https://github.com/dunningkrueg/machine-learning-2.0
hmzmmzmzmzmz
https://github.com/dunningkrueg/machine-learning-2.0
analysis clang cpp machine-learning project
Last synced: 2 days ago
JSON representation
hmzmmzmzmzmz
- Host: GitHub
- URL: https://github.com/dunningkrueg/machine-learning-2.0
- Owner: dunningkrueg
- License: mit
- Created: 2025-01-30T15:31:56.000Z (5 months ago)
- Default Branch: jane
- Last Pushed: 2025-02-11T07:35:38.000Z (4 months ago)
- Last Synced: 2025-06-12T15:05:20.562Z (2 days ago)
- Topics: analysis, clang, cpp, machine-learning, project
- Language: C++
- Homepage:
- Size: 72.3 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Machine Learning Framework
A C++ Machine Learning Framework with "modern" implementations of Decision Trees and Neural Networks.
## Features
### Core Models
- **Decision Tree**
- Advanced Gini impurity implementation
- Feature importance analysis
- Efficient model serialization
- Optimized splitting algorithm
- **Neural Network**
- Multi-layer perceptron architecture
- Sigmoid activation function
- Backpropagation with momentum
- Early stopping mechanism
- Batch processing support
### Utilities
- **Cross Validation**
- K-Fold implementation
- Stratified sampling
- Performance metrics
- **Data Processing**
- CSV data loading
- Automatic normalization
- Train/test splitting
- Data validation
- **Visualization**
- Learning curve plotting
- Model performance analysis
- Feature importance visualization
## Requirements
- C++17 compatible compiler
- CMake 3.14 or higher
- Visual Studio 2019/2022 (Windows) or GCC/Clang (Linux/Mac)
## Quick Start
### Building the Project
```bash
mkdir build
cd build
cmake ..
cmake --build .
```
### Running the Example
```bash
cd build/Debug # or Release
./MLExample
```
### Basic Usage Example
```cpp
#include "MLFramework.hpp"
int main() {
// Load and prepare data
ml::Dataset dataset;
dataset.loadFromCSV("data.csv");
dataset.normalize();
// Split dataset
auto split = dataset.split(0.8);
// Create and train model
ml::NeuralNetwork model({split.trainFeatures[0].size(), 10, 5, 1});
model.fit(split.trainFeatures, split.trainLabels);
// Make predictions
auto predictions = model.predict(split.testFeatures);
// Evaluate performance
double mse = ml::metrics::meanSquaredError(predictions, split.testLabels);
std::cout << "Model MSE: " << mse << std::endl;
return 0;
}
```
### Contributing
1. Fork the repository
2. Create feature branch (`git checkout -b feature/AmazingFeature`)
3. Commit changes (`git commit -m 'Add AmazingFeature'`)
4. Push to branch (`git push origin feature/AmazingFeature`)
5. Open Pull Request
### License
This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.