Replacing values in a df between certain values (replace >1 to 4 with 1)

Question

I would like to replace certain value-thresholds in a df with another value. For example all values between 1 and <3.3 should be summarized as 1. After that all values between >=3.3 and <10 should be summarized as 2 and so on.

I tried it like this: tndf is my df and tnn the column

tndf.loc[(tndf.tnn < 1), 'tnn'] = 0
tndf.loc[((tndf.tnn >= 1) | (tndf.tnn < 3.3)), 'tnn'] = 1
tndf.loc[((tndf.tnn >=3.3) | (tndf.tnn < 10)), 'tnn'] = 2
tndf.loc[((tndf.tnn >=10) | (tndf.tnn < 20)), 'tnn'] = 3
tndf.loc[((tndf.tnn >=20) | (tndf.tnn < 33.3)), 'tnn'] = 4
tndf.loc[((tndf.tnn >=33.3) | (tndf.tnn < 50)), 'tnn'] = 5
tndf.loc[((tndf.tnn >=50) | (tndf.tnn < 100)), 'tnn'] = 6
tndf.loc[(tndf.tnn == 100), 'tnn'] = 7

But every value at the end will be summarized as a 6. I think that's why because of the second part of each condition. But I don't know how to tell the program to only look in a specific range (for example from >=3.3 and <10).

Does that answer your question stackoverflow.com/questions/68195381/…? — FloLie
– FloLie, Commented Jul 6, 2021 at 12:33
Ah, thank you. So, if I am seeing this correctly then I only have to change | to &. — Neihilrach
– Neihilrach, Commented Jul 6, 2021 at 12:41

Orlando · Accepted Answer · 2021-07-06 12:50:19Z

1

i will use np.where() here is the documentation: np.where()

import numpy as np
tnddf0=np.where((tndf.tnn < 1),0,"tnn")
tnddf1=np.where(((tndf.tnn >= 1) & (tndf.tnn < 3.3)),1,"tnn")
#and so on....

answered Jul 6, 2021 at 12:50

Orlando

1341 silver badge10 bronze badges

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

Comments

Vishnudev Krishnadas · Accepted Answer · 2021-07-06 13:04:03Z

1

To form categories like these use pd.cut

pd.cut(df.tnn, [0, 1, 3.3, 10, 20, 33.3, 50, 100], right=False, labels=range(0, 7))

Sample output of pd.cut

         tnn cat
0  76.518227   6
1  44.808386   5
2  46.798994   5
3  70.798699   6
4  67.301112   6
5  13.701745   3
6  47.310570   5
7  74.048936   6
8  37.904632   5
9  38.617358   5

OR

Use np.select. It is meant exactly for your use-case.

conditions = [tndf.tnn < 1, (tndf.tnn >= 1) | (tndf.tnn < 3.3)]
values = [0, 1]
np.select(conditions, values, default="unknown")

edited Jul 6, 2021 at 13:04

answered Jul 6, 2021 at 12:52

Vishnudev Krishnadas

11k2 gold badges29 silver badges58 bronze badges

Collectives™ on Stack Overflow

Replacing values in a df between certain values (replace >1 to 4 with 1)

2 Answers 2

Comments

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related