Nik piepenbreier datagy. Io 31 tips and tricks for super stars

Date
Region
Type
Units
Sales
0
2020-07-11
East
Children's 
Clothing
18
306
1
2020-09-23
North
Children's 
Clothing
14
448
2
2020-04-02
South
Women's 
Clothing
17
425
3
2020-02-28
East
Children's 
Clothing
26
832
4
2020-03-19
West
Women's 
Clothing
3
33
...
...
...
...
...
...
995
2020-02-11
East
Children's 
Clothing
35
735
996
2020-12-25
North
Men's Clothing
NaN
1155
997
2020-08-31
South
Men's Clothing
13
208
998
2020-08-23
South
Women's 
Clothing
17
493
999
2020-08-17
North
Women's 
Clothing
25
300
1000 rows × 5 columns 
Nik Piepenbreier - 
​
datagy.io
 

---

43 
Let's see how often each type appears: 
In [60]: 
df[
​
'Region'
​
].value_counts()
Out[60]: 
East
​
411
North
​
316
South
​
137
West
​
136
Name: Region, dtype: int64
If you want to turn these into proportions, you just need to use the 
normalize=True argument: 
In [61]: 
df[
​
'Region'
​
].value_counts(normalize=
​
True
​
)
Out[61]: 
East
​
0.41
North
​
0.32
South
​
0.14
West
​
0.14
Name: Region, dtype: float64
Nik Piepenbreier - 
​
datagy.io
 

---

44 
Tip #31: Sort Your Dataframes 
It can be helpful to sort your dataframes. 
The sort_values() function is well known. But did you know you can apply 
multiple layers of sort to it? 
In [62]: 
df = 
pd.read_excel(
​
'https://github.com/datagy/pivot_table_pandas/raw/ma
ster/sample_pivot.xlsx'
​
) 
df.head(
​
10
​
)
Out[62]: 
Date
Region
Type
Units
Sales
0
2020-07-11
East
Children's 
Clothing
18
306
1
2020-09-23
North
Children's 
Clothing
14
448
2
2020-04-02
South
Women's 
Clothing
17
425
3
2020-02-28
East
Children's 
Clothing
26
832
4
2020-03-19
West
Women's 
Clothing
3
33
5
2020-02-05
North
Women's 
Clothing
33
627
6
2020-01-24
South
Women's 
Clothing
12
396
7
2020-03-25
East
Women's 
Clothing
29
609
8
2020-01-03
North
Children's 
Clothing
18
486
9
2020-11-03
East
Children's 
Clothing
34
374
Nik Piepenbreier - 
​
datagy.io
 

---

45 
Let's sort this dataframe by region and then by number of units sold: 
In [63]: 
df.sort_values(by=[
​
'Region'
​
, 
​
'Units'
​
],ascending=
​
False
​
).head(
​
10
​
)
Out[63]: 
Date
Region
Type
Units
Sales
534
2020-05-17
West
Men's Clothing
35
520
972
2020-09-21
West
Men's Clothing
35
437
755
2020-09-28
West
Men's Clothing
34
182
948
2020-01-27
West
Children's 
Clothing
34
510
280
2020-10-31
West
Women's 
Clothing
33
196
533
2020-05-14
West
Men's Clothing
33
840
844
2020-05-14
West
Men's Clothing
33
308
206
2020-08-05
West
Children's 
Clothing
32
640
244
2020-03-02
West
Men's Clothing
32
56
481
2020-02-02
West
Women's 
Clothing
32
558
Nik Piepenbreier - 
​
datagy.io
 

---

46 
Bonus Tip: Read Multiple Files at Once 
Do you have multiple files you need to read into a single dataframe? 
You could create seperate dataframes first - or you could use the glob library 
to read multiple files into a dataframe at the same time. 
In [64]: 
import
​
glob 
file_path = 
​
#path to your files here
df = pd.concat([pd.read_csv(file) 
​
for
​
file 
​
in
glob.glob(
​
f"
​
{file_path}
​
*.csv"
​
)], ignore_index = 
​
True
​
)
We use a list comprehension to read our csv files and then concatenate them 
using the concat function. 
You can also use the read_excel() function if you're dealing with Excel files! 
 
Conclusion 
Thanks so much for reading! I hope you enjoyed the tips and that you learned 
something new! 
Let me know what your favorite tip is over Twitter (​
@datagyio​
) or via email 
(​
[e-mail]​
). 
Be sure to check out ​
https://datagy.io​
for more Python tips, tricks, and 
thorough tutorials! 
Nik Piepenbreier - 
​
datagy.io