pandas DataFrame replace() by using regex

Question

I'm having a pandas DataFrame df. I want to replace ↑(a space after ↑) with +, and ↓(a space after ↓) with -. For example, df.a[0](values ↑ 0.69%) replace with +0.69%.

df['last_month'] = df['last_month'].replace(r"↑ ","")is not right. Why?

data = [{"a":"↑ 0.69%","b":"↓ 9.93%"},{"a":"↓ 0.46%","b":"↑ 3.3%"},{"a":"↓ 0.78%","b":"↓ 3.43%"}]
df = pd.DataFrame(data)
df

    a         b
0   ↑ 0.69%   ↓ 9.93%
1   ↓ 0.46%   ↑ 3.3%
2   ↓ 0.78%   ↓ 3.43%

In my raw data, ↑ is an unicode, so it didn't work. In the demo data, ↑ is a str(bytes), so df['last_month'] = df['last_month'].replace(r"↑ ","") works actually like MaxU's. But how to replace when DataFrame values are unicode?

not right? can you create a minimal reproducible example? and this is not regex this is string replacement. — Jean-François Fabre
– Jean-François Fabre ♦, Commented Nov 27, 2016 at 9:02

MaxU - stand with Ukraine · Accepted Answer · 2016-11-27 11:27:02Z

4

IIUC:

In [28]: df.replace(['↑\s*', '↓\s*'], ['+', '-'], regex=True)
Out[28]:
        a       b
0  +0.69%  -9.93%
1  -0.46%   +3.3%
2  -0.78%  -3.43%

For Python 2x:

In [80]: %paste
data = [{"a":u"↑ 0.69%","b":u"↓ 9.93%"},{"a":u"↓ 0.46%","b":u"↑ 3.3%"},{"a":u"↓ 0.78%","b":u"↓ 3.43%"}]
df = pd.DataFrame(data)
df
## -- End pasted text --
Out[80]:
         a        b
0  ↑ 0.69%  ↓ 9.93%
1  ↓ 0.46%   ↑ 3.3%
2  ↓ 0.78%  ↓ 3.43%

In [81]: %paste
df = df.replace([u'↑\s*', u'↓\s*'], [u'+', u'-'], regex=True)
print(df)
## -- End pasted text --
        a       b
0  +0.69%  -9.93%
1  -0.46%   +3.3%
2  -0.78%  -3.43%

edited Nov 27, 2016 at 11:27

answered Nov 27, 2016 at 9:06

MaxU - stand with Ukraine

212k37 gold badges402 silver badges436 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

running man Over a year ago

df.replace(['↑\s*', '↓\s*'], ['+', '-'], regex=True) works if ↑is a bytes but not an unicode? What if ↑is an unicode?

running man Over a year ago

I am using python 2.7.12 @MaxU

piRSquared · Accepted Answer · 2016-11-27 09:07:30Z

2

you can stack then unstack with the str accessor.

df.stack().str.replace("↑ ","+").str.replace("↓ ", "-").unstack()

answered Nov 27, 2016 at 9:07

piRSquared

296k68 gold badges509 silver badges654 bronze badges

Comments

running man · Accepted Answer · 2016-11-27 11:09:49Z

0

I got it, df.replace([u'↑ ', u'↓ '], [u'+', u'-'], regex=True) works.

answered Nov 27, 2016 at 11:09

running man

1,4774 gold badges15 silver badges23 bronze badges

Collectives™ on Stack Overflow

pandas DataFrame replace() by using regex

3 Answers 3

2 Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

2 Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related