C3W3 UNQ_C4 - the last 2 lines

Course Q&A AI For Medicine AI For Medical Treatment
1

So at the step where we have to reuse the metric function in Excercise 4,
I have tried to play around with the metric function but almost every time I get a type of error

    # Loop through each sample
    for i in range(len(feature_performance_arr)): # complete this line
        
        # permute the column of dataframe X
        perm_X = np.random.permutation(X[feature])
        print(perm_X.shape)
        # calculate the performance with the permuted data
        # Use the same metric function that was used earlier
        print(perm_X.reshape(-1,1).shape)
        print(y.shape)
        feature_performance_arr[i] = metric(y,model.predict_proba(perm_X.reshape(-1,1))[i, 1])

so I think the main issue is with my metric function down there but haven’t figured out the fix. Currently, this is the error I get “ValueError: X has 1 feature per sample; expecting 2”

but I thought the whole point was to get the performance of one feature at a time for N num_samples
what am I missing here?

2

looks like I might have figured it out