Metadata-Version: 2.1
Name: OERPredict
Version: 1.0
Summary: A machine learning library for predicting OER catalysis performance
Home-page: https://github.com/liyihang1024/OERPredict
Author: liyihang
Author-email: liyihang@shu.edu.cn
Project-URL: Documentation, https://yourprojectdocs.com
Project-URL: Source, https://github.com/liyihang1024/OERPredict
Project-URL: Tracker, https://github.com/liyihang1024/OERPredict/issues
Project-URL: Homepage, https://github.com/liyihang1024/OERPredict
Keywords: OER,machine learning,catalysis,prediction
Platform: any
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Science/Research
Classifier: Programming Language :: Python :: 3.8
Classifier: Programming Language :: Python :: 3.9
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: Implementation :: CPython
Requires-Python: >=3.6
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: numpy>=1.18.0
Requires-Dist: pandas>=1.0.0
Requires-Dist: scikit-learn>=0.22.0
Requires-Dist: matplotlib>=3.0.0
Requires-Dist: tensorflow>=2.0.0
Provides-Extra: dev
Requires-Dist: pytest>=5.0; extra == "dev"
Requires-Dist: black; extra == "dev"
Requires-Dist: flake8; extra == "dev"
Requires-Dist: tox; extra == "dev"
Provides-Extra: docs
Requires-Dist: sphinx; extra == "docs"
Requires-Dist: sphinx_rtd_theme; extra == "docs"
Requires-Dist: numpydoc; extra == "docs"

# OERPredict

**OERPredict** is a Python library for predicting Oxygen Evolution Reaction (OER) catalysis performance using machine learning models. This library provides regression and classification models for OER catalyst performance prediction, allowing researchers and scientists to efficiently analyze and predict the efficiency of various catalysts for energy conversion processes.

## Features

- **Machine Learning Models**: Implements regression and classification models based on `scikit-learn` for predicting OER catalyst performance.
- **Data Preprocessing**: Includes utilities for data cleaning, normalization, and splitting for model training and evaluation.
- **Model Evaluation**: Provides functions for evaluating model performance using common metrics like MSE, R², accuracy, and confusion matrix.
- **Customization**: Users can easily modify the models to experiment with different machine learning algorithms or feature engineering techniques.

## Installation

To install the library, you can use `pip` from PyPI:

```bash
pip install OERPredict
````

Alternatively, if you want to install the library from the source code, you can clone the repository and use the following commands:

```bash
git clone https://github.com/liyihang1024/OERPredict.git
cd OERPredict
pip install .
```

### Dependencies

The library requires the following Python packages:

* `numpy>=1.18.0`
* `pandas>=1.0.0`
* `scikit-learn>=0.22.0`
* `matplotlib>=3.0.0`
* `tensorflow>=2.0.0` (if using deep learning models)

These dependencies are automatically installed when you use `pip install OERPredict`.

### Optional Development Dependencies

To set up a development environment, you can install additional dependencies using:

```bash
pip install OERPredict[dev]
```

This will install tools for testing and code formatting, including:

* `pytest`
* `black`
* `flake8`

## Usage

Once installed, you can start using `OERPredict` to build models and predict the OER performance of various catalysts.

### Example: Regression Model

The regression model predicts continuous values, such as OER catalyst performance efficiency.

```python
from OERPredict.regression import OERRegressor
import pandas as pd

# Load dataset
data = pd.read_csv('data/dataset.csv')

# Initialize and train the model
regressor = OERRegressor(data)
regressor.train()

# Make predictions
new_data = pd.DataFrame({'feature1': [1.0], 'feature2': [2.0], 'feature3': [0.6]})
predictions = regressor.predict(new_data)
print(predictions)
```

### Example: Classification Model

The classification model predicts discrete categories, such as "High", "Medium", or "Low" catalyst efficiency.

```python
from OERPredict.classification import OERClassifier
import pandas as pd

# Load dataset
data = pd.read_csv('data/dataset.csv')

# Initialize and train the model
classifier = OERClassifier(data)
classifier.train()

# Make predictions
new_data = pd.DataFrame({'feature1': [1.0], 'feature2': [2.0], 'feature3': [0.6]})
predictions = classifier.predict(new_data)
print(predictions)
```

## Data Format

The library expects datasets in CSV format with at least the following columns:

1. **Features**: Numeric values representing the properties of the OER catalyst.
2. **Target (Regression)**: For regression tasks, the target column should contain continuous values representing OER performance.
3. **Target (Classification)**: For classification tasks, the target column should contain categorical values (e.g., "High", "Medium", "Low").

### Example Dataset (Regression)

```csv
feature1, feature2, feature3, OER_performance
1.2, 3.4, 0.5, 0.85
2.1, 4.2, 0.8, 0.90
1.5, 3.1, 0.6, 0.80
```

### Example Dataset (Classification)

```csv
feature1, feature2, feature3, OER_class
1.2, 3.4, 0.5, High
2.1, 4.2, 0.8, High
1.5, 3.1, 0.6, Low
```

## Model Evaluation

The library includes built-in evaluation functions for assessing model performance.

### For Regression Models:

* **Mean Squared Error (MSE)**
* **R² Score**

### For Classification Models:

* **Accuracy**
* **Confusion Matrix**
* **Classification Report** (Precision, Recall, F1-Score)

Example:

```python
from OERPredict.evaluation import evaluate_regression_model, evaluate_classification_model

# Evaluate regression model
evaluate_regression_model(y_true, y_pred)

# Evaluate classification model
evaluate_classification_model(y_true, y_pred)
```

## Changelog

### \[0.5] - 2024-06-15

* **Added**: Full support for regression and classification models.
* **Improved**: Model training performance and optimization.
* **Fixed**: Data preprocessing bug when handling missing values.
* **Updated**: Documentation for data format and usage examples.

### \[0.4] - 2024-05-10

* **Added**: Initial implementation of the classification model for OER catalysis.
* **Fixed**: Bug in model evaluation where metrics were not computed for classification tasks.
* **Improved**: Increased compatibility with newer versions of `scikit-learn` and `pandas`.

### \[0.3] - 2024-03-25

* **Initial Release**: First version released with regression model for predicting OER performance based on input features.

## Contributing

We welcome contributions to improve the `OERPredict` library. To contribute:

1. Fork the repository.
2. Create a new branch for your feature or bug fix.
3. Make your changes and ensure all tests pass.
4. Submit a pull request with a description of your changes.

## License

`OERPredict` is released under the MIT License. See the [LICENSE](LICENSE) file for more details.

## Acknowledgments

We would like to thank the contributors and the scientific community for their support in developing machine learning models for catalysis research.

## Contact

For any questions or suggestions, please feel free to reach out to the author at:

* **Author**: liyihang
* **Email**: [liyihang@shu.edu.cn](mailto:liyihang@shu.edu.cn)
* **GitHub**: [https://github.com/liyihang1024/OERPredict](https://github.com/liyihang1024/OERPredict)
