Feature importance¶

The importance of the predictors included in a forecaster can be obtained using the method get_feature_importance. This method accesses the coef_ and feature_importances_ attributes of the internal regressor.

  Warning

This method only returns values if the regressor used inside the forecaster has the attribute `coef_` or `feature_importances_`.

Libraries¶

In [7]:

# Libraries
# ==============================================================================
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.ensemble import RandomForestRegressor
from sklearn.linear_model import Ridge
from skforecast.ForecasterAutoreg import ForecasterAutoreg
from skforecast.ForecasterAutoregMultiOutput import ForecasterAutoregMultiOutput

# Libraries # ============================================================================== import pandas as pd import matplotlib.pyplot as plt from sklearn.ensemble import RandomForestRegressor from sklearn.linear_model import Ridge from skforecast.ForecasterAutoreg import ForecasterAutoreg from skforecast.ForecasterAutoregMultiOutput import ForecasterAutoregMultiOutput

Data¶

In [8]:

# Download data
# ==============================================================================
url = ('https://raw.githubusercontent.com/JoaquinAmatRodrigo/skforecast/master/data/h2o_exog.csv')
data = pd.read_csv(url, sep=',', header=0, names=['date', 'y', 'exog_1', 'exog_2'])

# Data preprocessing
# ==============================================================================
data['date'] = pd.to_datetime(data['date'], format='%Y/%m/%d')
data = data.set_index('date')
data = data.asfreq('MS')

# Download data # ============================================================================== url = ('https://raw.githubusercontent.com/JoaquinAmatRodrigo/skforecast/master/data/h2o_exog.csv') data = pd.read_csv(url, sep=',', header=0, names=['date', 'y', 'exog_1', 'exog_2']) # Data preprocessing # ============================================================================== data['date'] = pd.to_datetime(data['date'], format='%Y/%m/%d') data = data.set_index('date') data = data.asfreq('MS')

Extract feature importance from trained forecaster¶

In [9]:

# Create and fit forecaster
# ==============================================================================
forecaster = ForecasterAutoreg(
                regressor = RandomForestRegressor(),
                lags = 5
             )

forecaster.fit(y=data['y'], exog=data[['exog_1', 'exog_2']])

# Predictors importance
# ==============================================================================
forecaster.get_feature_importance()

# Create and fit forecaster # ============================================================================== forecaster = ForecasterAutoreg( regressor = RandomForestRegressor(), lags = 5 ) forecaster.fit(y=data['y'], exog=data[['exog_1', 'exog_2']]) # Predictors importance # ============================================================================== forecaster.get_feature_importance()

Out[9]:

	feature	importance
0	lag_1	0.533117
1	lag_2	0.097804
2	lag_3	0.029198
3	lag_4	0.073071
4	lag_5	0.053683
5	exog_1	0.042011
6	exog_2	0.171117

In [10]:

# Create and fit forecaster
# ==============================================================================
forecaster = ForecasterAutoreg(
                    regressor = Ridge(),
                    lags = 5
                )

forecaster.fit(y=data['y'], exog=data[['exog_1', 'exog_2']])

forecaster.get_feature_importance()

# Create and fit forecaster # ============================================================================== forecaster = ForecasterAutoreg( regressor = Ridge(), lags = 5 ) forecaster.fit(y=data['y'], exog=data[['exog_1', 'exog_2']]) forecaster.get_feature_importance()

Out[10]:

	feature	importance
0	lag_1	0.327688
1	lag_2	-0.073593
2	lag_3	-0.152202
3	lag_4	-0.217106
4	lag_5	-0.145800
5	exog_1	0.379798
6	exog_2	0.668162

Since ForecasterAutoregMultiOutput fits one model per step, it is necessary to specify from which model retrieves its feature importance.

In [11]:

# Create and fit forecaster
# ==============================================================================
forecaster = ForecasterAutoregMultiOutput(
                regressor = RandomForestRegressor(),
                steps = 10,
                lags = 5
             )

forecaster.fit(y=data['y'], exog=data[['exog_1', 'exog_2']])

# Predictors importance
# ==============================================================================
forecaster.get_feature_importance(step=1)

# Create and fit forecaster # ============================================================================== forecaster = ForecasterAutoregMultiOutput( regressor = RandomForestRegressor(), steps = 10, lags = 5 ) forecaster.fit(y=data['y'], exog=data[['exog_1', 'exog_2']]) # Predictors importance # ============================================================================== forecaster.get_feature_importance(step=1)

Out[11]:

	feature	importance
0	lag_1	0.537580
1	lag_2	0.102076
2	lag_3	0.024955
3	lag_4	0.074393
4	lag_5	0.051036
5	exog_1	0.055780
6	exog_2	0.154180

In [12]:

%%html
<style>
.jupyter-wrapper .jp-CodeCell .jp-Cell-inputWrapper .jp-InputPrompt {display: none;}
</style>

%%html