CODEWITHSUNDEEP
pythonpython-3.xpandasdataframefiltering
CarlosMorente
CarlosMorente

Reputation: 918

Filter dataframe based on columns without name values

I have a dataframe which I need to filter if the value of two columns for each row is bigger than two specific numbers. Also, it's important that I don't know anything of the dataframe except that those columns are the numbers -2 and -3

For example, we have dataframe = pd.DataFrame(data=[[1,2,3,4],[5,6,7,8],[1,2,-3,4],[5,-6,7,8]]). Imagine that we need to filter it if the -2 and -3 columns are bigger than 0. In this case, we'll return only the first two rows.

I've tried it using return dataframe[(dataframe[:, -3] >= 0) and (dataframe[:, -2] >= 0)], but it gives me the following error: TypeError: unhashable type: 'slice'.

Upvotes: 2

Views: 3517

Answers (2)

jezrael
jezrael

Reputation: 863116

Use iloc for select columns by positions and for botwise and use &:

df = dataframe[(dataframe.iloc[:, -3] >= 0) & (dataframe.iloc[:, -2] >= 0)]
print (df)
   0  1  2  3
0  1  2  3  4
1  5  6  7  8

Upvotes: 5

Dani Mesejo
Dani Mesejo

Reputation: 61910

You could do:

import pandas as pd

dataframe = pd.DataFrame(data=[[1,2,3,4],
                               [5,6,7,8],
                               [1,2,-3,4],
                               [5,-6,7,8]])

mask = (dataframe.iloc[:, [-3, 2]] >= 0).all(axis=1)

result = dataframe[mask]
print(result)

Output

   0  1  2  3
0  1  2  3  4
1  5  6  7  8

Upvotes: 1

Related Questions