runtime.py

import pandas as pd
import numpy as np

# generate 15000 intergeter data points randomly between -1000 and 1000
col1= np.random.randint(-1000,1000,15000)

#dictionary of data
data = {
    "col1" : col1,
}
# Pandas DataFrame
df = pd.DataFrame(data)
# We will test the execution time with 5 loops by multiplying each item of 
# col1 with 153
# ___________________________________________________________________________(1)
%%timeit 5
for i in range(len(df)):
    df.at[i,"multiply_at"] = 153*df.at[i,"col1"]
    
#Time: 113 ms ± 1.53 ms per loop (mean ± std. dev. of 7 runs, 10 loops each)
# ___________________________________________________________________________(2)
%%timeit 5
df['multiply_direct'] = 153 * df["col1"]

#Time: 243 µs ± 4.36 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each)
#____________________________________________________________________________(3)

%%timeit 5
func_multiply = lambda x,y :x*y
df["multiply_lambda"] = func_multiply(153,df["col1"])

#Time: 249 µs ± 8.09 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each)
#____________________________________________________________________________(4)
%%timeit 5
def multiply(row):
    data = row[["col1"]]
    return pd.Series({'multiply_apply': 153*data})

df.apply(multiply,axis=1)

#Time: 10.5 s ± 129 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
#____________________________________________________________________________(5)
%%timeit 5
df['multiply_apply2'] = df.col1.apply(lambda x: 153*x)

#Time 3.23 ms ± 49.7 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)