Model.features_importance¶

In [ ]:

Model.features_importance(X_idx: int = 0, 
                          ax=None, 
                          show: bool = True, 
                          **style_kwds,)

Computes the importance of the features of the model with time series aggregate functions.

Parameters¶

Name	Type	Optional	Description
X_idx	int/str	✓	Index of the main vector vcolumn used to draw the features importance. It can also be the name of a predictor vcolumn.
ax	Matplotlib axes object	✓	The axes to plot on.
show	bool	✓	If set to True, draw the features importance.
**style_kwds	any	✓	Any optional parameter to pass to the Matplotlib functions.

Returns¶

tablesample : An object containing the result. For more information, see utilities.tablesample.

Example¶

In [3]:

from verticapy.learn.tsa import VAR
model = VAR(name = "VAR_Texas",
            p = 5)
model.drop()
model.fit(input_relation = "Texas", 
          X = ["cases", "deaths"], 
          ts = "date")
model.features_importance(X_idx = "cases",)

Out[3]:

	importance	sign
ar0_1	52.90278531270462	1
ar0_2	16.454702499250395	-1
ar0_3	4.876031646413841	1
ar0_4	7.202416172229613	-1
ar0_5	4.093358739454691	1
ar1_1	1.2110964911996867	-1
ar1_2	0.24196327459720393	1
ar1_3	3.2234640965754604	-1
ar1_4	4.671230101818453	-1
ar1_5	5.122951665756048	1

Rows: 1-10 | Columns: 3