Vectorized method to find matching values between two columns

Question

I'm trying to locate the most recent rows within my Dataframe that contain the same values in two separate columns.

Presently, I am doing this slowly with looping, but I suspect there's a way to cleverly use the apply method or some other vectorized function to do this faster. My present code:

def enumerate_matching(df):
    a = list(df['A'])
    b = list(df['B'])
    matching = []

    for i in range(0, len(a)-1):
        for j in range(i+1, len(b)):
            if a[i] == b[j]:
                matching.append(i)
                matching.append(i+j)
                break
    return matching

Is there a faster method to do this?

Answer 1

you could use set to get the intersection (it has a complexity logarithmic in the size of the sets a and b)

 a = set(df['A'])
 b = set(df['B'])
 a.intersection(b)

Answer 2

If you want to do the matching line by line, you should do:

np.sum(df['A'] == df['B'])