C2_W3_Assignment - 5.1 Simple model

Hi there, in this section assignment says “Our simple model has a little higher classification error on training data but does better on cross-validation data than the more complex model”, however the cross validation error I obtained for both models is same:
categorization error, cv, simple model, 0.122, complex model: 0.122

Therefore, it is hard to say that simple model has a better performance. Is it expected?

Hello @conscell , the code for calculating cv error for simple model has a problem, please change from using model_predict to using model_predict_s (note the additional _s at the end) in the following line:

cv_cerr_simple = eval_cat_err(y_cv, model_predict_s(X_cv))

After the change you should see different errors, and we may discuss further if you still have the same view or other views you may want to discuss.

I am reporting the problem about the code to the staff. Thanks for the question!