Python get observation with nearest value-CodePudding

I have a dataframe that looks like this

    Segment  Distance   ID
    3        0.348      1203
    8        0.439      1204
    4        0.458      1205
    8        0.284      1207
    3        0.359      1208
    8        0.563      1209
    4        0.388      1210

I want to be able to enter id = '1203'in python and get 3 of the nearest observations according to the distance.

So when I enter id = '1203', this is my desired output

    Segment  Distance   ID
    3        0.359      1208
    4        0.388      1210
    8        0.284      1207

I only want my output to be based on the distance variable, not segment, but I want the respective segment number to show up in my output.

I know it's a simple problem, but I am fairly new to python, so I am a little confused on how to approach this. Can someone help me out? Thanks.

CodePudding user response：

Extract the distance from ID '1203' then subtract its distance to the whole dataframe and sort values. Drop the first row (ID '1203') and keep the next three ones:

dist = df.loc[df['ID'] == '1203', 'Distance'][0]
out = df.loc[df['Distance'].sub(dist).abs().sort_values().tail(-1).head(3).index]
print(out)

# Output:
   Segment  Distance    ID
4        3     0.359  1208
6        4     0.388  1210
3        8     0.284  1207

CodePudding user response：

I think a good approach is to sort by 'Distance' col. then get the next three rows with loc function

df.sort_values(by=['Distance'])

df.loc[:'1203'].head(3)