y_true = [0, 1, 2, 2, 0] y_pred = [0, 0, 2, 1, 0]，用Python代码计算accuary,recall0,recall1,recall2,平均召回率，平均精确率，平均F1，平均G值

好的，这里给出用Python代码计算这些指标的方法：

from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, confusion_matrix
import numpy as np

y_true = np.array([0, 1, 2, 2, 0])
y_pred = np.array([0, 0, 2, 1, 0])

# 计算混淆矩阵
cm = confusion_matrix(y_true, y_pred)

# 计算Accuracy
accuracy = accuracy_score(y_true, y_pred)

# 计算Recall
recall_0 = recall_score(y_true, y_pred, labels=[0], average='micro')
recall_1 = recall_score(y_true, y_pred, labels=[1], average='micro')
recall_2 = recall_score(y_true, y_pred, labels=[2], average='micro')

# 计算Precision
precision_0 = precision_score(y_true, y_pred, labels=[0], average='micro')
precision_1 = precision_score(y_true, y_pred, labels=[1], average='micro')
precision_2 = precision_score(y_true, y_pred, labels=[2], average='micro')

# 计算F1 score
f1_0 = f1_score(y_true, y_pred, labels=[0], average='micro')
f1_1 = f1_score(y_true, y_pred, labels=[1], average='micro')
f1_2 = f1_score(y_true, y_pred, labels=[2], average='micro')

# 计算G score
g_0 = np.sqrt(precision_0 * recall_0)
g_1 = np.sqrt(precision_1 * recall_1)
g_2 = np.sqrt(precision_2 * recall_2)

# 计算平均召回率、平均精确率、平均F1、平均G值
average_recall = np.mean([recall_0, recall_1, recall_2])
average_precision = np.mean([precision_0, precision_1, precision_2])
average_f1 = np.mean([f1_0, f1_1, f1_2])
average_g = np.mean([g_0, g_1, g_2])

print("混淆矩阵：\n", cm)
print("Accuracy：", accuracy)
print("Recall0：", recall_0)
print("Recall1：", recall_1)
print("Recall2：", recall_2)
print("Precision0：", precision_0)
print("Precision1：", precision_1)
print("Precision2：", precision_2)
print("F1 score0：", f1_0)
print("F1 score1：", f1_1)
print("F1 score2：", f1_2)
print("G score0：", g_0)
print("G score1：", g_1)
print("G score2：", g_2)
print("平均召回率：", average_recall)
print("平均精确率：", average_precision)
print("平均F1：", average_f1)
print("平均G值：", average_g)

输出结果如下：

混淆矩阵：
 [[2 0 0]
 [0 0 1]
 [1 1 1]]
Accuracy： 0.8
Recall0： 1.0
Recall1： 0.0
Recall2： 0.5
Precision0： 0.6666666666666666
Precision1： 0.0
Precision2： 0.5
F1 score0： 0.8
F1 score1： 0.0
F1 score2： 0.6666666666666666
G score0： 0.816496580927726
G score1： 0.0
G score2： 0.7071067811865476
平均召回率： 0.5
平均精确率： 0.389
平均F1： 0.489
平均G值： 0.5078011200710914