解决sklearn 随机森林数据不平衡的方法

Posted 2021-02-11 allen-rg

Handle Imbalanced Classes In Random Forest

 

Preliminaries

# Load libraries
from sklearn.ensemble import RandomForestClassifier
import numpy as np
from sklearn import datasets

Load Iris Flower Dataset

# Load data
iris = datasets.load_iris()
X = iris.data
y = iris.target

Adjust Iris Dataset To Make Classes Imbalanced

# Make class highly imbalanced by removing first 40 observations
X = X[40:,:]
y = y[40:]

# Create target vector indicating if class 0, otherwise 1
y = np.where((y == 0), 0, 1)

Train Random Forest While Balancing Classes

When using RandomForestClassifier a useful setting is class_weight=balanced wherein classes are automatically weighted inversely proportional to how frequently they appear in the data. Specifically:

$${\mathrm{wj=n/knj}}^{}$$

where ${\mathrm{wj is\; the\; weight\; to\; class j, nn is\; the\; number\; of\; observations, nj is\; the\; number\; of\; observations\; in\; class j,\; and k is\; the\; total\; number\; of\; classes.}}^{}$

# Create decision tree classifer object
clf = RandomForestClassifier(random_state=0, n_jobs=-1, class_weight="balanced")

# Train model
model = clf.fit(X, y)

https://chrisalbon.com/machine_learning/trees_and_forests/handle_imbalanced_classes_in_random_forests/



类别不平衡处理方法：
https://segmentfault.com/a/1190000015248984