Using replace efficiently in pandas

Question

I am looking to use the replace function in an efficient way in python3. The code I have is achieving the task, but is much too slow, as I am working with a large dataset. Thus, my priority is efficiency over elegancy whenever there is a tradeoff. Here is a toy of what I would like to do:

import pandas as pd
df = pd.DataFrame([[1,2],[3,4],[5,6]], columns = ['1st', '2nd'])

       1st  2nd
   0    1    2
   1    3    4
   2    5    6


idxDict= dict()
idxDict[1] = 'a'
idxDict[3] = 'b'
idxDict[5] = 'c'

for k,v in idxDict.items():
    df ['1st'] = df ['1st'].replace(k, v)

Which gives

     1st  2nd
   0   a    2
   1   b    4
   2   c    6

as I desire, but it takes way too long. What would be the fastest way?

Edit: this is a more focused and clean question than this one, for which the solution is similar.

Answer 1

use map to perform a lookup:

In [46]:
df['1st'] = df['1st'].map(idxDict)
df
Out[46]:
  1st  2nd
0   a    2
1   b    4
2   c    6

to avoid the situation where there is no valid key you can pass na_action='ignore'

You can also use df['1st'].replace(idxDict) but to answer you question about efficiency:

timings

In [69]:
%timeit df['1st'].replace(idxDict)
%timeit df['1st'].map(idxDict)

1000 loops, best of 3: 1.57 ms per loop
1000 loops, best of 3: 1.08 ms per loop

In [70]:    
%%timeit
for k,v in idxDict.items():
    df ['1st'] = df ['1st'].replace(k, v)

100 loops, best of 3: 3.25 ms per loop

So using map is over 3x faster here

on a larger dataset:

In [3]:
df = pd.concat([df]*10000, ignore_index=True)
df.shape

Out[3]:
(30000, 2)

In [4]:    
%timeit df['1st'].replace(idxDict)
%timeit df['1st'].map(idxDict)

100 loops, best of 3: 18 ms per loop
100 loops, best of 3: 4.31 ms per loop

In [5]:    
%%timeit
for k,v in idxDict.items():
    df ['1st'] = df ['1st'].replace(k, v)

100 loops, best of 3: 18.2 ms per loop

For 30K row df, map is ~4x faster so it scales better than replace or looping

Answer 2

While map is indeed faster, replace was updated in version 19.2 (details here) to improve its speed making the difference significantly less:

In [1]:
import pandas as pd


df = pd.DataFrame([[1,2],[3,4],[5,6]], columns = ['1st', '2nd'])
df = pd.concat([df]*10000, ignore_index=True)
df.shape

Out [1]:
(30000, 2)

In [2]:
idxDict = {1:'a', 3:"b", 5:"c"}
%timeit df['1st'].replace(idxDict, inplace=True)
%timeit df['1st'].update(df['1st'].map(idxDict))

Out [2]:
100 loops, best of 3: 12.8 ms per loop
100 loops, best of 3: 7.95 ms per loop

Additionally, I modified EdChum's code for map to include update, which, while slower, prevents values not included in an incomplete map from being changed to nans.