VerticaPy

Python API for Vertica Data Science at Scale

Example: Methods in a Multiclass Classification Model¶

In this example, we use the 'Iris' dataset to demonstrate the methods available to multiclass classification models.

In [1]:

from verticapy.datasets import load_iris
iris = load_iris()
display(iris)

	123 SepalLengthCm Numeric(5,2)	123 SepalWidthCm Numeric(5,2)	123 PetalLengthCm Numeric(5,2)	123 PetalWidthCm Numeric(5,2)	Abc Species Varchar(30)
1	4.30	3.00	1.10	0.10	Iris-setosa
2	4.40	2.90	1.40	0.20	Iris-setosa
3	4.40	3.00	1.30	0.20	Iris-setosa
4	4.40	3.20	1.30	0.20	Iris-setosa
5	4.50	2.30	1.30	0.30	Iris-setosa
6	4.60	3.10	1.50	0.20	Iris-setosa
7	4.60	3.20	1.40	0.20	Iris-setosa
8	4.60	3.40	1.40	0.30	Iris-setosa
9	4.60	3.60	1.00	0.20	Iris-setosa
10	4.70	3.20	1.30	0.20	Iris-setosa
11	4.70	3.20	1.60	0.20	Iris-setosa
12	4.80	3.00	1.40	0.10	Iris-setosa
13	4.80	3.00	1.40	0.30	Iris-setosa
14	4.80	3.10	1.60	0.20	Iris-setosa
15	4.80	3.40	1.60	0.20	Iris-setosa
16	4.80	3.40	1.90	0.20	Iris-setosa
17	4.90	2.40	3.30	1.00	Iris-versicolor
18	4.90	2.50	4.50	1.70	Iris-virginica
19	4.90	3.00	1.40	0.20	Iris-setosa
20	4.90	3.10	1.50	0.10	Iris-setosa
21	4.90	3.10	1.50	0.10	Iris-setosa
22	4.90	3.10	1.50	0.10	Iris-setosa
23	5.00	2.00	3.50	1.00	Iris-versicolor
24	5.00	2.30	3.30	1.00	Iris-versicolor
25	5.00	3.00	1.60	0.20	Iris-setosa
26	5.00	3.20	1.20	0.20	Iris-setosa
27	5.00	3.30	1.40	0.20	Iris-setosa
28	5.00	3.40	1.50	0.20	Iris-setosa
29	5.00	3.40	1.60	0.40	Iris-setosa
30	5.00	3.50	1.30	0.30	Iris-setosa
31	5.00	3.50	1.60	0.60	Iris-setosa
32	5.00	3.60	1.40	0.20	Iris-setosa
33	5.10	2.50	3.00	1.10	Iris-versicolor
34	5.10	3.30	1.70	0.50	Iris-setosa
35	5.10	3.40	1.50	0.20	Iris-setosa
36	5.10	3.50	1.40	0.20	Iris-setosa
37	5.10	3.50	1.40	0.30	Iris-setosa
38	5.10	3.70	1.50	0.40	Iris-setosa
39	5.10	3.80	1.50	0.30	Iris-setosa
40	5.10	3.80	1.60	0.20	Iris-setosa
41	5.10	3.80	1.90	0.40	Iris-setosa
42	5.20	2.70	3.90	1.40	Iris-versicolor
43	5.20	3.40	1.40	0.20	Iris-setosa
44	5.20	3.50	1.50	0.20	Iris-setosa
45	5.20	4.10	1.50	0.10	Iris-setosa
46	5.30	3.70	1.50	0.20	Iris-setosa
47	5.40	3.00	4.50	1.50	Iris-versicolor
48	5.40	3.40	1.50	0.40	Iris-setosa
49	5.40	3.40	1.70	0.20	Iris-setosa
50	5.40	3.70	1.50	0.20	Iris-setosa
51	5.40	3.90	1.30	0.40	Iris-setosa
52	5.40	3.90	1.70	0.40	Iris-setosa
53	5.50	2.30	4.00	1.30	Iris-versicolor
54	5.50	2.40	3.70	1.00	Iris-versicolor
55	5.50	2.40	3.80	1.10	Iris-versicolor
56	5.50	2.50	4.00	1.30	Iris-versicolor
57	5.50	2.60	4.40	1.20	Iris-versicolor
58	5.50	3.50	1.30	0.20	Iris-setosa
59	5.50	4.20	1.40	0.20	Iris-setosa
60	5.60	2.50	3.90	1.10	Iris-versicolor
61	5.60	2.70	4.20	1.30	Iris-versicolor
62	5.60	2.80	4.90	2.00	Iris-virginica
63	5.60	2.90	3.60	1.30	Iris-versicolor
64	5.60	3.00	4.10	1.30	Iris-versicolor
65	5.60	3.00	4.50	1.50	Iris-versicolor
66	5.70	2.50	5.00	2.00	Iris-virginica
67	5.70	2.60	3.50	1.00	Iris-versicolor
68	5.70	2.80	4.10	1.30	Iris-versicolor
69	5.70	2.80	4.50	1.30	Iris-versicolor
70	5.70	2.90	4.20	1.30	Iris-versicolor
71	5.70	3.00	4.20	1.20	Iris-versicolor
72	5.70	3.80	1.70	0.30	Iris-setosa
73	5.70	4.40	1.50	0.40	Iris-setosa
74	5.80	2.60	4.00	1.20	Iris-versicolor
75	5.80	2.70	3.90	1.20	Iris-versicolor
76	5.80	2.70	4.10	1.00	Iris-versicolor
77	5.80	2.70	5.10	1.90	Iris-virginica
78	5.80	2.70	5.10	1.90	Iris-virginica
79	5.80	2.80	5.10	2.40	Iris-virginica
80	5.80	4.00	1.20	0.20	Iris-setosa
81	5.90	3.00	4.20	1.50	Iris-versicolor
82	5.90	3.00	5.10	1.80	Iris-virginica
83	5.90	3.20	4.80	1.80	Iris-versicolor
84	6.00	2.20	4.00	1.00	Iris-versicolor
85	6.00	2.20	5.00	1.50	Iris-virginica
86	6.00	2.70	5.10	1.60	Iris-versicolor
87	6.00	2.90	4.50	1.50	Iris-versicolor
88	6.00	3.00	4.80	1.80	Iris-virginica
89	6.00	3.40	4.50	1.60	Iris-versicolor
90	6.10	2.60	5.60	1.40	Iris-virginica
91	6.10	2.80	4.00	1.30	Iris-versicolor
92	6.10	2.80	4.70	1.20	Iris-versicolor
93	6.10	2.90	4.70	1.40	Iris-versicolor
94	6.10	3.00	4.60	1.40	Iris-versicolor
95	6.10	3.00	4.90	1.80	Iris-virginica
96	6.20	2.20	4.50	1.50	Iris-versicolor
97	6.20	2.80	4.80	1.80	Iris-virginica
98	6.20	2.90	4.30	1.30	Iris-versicolor
99	6.20	3.40	5.40	2.30	Iris-virginica
100	6.30	2.30	4.40	1.30	Iris-versicolor

Rows: 1-100 of 150 | Columns: 5

Our goal is to predict the flower species. Let's create a random forest model.

In [2]:

from verticapy.learn.ensemble import RandomForestClassifier
model = RandomForestClassifier("public.RF_iris")
model.fit("public.iris", 
          ["PetalLengthCm", "SepalWidthCm", "SepalLengthCm", "PetalWidthCm"], 
          "Species")

Out[2]:


===========
call_string
===========
SELECT rf_classifier('public.RF_iris', 'public.iris', '"species"', '"PetalLengthCm", "SepalWidthCm", "SepalLengthCm", "PetalWidthCm"' USING PARAMETERS exclude_columns='', ntree=10, mtry=2, sampling_size=0.632, max_depth=5, max_breadth=1000000000, min_leaf_size=1, min_info_gain=0, nbins=32);

=======
details
=======
  predictor  |      type      
-------------+----------------
petallengthcm|float or numeric
sepalwidthcm |float or numeric
sepallengthcm|float or numeric
petalwidthcm |float or numeric


===============
Additional Info
===============
       Name       |Value
------------------+-----
    tree_count    | 10  
rejected_row_count|  0  
accepted_row_count| 150

Fitting the model creates new model attributes.

In [3]:

model.X

Out[3]:

['"PetalLengthCm"', '"SepalWidthCm"', '"SepalLengthCm"', '"PetalWidthCm"']

In [4]:

model.y

Out[4]:

'"Species"'

In [5]:

model.input_relation

Out[5]:

'public.iris'

In [6]:

model.test_relation

Out[6]:

'public.iris'

Since we didn't write a test relation when fitting the model, the model will use the training relation as the test relation.

Let's compute the accuracy of our model.

In [7]:

model.score(method = "accuracy")

Out[7]:

0.986666666666667

The 'score' method uses the 'y' attribute and the model prediction in the 'test_relation' to compute the accuracy of the model. You can change these attributes at any time to deploy the models on different columns.

Models have many useful attributes. For a random forest model, we can use the 'classes_' attribute for a list of the response categories.

In [8]:

model.classes_

Out[8]:

['Iris-setosa', 'Iris-versicolor', 'Iris-virginica']

You can view other attributes using the 'get_attr' method.

In [18]:

model.get_attr()

	Abc attr_name Varchar(128)	Abc attr_fields Long varchar(32000000)	123 #_of_rows Integer
1	tree_count	tree_count	1
2	rejected_row_count	rejected_row_count	1
3	accepted_row_count	accepted_row_count	1
4	call_string	call_string	1
5	details	predictor, type	4

Out[18]:

Visualizing the model with a PRC, ROC, or lift chart might be helpful. We should keep in mind that this is a multiclass classification, so one class must be considered positive. We can identify this class with the 'pos_label'. For this example, we'll mark the "Iris-setosa' as the positive class for each graph.

In [9]:

model.roc_curve(pos_label = 'Iris-setosa')
model.prc_curve(pos_label = 'Iris-setosa')
model.lift_chart(pos_label = 'Iris-setosa')

Let's look at the SQL query for our model.

In [10]:

display(model.deploySQL())

PREDICT_RF_CLASSIFIER("PetalLengthCm", "SepalWidthCm", "SepalLengthCm", "PetalWidthCm" USING PARAMETERS model_name = 'public.RF_iris', match_by_pos = 'true')

You can evaluate the quality of your model with the 'report' method. Again, since this is a multiclass classification, each class will have its turn as the 'positive' class during each round.

If we don't specify a cutoff, VerticaPy automatically computes an optimal cutoff.

In [11]:

model.report()

	Iris-setosa	Iris-versicolor	Iris-virginica
auc	1.0	0.9998000000000002	0.9998000000000001
prc_auc	1.0	0.9996039215686274	0.9996039215686274
accuracy	1.0	0.9933333333333333	0.9933333333333333
log_loss	0.00398288918014474	0.0192465228007074	0.0188544198895192
precision	1.0	0.9803921568627451	0.9803921568627451
recall	1.0	1.0	1.0
f1_score	1.0	0.9949748743718593	0.9949748743718593
mcc	1.0	0.985184366143778	0.985184366143778
informedness	1.0	0.99	0.99
markedness	1.0	0.9803921568627452	0.9803921568627452
csi	1.0	0.9803921568627451	0.9803921568627451
cutoff	0.9	0.364	0.438

Out[11]:

You can add this prediction to your vDataFrame.

In [12]:

model.predict(iris, name = "pred_Species")

	123 SepalLengthCm Numeric(5,2)	123 SepalWidthCm Numeric(5,2)	123 PetalLengthCm Numeric(5,2)	123 PetalWidthCm Numeric(5,2)	Abc Species Varchar(30)	Abc pred_Species Varchar(128)
1	4.30	3.00	1.10	0.10	Iris-setosa	Iris-setosa
2	4.40	2.90	1.40	0.20	Iris-setosa	Iris-setosa
3	4.40	3.00	1.30	0.20	Iris-setosa	Iris-setosa
4	4.40	3.20	1.30	0.20	Iris-setosa	Iris-setosa
5	4.50	2.30	1.30	0.30	Iris-setosa	Iris-setosa
6	4.60	3.10	1.50	0.20	Iris-setosa	Iris-setosa
7	4.60	3.20	1.40	0.20	Iris-setosa	Iris-setosa
8	4.60	3.40	1.40	0.30	Iris-setosa	Iris-setosa
9	4.60	3.60	1.00	0.20	Iris-setosa	Iris-setosa
10	4.70	3.20	1.30	0.20	Iris-setosa	Iris-setosa
11	4.70	3.20	1.60	0.20	Iris-setosa	Iris-setosa
12	4.80	3.00	1.40	0.10	Iris-setosa	Iris-setosa
13	4.80	3.00	1.40	0.30	Iris-setosa	Iris-setosa
14	4.80	3.10	1.60	0.20	Iris-setosa	Iris-setosa
15	4.80	3.40	1.60	0.20	Iris-setosa	Iris-setosa
16	4.80	3.40	1.90	0.20	Iris-setosa	Iris-setosa
17	4.90	2.40	3.30	1.00	Iris-versicolor	Iris-versicolor
18	4.90	2.50	4.50	1.70	Iris-virginica	Iris-virginica
19	4.90	3.00	1.40	0.20	Iris-setosa	Iris-setosa
20	4.90	3.10	1.50	0.10	Iris-setosa	Iris-setosa
21	4.90	3.10	1.50	0.10	Iris-setosa	Iris-setosa
22	4.90	3.10	1.50	0.10	Iris-setosa	Iris-setosa
23	5.00	2.00	3.50	1.00	Iris-versicolor	Iris-versicolor
24	5.00	2.30	3.30	1.00	Iris-versicolor	Iris-versicolor
25	5.00	3.00	1.60	0.20	Iris-setosa	Iris-setosa
26	5.00	3.20	1.20	0.20	Iris-setosa	Iris-setosa
27	5.00	3.30	1.40	0.20	Iris-setosa	Iris-setosa
28	5.00	3.40	1.50	0.20	Iris-setosa	Iris-setosa
29	5.00	3.40	1.60	0.40	Iris-setosa	Iris-setosa
30	5.00	3.50	1.30	0.30	Iris-setosa	Iris-setosa
31	5.00	3.50	1.60	0.60	Iris-setosa	Iris-setosa
32	5.00	3.60	1.40	0.20	Iris-setosa	Iris-setosa
33	5.10	2.50	3.00	1.10	Iris-versicolor	Iris-versicolor
34	5.10	3.30	1.70	0.50	Iris-setosa	Iris-setosa
35	5.10	3.40	1.50	0.20	Iris-setosa	Iris-setosa
36	5.10	3.50	1.40	0.20	Iris-setosa	Iris-setosa
37	5.10	3.50	1.40	0.30	Iris-setosa	Iris-setosa
38	5.10	3.70	1.50	0.40	Iris-setosa	Iris-setosa
39	5.10	3.80	1.50	0.30	Iris-setosa	Iris-setosa
40	5.10	3.80	1.60	0.20	Iris-setosa	Iris-setosa
41	5.10	3.80	1.90	0.40	Iris-setosa	Iris-setosa
42	5.20	2.70	3.90	1.40	Iris-versicolor	Iris-versicolor
43	5.20	3.40	1.40	0.20	Iris-setosa	Iris-setosa
44	5.20	3.50	1.50	0.20	Iris-setosa	Iris-setosa
45	5.20	4.10	1.50	0.10	Iris-setosa	Iris-setosa
46	5.30	3.70	1.50	0.20	Iris-setosa	Iris-setosa
47	5.40	3.00	4.50	1.50	Iris-versicolor	Iris-versicolor
48	5.40	3.40	1.50	0.40	Iris-setosa	Iris-setosa
49	5.40	3.40	1.70	0.20	Iris-setosa	Iris-setosa
50	5.40	3.70	1.50	0.20	Iris-setosa	Iris-setosa
51	5.40	3.90	1.30	0.40	Iris-setosa	Iris-setosa
52	5.40	3.90	1.70	0.40	Iris-setosa	Iris-setosa
53	5.50	2.30	4.00	1.30	Iris-versicolor	Iris-versicolor
54	5.50	2.40	3.70	1.00	Iris-versicolor	Iris-versicolor
55	5.50	2.40	3.80	1.10	Iris-versicolor	Iris-versicolor
56	5.50	2.50	4.00	1.30	Iris-versicolor	Iris-versicolor
57	5.50	2.60	4.40	1.20	Iris-versicolor	Iris-versicolor
58	5.50	3.50	1.30	0.20	Iris-setosa	Iris-setosa
59	5.50	4.20	1.40	0.20	Iris-setosa	Iris-setosa
60	5.60	2.50	3.90	1.10	Iris-versicolor	Iris-versicolor
61	5.60	2.70	4.20	1.30	Iris-versicolor	Iris-versicolor
62	5.60	2.80	4.90	2.00	Iris-virginica	Iris-virginica
63	5.60	2.90	3.60	1.30	Iris-versicolor	Iris-versicolor
64	5.60	3.00	4.10	1.30	Iris-versicolor	Iris-versicolor
65	5.60	3.00	4.50	1.50	Iris-versicolor	Iris-versicolor
66	5.70	2.50	5.00	2.00	Iris-virginica	Iris-virginica
67	5.70	2.60	3.50	1.00	Iris-versicolor	Iris-versicolor
68	5.70	2.80	4.10	1.30	Iris-versicolor	Iris-versicolor
69	5.70	2.80	4.50	1.30	Iris-versicolor	Iris-versicolor
70	5.70	2.90	4.20	1.30	Iris-versicolor	Iris-versicolor
71	5.70	3.00	4.20	1.20	Iris-versicolor	Iris-versicolor
72	5.70	3.80	1.70	0.30	Iris-setosa	Iris-setosa
73	5.70	4.40	1.50	0.40	Iris-setosa	Iris-setosa
74	5.80	2.60	4.00	1.20	Iris-versicolor	Iris-versicolor
75	5.80	2.70	3.90	1.20	Iris-versicolor	Iris-versicolor
76	5.80	2.70	4.10	1.00	Iris-versicolor	Iris-versicolor
77	5.80	2.70	5.10	1.90	Iris-virginica	Iris-virginica
78	5.80	2.70	5.10	1.90	Iris-virginica	Iris-virginica
79	5.80	2.80	5.10	2.40	Iris-virginica	Iris-virginica
80	5.80	4.00	1.20	0.20	Iris-setosa	Iris-setosa
81	5.90	3.00	4.20	1.50	Iris-versicolor	Iris-versicolor
82	5.90	3.00	5.10	1.80	Iris-virginica	Iris-virginica
83	5.90	3.20	4.80	1.80	Iris-versicolor	Iris-versicolor
84	6.00	2.20	4.00	1.00	Iris-versicolor	Iris-versicolor
85	6.00	2.20	5.00	1.50	Iris-virginica	Iris-versicolor
86	6.00	2.70	5.10	1.60	Iris-versicolor	Iris-virginica
87	6.00	2.90	4.50	1.50	Iris-versicolor	Iris-versicolor
88	6.00	3.00	4.80	1.80	Iris-virginica	Iris-virginica
89	6.00	3.40	4.50	1.60	Iris-versicolor	Iris-versicolor
90	6.10	2.60	5.60	1.40	Iris-virginica	Iris-virginica
91	6.10	2.80	4.00	1.30	Iris-versicolor	Iris-versicolor
92	6.10	2.80	4.70	1.20	Iris-versicolor	Iris-versicolor
93	6.10	2.90	4.70	1.40	Iris-versicolor	Iris-versicolor
94	6.10	3.00	4.60	1.40	Iris-versicolor	Iris-versicolor
95	6.10	3.00	4.90	1.80	Iris-virginica	Iris-virginica
96	6.20	2.20	4.50	1.50	Iris-versicolor	Iris-versicolor
97	6.20	2.80	4.80	1.80	Iris-virginica	Iris-virginica
98	6.20	2.90	4.30	1.30	Iris-versicolor	Iris-versicolor
99	6.20	3.40	5.40	2.30	Iris-virginica	Iris-virginica
100	6.30	2.30	4.40	1.30	Iris-versicolor	Iris-versicolor

Out[12]:

Rows: 1-100 of 150 | Columns: 6

You can also add the probability of a specific class with the 'pos_label' parameter.

In [13]:

model.predict(iris, name = "prob_versicolor", pos_label = "Iris-versicolor")

	123 SepalLengthCm Numeric(5,2)	123 SepalWidthCm Numeric(5,2)	123 PetalLengthCm Numeric(5,2)	123 PetalWidthCm Numeric(5,2)	Abc Species Varchar(30)	Abc pred_Species Varchar(128)	Abc prob_versicolor Varchar(128)
1	4.30	3.00	1.10	0.10	Iris-setosa	Iris-setosa	0
2	4.40	2.90	1.40	0.20	Iris-setosa	Iris-setosa	0
3	4.40	3.00	1.30	0.20	Iris-setosa	Iris-setosa	0
4	4.40	3.20	1.30	0.20	Iris-setosa	Iris-setosa	0
5	4.50	2.30	1.30	0.30	Iris-setosa	Iris-setosa	0
6	4.60	3.10	1.50	0.20	Iris-setosa	Iris-setosa	0
7	4.60	3.20	1.40	0.20	Iris-setosa	Iris-setosa	0
8	4.60	3.40	1.40	0.30	Iris-setosa	Iris-setosa	0
9	4.60	3.60	1.00	0.20	Iris-setosa	Iris-setosa	0
10	4.70	3.20	1.30	0.20	Iris-setosa	Iris-setosa	0
11	4.70	3.20	1.60	0.20	Iris-setosa	Iris-setosa	0
12	4.80	3.00	1.40	0.10	Iris-setosa	Iris-setosa	0
13	4.80	3.00	1.40	0.30	Iris-setosa	Iris-setosa	0
14	4.80	3.10	1.60	0.20	Iris-setosa	Iris-setosa	0
15	4.80	3.40	1.60	0.20	Iris-setosa	Iris-setosa	0
16	4.80	3.40	1.90	0.20	Iris-setosa	Iris-setosa	0
17	4.90	2.40	3.30	1.00	Iris-versicolor	Iris-versicolor	0.9
18	4.90	2.50	4.50	1.70	Iris-virginica	Iris-virginica	0.3
19	4.90	3.00	1.40	0.20	Iris-setosa	Iris-setosa	0
20	4.90	3.10	1.50	0.10	Iris-setosa	Iris-setosa	0
21	4.90	3.10	1.50	0.10	Iris-setosa	Iris-setosa	0
22	4.90	3.10	1.50	0.10	Iris-setosa	Iris-setosa	0
23	5.00	2.00	3.50	1.00	Iris-versicolor	Iris-versicolor	1
24	5.00	2.30	3.30	1.00	Iris-versicolor	Iris-versicolor	1
25	5.00	3.00	1.60	0.20	Iris-setosa	Iris-setosa	0
26	5.00	3.20	1.20	0.20	Iris-setosa	Iris-setosa	0
27	5.00	3.30	1.40	0.20	Iris-setosa	Iris-setosa	0
28	5.00	3.40	1.50	0.20	Iris-setosa	Iris-setosa	0
29	5.00	3.40	1.60	0.40	Iris-setosa	Iris-setosa	0
30	5.00	3.50	1.30	0.30	Iris-setosa	Iris-setosa	0
31	5.00	3.50	1.60	0.60	Iris-setosa	Iris-setosa	0.1
32	5.00	3.60	1.40	0.20	Iris-setosa	Iris-setosa	0
33	5.10	2.50	3.00	1.10	Iris-versicolor	Iris-versicolor	1
34	5.10	3.30	1.70	0.50	Iris-setosa	Iris-setosa	0.1
35	5.10	3.40	1.50	0.20	Iris-setosa	Iris-setosa	0
36	5.10	3.50	1.40	0.20	Iris-setosa	Iris-setosa	0
37	5.10	3.50	1.40	0.30	Iris-setosa	Iris-setosa	0
38	5.10	3.70	1.50	0.40	Iris-setosa	Iris-setosa	0
39	5.10	3.80	1.50	0.30	Iris-setosa	Iris-setosa	0
40	5.10	3.80	1.60	0.20	Iris-setosa	Iris-setosa	0
41	5.10	3.80	1.90	0.40	Iris-setosa	Iris-setosa	0
42	5.20	2.70	3.90	1.40	Iris-versicolor	Iris-versicolor	1
43	5.20	3.40	1.40	0.20	Iris-setosa	Iris-setosa	0
44	5.20	3.50	1.50	0.20	Iris-setosa	Iris-setosa	0
45	5.20	4.10	1.50	0.10	Iris-setosa	Iris-setosa	0
46	5.30	3.70	1.50	0.20	Iris-setosa	Iris-setosa	0
47	5.40	3.00	4.50	1.50	Iris-versicolor	Iris-versicolor	1
48	5.40	3.40	1.50	0.40	Iris-setosa	Iris-setosa	0
49	5.40	3.40	1.70	0.20	Iris-setosa	Iris-setosa	0
50	5.40	3.70	1.50	0.20	Iris-setosa	Iris-setosa	0
51	5.40	3.90	1.30	0.40	Iris-setosa	Iris-setosa	0
52	5.40	3.90	1.70	0.40	Iris-setosa	Iris-setosa	0
53	5.50	2.30	4.00	1.30	Iris-versicolor	Iris-versicolor	1
54	5.50	2.40	3.70	1.00	Iris-versicolor	Iris-versicolor	1
55	5.50	2.40	3.80	1.10	Iris-versicolor	Iris-versicolor	1
56	5.50	2.50	4.00	1.30	Iris-versicolor	Iris-versicolor	1
57	5.50	2.60	4.40	1.20	Iris-versicolor	Iris-versicolor	1
58	5.50	3.50	1.30	0.20	Iris-setosa	Iris-setosa	0
59	5.50	4.20	1.40	0.20	Iris-setosa	Iris-setosa	0
60	5.60	2.50	3.90	1.10	Iris-versicolor	Iris-versicolor	1
61	5.60	2.70	4.20	1.30	Iris-versicolor	Iris-versicolor	1
62	5.60	2.80	4.90	2.00	Iris-virginica	Iris-virginica	0.01
63	5.60	2.90	3.60	1.30	Iris-versicolor	Iris-versicolor	1
64	5.60	3.00	4.10	1.30	Iris-versicolor	Iris-versicolor	1
65	5.60	3.00	4.50	1.50	Iris-versicolor	Iris-versicolor	1
66	5.70	2.50	5.00	2.00	Iris-virginica	Iris-virginica	0.01
67	5.70	2.60	3.50	1.00	Iris-versicolor	Iris-versicolor	1
68	5.70	2.80	4.10	1.30	Iris-versicolor	Iris-versicolor	1
69	5.70	2.80	4.50	1.30	Iris-versicolor	Iris-versicolor	1
70	5.70	2.90	4.20	1.30	Iris-versicolor	Iris-versicolor	1
71	5.70	3.00	4.20	1.20	Iris-versicolor	Iris-versicolor	1
72	5.70	3.80	1.70	0.30	Iris-setosa	Iris-setosa	0
73	5.70	4.40	1.50	0.40	Iris-setosa	Iris-setosa	0
74	5.80	2.60	4.00	1.20	Iris-versicolor	Iris-versicolor	1
75	5.80	2.70	3.90	1.20	Iris-versicolor	Iris-versicolor	1
76	5.80	2.70	4.10	1.00	Iris-versicolor	Iris-versicolor	1
77	5.80	2.70	5.10	1.90	Iris-virginica	Iris-virginica	0.05
78	5.80	2.70	5.10	1.90	Iris-virginica	Iris-virginica	0.05
79	5.80	2.80	5.10	2.40	Iris-virginica	Iris-virginica	0.05
80	5.80	4.00	1.20	0.20	Iris-setosa	Iris-setosa	0
81	5.90	3.00	4.20	1.50	Iris-versicolor	Iris-versicolor	1
82	5.90	3.00	5.10	1.80	Iris-virginica	Iris-virginica	0.0714286
83	5.90	3.20	4.80	1.80	Iris-versicolor	Iris-versicolor	0.6
84	6.00	2.20	4.00	1.00	Iris-versicolor	Iris-versicolor	0.95
85	6.00	2.20	5.00	1.50	Iris-virginica	Iris-versicolor	0.493333
86	6.00	2.70	5.10	1.60	Iris-versicolor	Iris-virginica	0.364762
87	6.00	2.90	4.50	1.50	Iris-versicolor	Iris-versicolor	1
88	6.00	3.00	4.80	1.80	Iris-virginica	Iris-virginica	0.3
89	6.00	3.40	4.50	1.60	Iris-versicolor	Iris-versicolor	1
90	6.10	2.60	5.60	1.40	Iris-virginica	Iris-virginica	0.11
91	6.10	2.80	4.00	1.30	Iris-versicolor	Iris-versicolor	0.91
92	6.10	2.80	4.70	1.20	Iris-versicolor	Iris-versicolor	0.91
93	6.10	2.90	4.70	1.40	Iris-versicolor	Iris-versicolor	1
94	6.10	3.00	4.60	1.40	Iris-versicolor	Iris-versicolor	1
95	6.10	3.00	4.90	1.80	Iris-virginica	Iris-virginica	0.131429
96	6.20	2.20	4.50	1.50	Iris-versicolor	Iris-versicolor	0.95
97	6.20	2.80	4.80	1.80	Iris-virginica	Iris-virginica	0.2
98	6.20	2.90	4.30	1.30	Iris-versicolor	Iris-versicolor	1
99	6.20	3.40	5.40	2.30	Iris-virginica	Iris-virginica	0
100	6.30	2.30	4.40	1.30	Iris-versicolor	Iris-versicolor	0.95

Out[13]:

Rows: 1-100 of 150 | Columns: 7

The vDataFrame has its own 'score' method to evaluate your models.

In [14]:

iris.score("Species", "pred_Species", method = "accuracy")

Out[14]:

0.986666666666667

Let's examine our prediction with a histogram.

In [15]:

iris.hist(["pred_Species", "Species"])

Some multiclass classifiers let you evaluate the importance of each feature with the 'features_importance' method.

In [16]:

model.features_importance()

	importance
petallengthcm	61.99
petalwidthcm	30.42
sepallengthcm	5.58
sepalwidthcm	2.02

Out[16]:

Let's draw a tree diagram.

In [17]:

model.plot_tree()

--------------------------------------------------------------------------------
Tree Id: 0
Number of Nodes: 17
Tree Depth: 5
Tree Breadth: 9
--------------------------------------------------------------------------------
[1] (petallengthcm < 1.921875 ?)
├── [2] => Iris-setosa (variance = 1.0)
└── [3] (petallengthcm < 4.871875 ?)
    ├── [6] (petallengthcm < 4.687500 ?)
    │   ├── [12] => Iris-versicolor (variance = 1.0)
    │   └── [13] (sepallengthcm < 6.212500 ?)
    │       ├── [26] (petalwidthcm < 1.450000 ?)
    │       │   ├── [52] => Iris-versicolor (variance = 1.0)
    │       │   └── [53] => Iris-virginica (variance = 1.0)
    │       └── [27] => Iris-versicolor (variance = 1.0)
    └── [7] (petalwidthcm < 1.750000 ?)
        ├── [14] (petallengthcm < 5.240625 ?)
        │   ├── [28] (sepallengthcm < 6.325000 ?)
        │   │   ├── [56] => Iris-virginica (variance = 0.666666666666667)
        │   │   └── [57] => Iris-versicolor (variance = 1.0)
        │   └── [29] => Iris-virginica (variance = 1.0)
        └── [15] => Iris-virginica (variance = 1.0)