when trying to train a CNN with randomly selected small subsets (each same size) of the training data set, I get different results in accuracy (the accuracy varies from 0.75 to 0.85).

I determine the subsets by the following method by varying the value for random_state, e.g. random_state = 0, 7, 42,.....

X_train, X_test, y_train, y_test = train_test_split(X, y, train_size=0.1, random_state=..., stratify=y)

Would it be correct to calculate the average of the different accuracies (with standard deviation) in order to get a real estimation? Would this be a correct scientific approach?

I hope and look forward for any help and would be very grateful.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.