Pandas Select a value in a specific column based on the index label

Question

So the question actually is the following:

I create the (straight copy paste from pandas documentation) dataframe:

d = {'col1': [1, 2], 'col2': [3, 4]}
df = pd.DataFrame(data=d)

This creates a a DataFrame like this: (where the left-most column is acutally the index)

 col1  col2
0     1     3
1     2     4

Then I want to assign my name x the value of the col2 and row1, which is 4. But I want to do this based on the value in col1 (cause thats the one I know). So that it looks more or less like this:

x = df.loc[df.col2 == df.loc[df.col1 == 2]]

I know this makes no sense, but maybe someone understands the issue and comes up with a hint:) Probably this is super easy to do, but I just can wrap my head around all the .loc and .iloc possibilities at the moment

The "real" problem is to write a class which saves the coordinates (columns lat, lon) based on the provided aiport name: (and this still is kind of confusing)

def __init__(self, name):
        import pandas as pd
        alleDaten = pd.read_csv("airports.dat", delimiter=",", usecols=[1,6,7])
        alleDaten.columns = ["name","lat","lon"] #rename columns
        self._name = alleDaten.loc[alleDaten.name == "name"] #get name based on the provided name of the airport
        self._x = alleDaten.loc[alleDaten.name == "name", ["lat"]] # get lat based on the provided name
        self._y = alleDaten.loc[alleDaten.name == "name", ["lon"]] # get lon based on the provided name

Thanks a lot for any help:)!!

df.iloc[1,1] would give second column, second row value. Or if you want col2 value where col1=2, then x=df.loc[df.col1==2, 'col2'].iloc[0] — Suraj Motaparthy, Oct 23 '19 at 15:08
yeah, but in case you just know the value for one column, but want to assign the variable the value of another column of the same row:/ Hope it makes sense... — Robin Kohrs, Oct 23 '19 at 15:11
awesome, I needed to long typing my reponse! I'll try these variations out:) Thanks a lot — Robin Kohrs, Oct 23 '19 at 15:12

score 4 · Answer 1 · answered Oct 23 '19 at 15:12

4

This should do the trick for you (to get only value of filtered pandas):

x = df.loc[df.col1 == 2, "col2"]
print(x.to_list()[0])

Output:

4

[Program finished]

answered Oct 23 '19 at 15:12

Grzegorz Skibinski

12,624
2
11
34

Great, thank you very much! – Robin Kohrs Oct 23 '19 at 15:37
Any idea of how not to return a DataFrame-Object and instead an float? casting it just an float seems not to work – Robin Kohrs Oct 23 '19 at 15:53
Yup, as per the above - ```x.to_list()[0]```. Then you'll get back just the first value of the returned result. – Grzegorz Skibinski Oct 23 '19 at 17:04
thanks:) Now it worked – Robin Kohrs Oct 23 '19 at 21:43

score 0 · Answer 2 · answered Oct 23 '19 at 15:10

0

x = df.iloc[1, 1]
The numbering system starts at 0, so [1, 1] will select the second row and second column.

answered Oct 23 '19 at 15:10

Nicko

350
2
9

Pandas Select a value in a specific column based on the index label

2 Answers2