CODEWITHSUNDEEP
pythonpandasnumpy
sym
sym

Reputation: 77

How to select all rows in a dataframe between two values of column

I have a Dataframe like this: 

import numpy as np
import pandas as pd

df=pd.DataFrame([['travail','hk','hj'],['test',6,6],[5,5,8],[4,3,1],['moyenne',5,6],[5,6,7],
         [1,2,3],['travail','test','kkj'],[5,'hjjd',8],['moyenne',6,7],[5,5,8],[4,3,1],['hkk',5,6],[5,5,8],
         [7,8,5]],columns=['A','B','C'])

I want to select all line between travail and moyennee in column A and obtain :

         A     B    C
0   travail    hk   hj
1      test     6    6
2         5     5    8
3         4     3    1
4   moyenne     5    6
7   travail  test  kkj
8         5  hjjd    8
9   moyenne     6    7

How can I do that?

Upvotes: 3

Views: 446

Answers (2)

rnso
rnso

Reputation: 24613

One can use for loop with iloc to check each row and append rows within desired blocks to a new empty dataframe: 

newdf = pd.DataFrame(columns=df.columns)
flag = False
for i in range(len(df)): 
    firstval = df.iloc[i,0]
    if firstval == 'travail': 
        newdf = newdf.append(df.iloc[i,:])
        flag = True
    elif firstval == 'moyenne':
        newdf = newdf.append(df.iloc[i,:])
        flag = False
    elif flag==True: 
        newdf = newdf.append(df.iloc[i,:])
print(newdf)

Output: 

         A     B    C
0  travail    hk   hj
1     test     6    6
2        5     5    8
3        4     3    1
4  moyenne     5    6
7  travail  test  kkj
8        5  hjjd    8
9  moyenne     6    7

Upvotes: 1

jezrael
jezrael

Reputation: 863431

Compare column by Series.eq (==), second change ordering by Series.iloc, get Series.cumsum and compare again by Series.gt (>), chain mask by & for bitwise AND, last filter by boolean indexing:

m1 = df['A'].eq('travail').cumsum().gt(0)
m2 = df['A'].eq('moyenne').iloc[::-1].cumsum().gt(0)

df1 = df[m1 & m2]
print (df1)
         A     B    C
0  travail    hk   hj
1     test     6    6
2        5     5    8
3        4     3    1
4  moyenne     5    6
5        5     6    7
6        1     2    3
7  travail  test  kkj
8        5  hjjd    8
9  moyenne     6    7

If exist always both values in column A is possible use Series.idxmax with DataFrame.loc:

a = df['A'].eq('travail').idxmax()
b = df['A'].eq('moyenne').iloc[::-1].idxmax()

df1 = df.loc[a:b]

Upvotes: 1

Related Questions