How can I make my code to find the minimum faster?

Question

I have a question about how to find minimum in a list in my way:

I have a large list where every record is like this:

['22:00:19', '0026f88e557225333f01', '23', '37', '', '176.2', '0', '60', 'SOMETHING', 
 3.318717958567554e-05]

But the first and the last records in the list don't contain the last number:

For example:

['22:00:09', '0026f88e557225333f01', '23', '37', '', '176', '0', '60', 'SOMETHING']

I need to find the min of that last column 3.318717958567554e-05 and its index in every call of my function.

Here is my code:

def find_min(data, size, num):

    for index, i in enumerate(data):

        if index == 0 or index == data.__len__() -1: continue

        if index == 1:
            minimum = float(i[9])
            idx = index
            continue

        if float(i[9]) < minimum or float(i[9]) < num:
            minimum = float(i[9])
            idx = index

    return idx, minimum

num is a user-defined threshold which is used to calculate min. (Min should be less than that.) This code works properly and I find what I want but how can I make my code faster because I call this function thousand times and I work with a huge dataset and as a result the execution time is very much because of that slow function.

Answer 1

Get rid of all the if statements and slice the array to just the ones you care about.

Convert i[9] to float just once.

def find_min(data, size, num):
    idx = 1
    minimum = float(data[1][9])
    for index, i in enumerate(data[2:-1]):        
        f = float(i[9])
        if f < minimum or f < num:
            minimum = f
            idx = index + 2 # +2 because of the slicing
    
    return idx, minimum

Or if the list is so large that making a slice of it is too expensive, just iterate over indexes:

def find_min(data, size, num):
    idx = 1
    minimum = float(data[1][9])
    for index in range(2, len(data)-1):     
        f = float(data[index][9])
        if f < minimum or f < num:
            minimum = f
            idx = index
    
    return idx, minimum