> For the complete documentation index, see [llms.txt](https://tusharkolekar24.gitbook.io/package-kolekar-1/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://tusharkolekar24.gitbook.io/package-kolekar-1/welcome-to-package-kolekar/ensemble.md).

# Ensemble

## Averaging Technique

Multiple models predicted output could be average by averaging technique and can be imported from ensemble module of package-Kolekar 

```python
from package_Kolekar.Ensemble_Learning.Avg_Ensemble import Average_weight_Ensemble
```

Preparing Model for Averaging Techniques.

```python
model = Average_weight_Ensemble()
```

Use the default base model to perform Average Ensembling.

```python
base_models = model.set_base_models()
```

Performing predictions using Average Ensemble Technique.

```python
y_train_pred = model.get_Averaging_technique(base_model=base_models, 
                                              train_X=X_train, 
                                              train_y=y_train, 
                                              test_X = X_train)
```

### Weighted-Averaging Techniques

In this case, multiple models predicted output combined with assigned weights and can be imported from the ensemble module of package-Kolekar. Here weights are assigned Manually. 

```python
weight=[0.3,0.2,0.1,0.05, 0.05]
y_test_pred = model.get_weighted_Avg_technique(base_model=base_models, 
                                               train_X=X_train, 
                                               train_y=y_train, 
                                               test_X = X_test,
                                               weights=weight)
```

Evaluating Prediction Performance of the models based on R2\_score, MAE, MSE, RMSE, MAPE, etc.

```python
result_test = model.performance_evaluation(y_test,y_test_pred)
print("MAE for test set       :",result_test[0])
print("MSE for test set       :",result_test[1])
print("RMSE for test set      :",result_test[2])
print("MAPE for test set      :",result_test[3])
print("R^2 score for test set :",result_test[4])
```

### Rank-Weighted Ensemble Technique

In this case, multiple models predicted output combined with assigned weights and can be imported from the ensemble module of package-Kolekar. Here weights are assigned based on an individual model given a prediction accuracy.

```python
summary = model.get_weights(threshold=0.5,
                            base_model=base_models,
                            train_X=X_train, 
                            test_X=X_test,
                            train_y=y_train,
                            test_y=y_test)
                            
weight=summary.weights.values
y_test_pred = model.get_weighted_Avg_technique(base_model=base_models, 
                                               train_X=X_train, 
                                               train_y=y_train, 
                                               test_X = X_test,
                                               weights=weight)
```