data handling, and model saving/loading.\r\n---\r\n\r\n## Features\r\n1. **Custom Initialization Techniques**: Supports zero, random, and Xavier initialization.\r\n2. **Forward and Backward Propagation**: Implements core functions for feedforward and backpropagation.\r\n3. **Data Handling**: Functions to initialize, normalize, partition, and describe data.\r\n4. **Model Serialization**: Save and load model configurations to/from files.\r\n\r\n## Usage\r\n1. **Model Initialization**: Set input, hidden, and output sizes with your preferred initialization method.\r\n2. **Training and Testing**: Use train and test functions with Data and Parameters structures to train the model.\r\n3. **Model Persistence**: Save and load models using save_model and load_model.\r\n\r\n## API Overview\r\n### Core Structures\r\n1. **Model**: Represents the neural network with layer sizes and weight initialization.\r\n2. **Data**: Holds data samples and labels for training/testing.\r\n3. **Parameters**: Configurable training parameters.\r\n### Key Functions\r\n1. **Model Initialization**: `initialize_model()`\r\n2. **Forward and Backward Propagation**: `forward()`, `backward()`\r\n3. **Data Manipulation**: `normalize_data()`, `partition_data()`, `describe_data()`\r\n4. **Model Persistence**: `save_model()`, `load_model()`\r\n\r\n## Getting Started\r\n```bash\r\ngit clone https://github.com/hyouteki/malpractice --recursive --depth=1\r\ncd malpractice\r\n./build.sh\r\n```\r\n\r\n## Example\r\n```c\r\n#include \"malpractice.h\"\r\n\r\nint main() {\r\n    // Model parameters\r\n    size_t input_size = 784, hidden_size = 128, output_size = 10;\r\n    Model_InitTechnique init_tech = Model_Init_Xavier;\r\n    Model *model = initialize_model(input_size, hidden_size, output_size, init_tech);\r\n\r\n    // Load data (replace with actual loading logic)\r\n    Data *data = zero_initialize_data(input_size, 100);\r\n\r\n    // Set training parameters\r\n    Parameters params = {.learning_rate = 0.01, .epochs = 1000, .log_train_metrics = 1};\r\n\r\n    // Train and evaluate\r\n    train(data, params, model);\r\n    test(data, model);\r\n\r\n    // Save model\r\n    save_model(model, \"model.bin\");\r\n\r\n    // Cleanup\r\n    deinitialize_data(data);\r\n    deinitialize_model(model);\r\n    return 0;\r\n}\r\n```\r\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhyouteki%2Fmalpractice","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fhyouteki%2Fmalpractice","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhyouteki%2Fmalpractice/lists"}