Xgboost：如何将预测概率转换为多类标签原始名称？

Question

我正在使用 xgboost multiclass classifier，如下例所示。对于 X_test 数据框中的每一行，模型输出一个列表，列表元素是对应于每个类别 'a'、'b'、'c' 或 'd' 的概率例如[0.44767836 0.2043365 0.15775423 0.19023092].

如何判断列表中的哪个元素对应于哪个 class / 类别（a、b、c 或 d）？我的目标是在数据框 a、b、c、d 上创建 4 个额外的列，并将匹配概率作为每列中的行值。

import numpy as np
import pandas as pd
import xgboost as xgb
import random
from sklearn import preprocessing
from sklearn.model_selection import train_test_split

#Create Example Data
np.random.seed(312)
data = np.random.random((10000, 3))
y = [random.choice('abcd') for _ in range(data.shape[0])]

features = ["x1", "x2", "x3"]
df = pd.DataFrame(data=data, columns=features)
df['y']=y

#Encode target variable
labelencoder = preprocessing.LabelEncoder()
df['y_target'] = labelencoder.fit_transform(df['y'])
    
#Train Test Split    
X_train, X_test, y_train, y_test = train_test_split(df[features], df['y_target'], test_size=0.2, random_state=42, stratify=y)

#Train Model
dtrain = xgb.DMatrix(X_train, label=y_train)
dtest = xgb.DMatrix(X_test, label=y_test)

param = {        'objective':'multi:softprob',
                 'random_state': 20,
                 'tree_method': 'gpu_hist',
                 'num_class':4
                }

xgb_model = xgb.train(param, dtrain, 100)

predictions=xgb_model.predict(dtest)

print(predictions)

Answer 1

预测遵循与列标签相同的顺序 0, 1, 2, 3。要获取原始目标名称，请使用 LabelEncoder.

中的 classes_ 属性

import pandas as pd

pd.DataFrame(predictions, columns=labelencoder.classes_)
>>>     
           a           b           c           d
0       0.133130    0.214460    0.569207    0.083203
1       0.232991    0.275813    0.237639    0.253557
2       0.163103    0.248531    0.114013    0.474352
3       0.296990    0.202413    0.157542    0.343054
4       0.199861    0.460732    0.228247    0.111159
... 
1995    0.021859    0.460219    0.235214    0.282708
1996    0.145394    0.182243    0.225992    0.446370
1997    0.128586    0.318980    0.237229    0.315205
1998    0.250899    0.257968    0.274477    0.216657
1999    0.252377    0.236990    0.221835    0.288798

Xgboost：如何将预测概率转换为多类标签原始名称？

Xgboost: How to convert prediction probabilities to multiclass labels original names?

python

scikit-learn

xgboost

multiclass-classification