Table 3 Hyperparameters used to tune the RF model, their range, and the result of tuning.

From: Diagnosis of disease affecting gait with a body acceleration-based model using reflected marker data for training and a wearable accelerometer for implementation

Hyperparameter

Description

Range

Selected

Criterion

Function used to measure the quality of a split in a decision tree

Entropy, gini

Entropy

max_depth

Maximum depth of each tree in the forest

Range (0, 20)

1

max_features

Number of features to consider when looking for the best split

Auto, sqrt, log2, None

log2

min_samples_leaf

Minimum number of samples required to be at a leaf node

Uniform (0, 0.3)

0.1585

min_samples_split

Minimum number of samples required to split an internal node

Uniform (0, 1)

0.106

n_estimators

Number of trees in the forest

Range (0, 100)

78

Back to article page