MetaPerceptron (Metaheuristic-optimized Multi-Layer Perceptron) is a powerful and extensible Python library that brings the best of both worlds: metaheuristic optimization and deep learning via Multi-Layer Perceptron (MLP). Whether you're working with classic Gradient Descent techniques or state-of-the-art metaheuristic algorithms like GA, PSO, WOA, DE, etc., MetaPerceptron has you covered. With MetaPerceptron, you can perform searches, feature selection, and hyperparameter tuning using the features provided by the Scikit-Learn library.

🚀 Features at a Glance

• 🔧 Estimators: MlpRegressor, MlpClassifier, MhaMlpRegressor, MhaMlpClassifier
• 📊 Utilities: MhaMlpTuner, MhaMlpComparator
• 🧠 Model Zoo:
  • 200+ Metaheuristic-trained MLP Regressors
  • 200+ Metaheuristic-trained MLP Classifiers
  • 12 Gradient Descent-trained MLP Regressors
  • 12 Gradient Descent-trained MLP Classifiers
• 📏 67+ Performance Metrics (47 for regression, 20 for classification)
• ⚙️ Support: GPU support (for GD-based models), Scikit-learn compatible API
• 📚 Documentation: https://metaperceptron.readthedocs.io
• 🐍 Python: 3.8+
• 📦 Dependencies: numpy, scipy, scikit-learn, pytorch, mealpy, pandas, permetrics

📖 Citation

If MetaPerceptron supports your work, please consider citing the following:

```bibtex @article{van2025metaperceptron, title={MetaPerceptron: A Standardized Framework for Metaheuristic-Driven Multi-Layer Perceptron Optimization}, author={Van Thieu, Nguyen and Mirjalili, Seyedali and Garg, Harish and Hoang, Nguyen Thanh}, journal={Computer Standards & Interfaces}, pages={103977}, year={2025}, publisher={Elsevier}, doi={10.1016/j.csi.2025.103977}, url={https://doi.org/10.1016/j.csi.2025.103977} }

@article{van2023mealpy, title={MEALPY: An open-source library for latest meta-heuristic algorithms in Python}, author={Van Thieu, Nguyen and Mirjalili, Seyedali}, journal={Journal of Systems Architecture}, year={2023}, publisher={Elsevier}, doi={10.1016/j.sysarc.2023.102871} }

@article{van2023groundwater, title={Groundwater level modeling using Augmented Artificial Ecosystem Optimization}, author={Van Thieu, Nguyen and Barma, Surajit Deb and Van Lam, To and Kisi, Ozgur and Mahesha, Amai}, journal={Journal of Hydrology}, volume={617}, pages={129034}, year={2023}, publisher={Elsevier}, doi={10.1016/j.jhydrol.2022.129034} } ```

🧪 Quick Start

Install the latest version using pip:

bash pip install metaperceptron

After that, check the version to ensure successful installation:

python import metaperceptron print(metaperceptron.__version__)

✅ Import core components

Here is how you can import all provided classes from MetaPerceptron

python from metaperceptron import DataTransformer, Data from metaperceptron import MhaMlpRegressor, MhaMlpClassifier, MlpRegressor, MlpClassifier from metaperceptron import MhaMlpTuner, MhaMlpComparator

🔍 Example: Training an MLP Classifier with Genetic Algorithm

In this tutorial, we will use Genetic Algorithm to train Multi-Layer Perceptron network for classification task. For more complex examples and use cases, please check the folder examples.

```python from sklearn.datasets import loadiris from sklearn.modelselection import traintestsplit from metaperceptron import DataTransformer, MhaMlpClassifier

Load the dataset

X, y = loadiris(returnX_y=True)

Split train and test

Xtrain, ytrain, Xtest, ytest = traintestsplit(X, y, test_size=0.2)

Scale dataset with two methods: standard and minmax

dt = DataTransformer(scalingmethods=("standard", "minmax")) Xtrainscaled = dt.fittransform(Xtrain) Xtestscaled = dt.transform(Xtest)

Define Genetic Algorithm-trained Multi-Layer Perceptron

model = MhaMlpClassifier(hiddenlayers=(50, 15), actnames="Tanh", dropoutrates=None, actoutput=None, optim="BaseGA", optimparams={"epoch": 100, "popsize": 20, "name": "GA"}, obj_name="F1S", seed=42, verbose=True)

Train the model

model.fit(X=Xtrainscaled, y=y_train)

Test the model

ypred = model.predict(Xtest) print(y_pred)

Print the score

print(model.score(Xtestscaled, y_test))

Calculate some metrics

print(model.evaluate(ytrue=ytest, ypred=ypred, list_metrics=["AS", "PS", "RS", "F2S", "CKS", "FBS"])) ```

💬 Support

• 📦 Source Code
• 📖 Documentation
• ⬇️ PyPI Releases
• ❗ Report Issues
• 📝 Changelog
• 💬 Chat Group

Developed by: Thieu @ 2025