Summary of 5_Default_NeuralNetwork

Neural Network

dense_1_size: 32
dense_2_size: 16
learning_rate: 0.05
num_class: 3
explain_level: 2

Validation

validation_type: split
train_ratio: 0.75
shuffle: True
stratify: True

Optimized metric

logloss

Training time

1.7 seconds

Metric details

	0	1	2	accuracy	macro avg	weighted avg	logloss
precision	1	0.909091	0.8125	0.894737	0.907197	0.904755	0.29232
recall	0.916667	0.769231	1	0.894737	0.895299	0.894737	0.29232
f1-score	0.956522	0.833333	0.896552	0.894737	0.895469	0.893862	0.29232
support	12	13	13	0.894737	38	38	0.29232

Confusion matrix

	Predicted as 0	Predicted as 1	Predicted as 2
Labeled as 0	11	1	0
Labeled as 1	0	10	3
Labeled as 2	0	0	13

Learning curves

Permutation-based Importance