MKYZ

Load and preprocess your dataset with automatic handling of missing values and encoding.

from mkyz import data_processing as dp

# Load and preprocess data
data = dp.prepare_data(
    'data.csv',
    target_column='target',
    handle_missing=True,
    encode_categorical=True
)

Dataset loaded successfully:
- Features: 15
- Samples: 1000
- Missing values handled: 23
- Categorical columns encoded: 4

Create insightful visualizations with a single line of code.

from mkyz import visualization as viz

# Create correlation matrix and feature importance plots
viz.visualize(
    data=data,
    plot_type=['corr', 'feature_importance'],
    save_path='plots/'
)

Plots generated:
- correlation_matrix.png
- feature_importance.png
Saved to: plots/

Train your model with automated hyperparameter optimization.

from mkyz import training as tr

# Train model with automatic optimization
model = tr.train(
    data=data,
    task='classification',
    model='rf',
    optimize_params=True
)

Model trained successfully:
Accuracy: 0.95
F1 Score: 0.94
Best parameters found:
- n_estimators: 100
- max_depth: 15

Evaluate your model's performance with comprehensive metrics.

# Make predictions and evaluate
predictions = model.predict(data.test_x)
results = tr.evaluate(
    true_values=data.test_y,
    predictions=predictions,
    task='classification'
)
print(results)

Classification Report:
              precision    recall  f1-score
Class 0       0.96        0.94    0.95
Class 1       0.93        0.95    0.94
Accuracy: 0.945