mlsimplified 0.1.0

A production-ready machine learning library for quick model training and prediction

MLSimplified

A production-ready machine learning library that makes training and using ML models as easy as writing a sentence.

Installation

pip install mlsimplified

Quick Start

from mlsimplified import Model

# Create and train a model in one line
model = Model("data.csv", target="label")
model.train().evaluate().predict("test.csv").export("model.pkl")

Features

• Automatic problem type detection (classification/regression)
• Smart data preprocessing and cleaning
• Built-in model selection
• One-line training and prediction
• Automatic evaluation and reporting
• Built-in visualization
• Production-ready error handling
• Efficient memory management

Usage

Basic Usage

from mlsimplified import Model

# Initialize with your data
model = Model("your_data.csv", target="target_column")

# Train the model
model.train()

# Make predictions
predictions = model.predict("new_data.csv")

# Export the model
model.export("model.pkl")

Advanced Usage

from mlsimplified import Model

# Initialize with custom parameters
model = Model("data.csv", target="label")

# Train with custom test size
model.train(test_size=0.3)

# Get model insights
model.summary()  # View model information
model.plot()     # Visualize feature importance
model.report()   # Generate evaluation report

# Chain operations
model.train().evaluate().predict("test.csv").export("model.pkl")

API Reference

Model Class

Initialization

Model(data: Union[str, pd.DataFrame], target: str)

Methods

• train(test_size: float = 0.2, random_state: int = 42) -> Model
• evaluate() -> Model
• predict(data: Union[str, pd.DataFrame]) -> np.ndarray
• export(path: str) -> Model
• summary() -> Model
• plot() -> Model
• report() -> Model

Dependencies

• pandas>=1.3.0
• numpy>=1.21.0
• scikit-learn>=1.0.0
• matplotlib>=3.4.0
• seaborn>=0.11.0
• joblib>=1.0.0
• setuptools>=65.5.1

License

MIT