将Y转换为指示矩阵是可行的:

file = 'dataset.csv'    X, Y = readFile(file)N = len(Y)T = np.zeros((N, K)) for i in range(N):    T[i, Y[i]] = 1 

但是当我将数据集划分为训练集和测试集后，执行同样的操作时，如下所示，会出现错误:

X, Y = shuffle(X, Y)Ntrain = int(0.7*len(X))Xtrain, Ytrain = X[:Ntrain], Y[:Ntrain]Xtest, Ytest = X[Ntrain:], Y[Ntrain:]N1 = len(Ytrain)T1 = np.zeros((N1, K))  for i in range(N1):    T1[i, Ytrain[i]] = 1 

在最后一行T1[i, Ytrain[i]] = 1 处显示错误，我哪里做错了？K是类别的数量，等于9

print(np.unique(Y))print(np.unique(Ytrain))

上述打印语句显示如下:

[0 1 2 3 4 5 6 7 8][0 1 2 3 4 5 6 7 8]

回答：

这个问题已经解决了:不知为何Ytrain中没有索引。所以我将Ytrain转换为pandas系列:

data = np.array(Ytrain)Ytrain1 = pd.Series(data)N1 = len(Ytrain1)T1 = np.zeros((N1, K))for i in range(N1):    print(i, Ytrain1[i]) # 打印正常    T1[i, Ytrain1[i]] = 1