Created
August 28, 2018 13:35
-
-
Save mjhea0/64a19b32ffe061ae637eaf598e3b9d2a to your computer and use it in GitHub Desktop.
dask example
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import time | |
import pandas as pd | |
import dask.dataframe as ddf | |
def benchmark(function, repetitions, name): | |
start = time.time() | |
function(repetitions) | |
end = time.time() | |
print('{0} seconds for {1}'.format((end - start), name)) | |
def test_pandas(repetitions): | |
df = pd.read_csv('random_people.csv') | |
df['bonus'] = df['salary'] * .5 | |
df2 = pd.concat([df for _ in range(repetitions)]) | |
return df2.salary.mean() | |
def test_pandas_dask(repetitions): | |
df = pd.read_csv('random_people.csv') | |
df2 = pd.concat([df for _ in range(repetitions)]) | |
dd = ddf.from_pandas(df2, npartitions=3) # why only 3 cores? | |
dd['bonus'] = dd['salary'] * .5 | |
return dd.salary.mean().compute() | |
# 1,000 | |
benchmark(test_pandas, 1000, f'{test_pandas.__name__} 1k') | |
benchmark(test_pandas_dask, 1000, f'{test_pandas_dask.__name__} 1k') | |
print('\n') | |
# 10,000 | |
benchmark(test_pandas, 10000, f'{test_pandas.__name__} 10k') | |
benchmark(test_pandas_dask, 10000, f'{test_pandas_dask.__name__} 10k') | |
print('\n') | |
# 100,000 | |
benchmark(test_pandas, 100000, f'{test_pandas.__name__} 100k') | |
benchmark(test_pandas_dask, 100000, f'{test_pandas_dask.__name__} 100k') |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
name | surname | salary | ||
---|---|---|---|---|
0 | Henry | Joneson | 5000 | |
1 | Albert | Goodman | 10000 | |
2 | William | Goodman | 10000 | |
3 | John | Joneson | 10000 | |
4 | Albert | Black | 10000 | |
5 | Henry | Joneson | 12000 | |
6 | Richard | Green | 5500 | |
7 | Henry | Joneson | 11000 | |
8 | Henry | Goodman | 12000 | |
9 | Albert | Joneson | 11000 | |
10 | William | Joneson | 10000 | |
11 | John | White | 10000 | |
12 | Henry | Black | 11000 | |
13 | Albert | Goodman | 10000 | |
14 | Richard | Green | 5500 | |
15 | Henry | Black | 13500 | |
16 | Richard | White | 11000 | |
17 | Albert | Black | 5500 | |
18 | Henry | Green | 10000 | |
19 | Albert | Joneson | 11000 | |
20 | William | Goodman | 12000 | |
21 | William | Goodman | 5000 | |
22 | John | Green | 9500 | |
23 | John | Black | 13500 | |
24 | Richard | Green | 13500 | |
25 | Henry | Joneson | 12000 | |
26 | Henry | Goodman | 10000 | |
27 | John | Joneson | 9500 | |
28 | Henry | Goodman | 11000 | |
29 | William | Green | 12000 | |
30 | Henry | Goodman | 10000 | |
31 | Richard | Black | 10000 | |
32 | Richard | Joneson | 5500 | |
33 | Richard | Joneson | 5000 | |
34 | Henry | Black | 9500 | |
35 | John | White | 13500 | |
36 | Henry | Green | 11000 | |
37 | John | Black | 5500 | |
38 | William | Green | 12000 | |
39 | Albert | Green | 10000 | |
40 | Richard | Joneson | 9500 | |
41 | William | Joneson | 12000 | |
42 | John | Joneson | 10000 | |
43 | William | Black | 10000 | |
44 | Albert | Black | 12000 | |
45 | John | Goodman | 13500 | |
46 | John | Joneson | 10000 | |
47 | John | Joneson | 9500 | |
48 | Richard | Black | 9500 | |
49 | Albert | White | 10000 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment