Cross Validation, Data Science

I have written this code and getting error:

from sklearn.model_selection import train_test_split

X0 = pd.read_csv('C:/Users/Desktop/Dataset for Experiment - 2019/ANFIS modeling data/china_data.csv')

train_test_split. X0 = X0.drop(columns = ['EFFORT'], axis = 1)

X_train, X_test, y_train, y_test = train_test_split(X0, test_size=0.8)

**the error which is showing for this code "X_train, X_test, y_train, y_test = train_test_split(X0, test_size=0.8)

ValueError: not enough values to unpack (expected 4, got 2)"**

I tried many times to remove this error but it is still persistent. Can anybody explain to me why I am getting this error & how can I remove this error?

Answer 1

[Hiro Yoshino]

In MATLAB, you can use cvpartition

https://www.mathworks.com/help/stats/cvpartition.html?searchHighlight=cvpartition&s_tid=doc_srchtitle

Train vs Test

pt = cvpartition(Y, 'HoldOut', 0.2);
trainIndex = training(pt);
testIndex = test(pt);

X_train = X(:,trainIndex);
Y_train  = Y(trainIndex);
X_test = X(:,testIndex);
Y_test  = Y(testIndex);

Cross validation

pt = cvpartition(Y, 'KFold', 5);
...
...

It is easier to control the data index for validation.