Model.classification_report / report¶

In [ ]:

Model.report(cutoff = [],
             labels: list = [],
             nbins: int = 10000)

Computes a classification report using multiple metrics to evaluate the model (AUC, accuracy, PRC AUC, F1...). In case of multiclass classification, it will consider each category as positive and switch to the next one during the computation.

Parameters¶

Name	Type	Optional	Description
cutoff	float / list	✓	Cutoff for which the tested category will be accepted as prediction. In case of multiclass classification, each tested category becomes the positives and the others are merged into the negatives. The list will represent the classes threshold. If it is empty or invalid, the best cutoff will be used.
labels	list	✓	List of the different labels to be used during the computation.
nbins	int	✓	[Used to compute ROC AUC, PRC AUC and the best cutoff] An integer value that determines the number of decision boundaries. Decision boundaries are set at equally spaced intervals between 0 and 1, inclusive. Greater values for nbins give more precise estimations of the metrics, but can potentially decrease performance. The maximum value is 999,999. If negative, the maximum value is used.

Returns¶

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

Example¶

In [6]:

from verticapy.datasets import load_iris
iris = load_iris()

# Multiclass Classification
from verticapy.learn.ensemble import RandomForestClassifier
model = RandomForestClassifier(name = "public.RF_iris",
                               n_estimators = 20,
                               max_features = "auto",
                               max_leaf_nodes = 32, 
                               sample = 0.7,
                               max_depth = 3,
                               min_samples_leaf = 5,
                               min_info_gain = 0.0,
                               nbins = 32)
model.fit(iris, ["PetalLengthCm", "PetalWidthCm"], "Species")
# Multiclass Classification: Using a fixed cutoff
model.report(cutoff = 0.33)

Out[6]:

	Iris-setosa	Iris-versicolor	Iris-virginica
auc	1.0	0.9961000000000003	0.9963000000000002
prc_auc	1.0	0.9925582491285112	0.9927773021392375
accuracy	1.0	0.9533333333333334	0.9666666666666667
log_loss	0.00918264030121329	0.0374488702584726	0.0364849953334055
precision	1.0	0.8771929824561403	0.9411764705882353
recall	1.0	1.0	0.96
f1_score	1.0	0.9345794392523363	0.9504950495049505
mcc	1.0	0.9032106474595007	0.9254762227411247
informedness	1.0	0.9300000000000002	0.9299999999999999
markedness	1.0	0.8771929824561404	0.9209744503862152
csi	1.0	0.8771929824561403	0.9056603773584906
cutoff	0.33	0.33	0.33

Rows: 1-12 | Columns: 4

In [7]:

# Multiclass Classification: Using automatic cutoffs
model.report()

Out[7]:

	Iris-setosa	Iris-versicolor	Iris-virginica
auc	1.0	0.9961000000000003	0.9963000000000002
prc_auc	1.0	0.9925582491285112	0.9927773021392375
accuracy	1.0	0.96	0.9666666666666667
log_loss	0.00918264030121329	0.0374488702584726	0.0364849953334055
precision	1.0	0.9074074074074074	0.9090909090909091
recall	1.0	0.98	1.0
f1_score	1.0	0.9423076923076924	0.9523809523809523
mcc	1.0	0.9133462590326239	0.929320377284585
informedness	1.0	0.9299999999999999	0.95
markedness	1.0	0.8969907407407409	0.9090909090909092
csi	1.0	0.8909090909090909	0.9090909090909091
cutoff	0.9029	0.54	0.0881

Rows: 1-12 | Columns: 4

In [8]:

# Multiclass Classification: Customized Cutoffs
model.report(cutoff = [0.8, 0.4, 0.2])

Out[8]:

	Iris-setosa	Iris-versicolor	Iris-virginica
auc	1.0	0.9961000000000003	0.9963000000000002
prc_auc	1.0	0.9925582491285112	0.9927773021392375
accuracy	1.0	0.96	0.96
log_loss	0.00918264030121329	0.0374488702584726	0.0364849953334055
precision	1.0	0.9074074074074074	0.9074074074074074
recall	1.0	0.98	0.98
f1_score	1.0	0.9423076923076924	0.9423076923076924
mcc	1.0	0.9133462590326239	0.9133462590326239
informedness	1.0	0.9299999999999999	0.9299999999999999
markedness	1.0	0.8969907407407409	0.8969907407407409
csi	1.0	0.8909090909090909	0.8909090909090909
cutoff	0.8	0.4	0.2

Rows: 1-12 | Columns: 4

In [9]:

# Multiclass Classification: Choosing the categories
model.report(labels = ["Iris-versicolor", "Iris-virginica"])

Out[9]:

	Iris-versicolor	Iris-virginica
auc	0.9961000000000003	0.9963000000000002
prc_auc	0.9925582491285112	0.9927773021392375
accuracy	0.96	0.9666666666666667
log_loss	0.0374488702584726	0.0364849953334055
precision	0.9074074074074074	0.9090909090909091
recall	0.98	1.0
f1_score	0.9423076923076924	0.9523809523809523
mcc	0.9133462590326239	0.929320377284585
informedness	0.9299999999999999	0.95
markedness	0.8969907407407409	0.9090909090909092
csi	0.8909090909090909	0.9090909090909091
cutoff	0.54	0.0881

Rows: 1-12 | Columns: 3