CODEWITHSUNDEEP
pythonpandas
hari prasad
hari prasad

Reputation: 25

selecting rows and columns in pandas dataframe

pandas dataframe has , number of items selling quantity

FRUIT   YESTERDAY    TODAY    TOMORROW
Apple     5           5         3
Apple     3           5         3
Orange    6           9         8
Banana    0           0         0
Grapes    7           7         7
Guava     0           3         3
Mango     2           8         2
Mango     4           4         6

The above data is in pandas dataframe , Using IF condition , I need data like , after filtering with below conditions (Cond-1) we have to check any of value of YESTERDAY , TODAY , TOMORROW are equal to zero then exclude that rows.(cond-1A) also checking if YESTERDAY, TODAY, TOMORROW all have same values then exclude such rows. (Like , Excluding Grapes 7 7 7). (Cond-2) after filtering data with cond-1 , we have check condition for YESTERDAY and TODAY are equal then get such rows (cond-3) we have to check condition for YESTERDAY and TOMORROW are equal then display such rows

OUTPUT :

YESTERDAY and TODAY Equal are

FRUIT   YESTERDAY   TODAY   TOMORROW
Apple       5         5       3
Mango       4         4       6

YESTERDAY and TOMORROW are Equal

FRUIT      YESTERDAY    TODAY   TOMORROW
Apple        3           5         3
Mango        2           8         2

Upvotes: 0

Views: 316

Answers (2)

Corralien
Corralien

Reputation: 120509

Create a boolean mask to filter your dataframe:

cols = ['YESTERDAY', 'TODAY', 'TOMORROW']
mask = df[cols].ne(0).all(1) & df[cols].std(axis=1).ne(0)
out = df[mask]

Mask detail:

  • df[cols].ne(0).all(1) keep rows which have no 0 in their columns.
  • df[cols].std(axis=1).ne(0) keep rows which have at least one different value in their columns . A standard deviation of 0 means all values are equals (7, 7, 7).

First output:

>>> out.loc[out['YESTERDAY'] == out['TODAY']]
   FRUIT  YESTERDAY  TODAY  TOMORROW
0  Apple          5      5         3
7  Mango          4      4         6

Second output:

>>> out.loc[out['YESTERDAY'] == out['TOMORROW']]
   FRUIT  YESTERDAY  TODAY  TOMORROW
1  Apple          3      5         3
6  Mango          2      8         2

Upvotes: 0

Scott Boston
Scott Boston

Reputation: 153510

Try using pd.DataFrame.query after masking where dataframe equals to zero:

df.mask(df == 0).query('YESTERDAY == TOMORROW')

Output:

    FRUIT  YESTERDAY  TODAY  TOMORROW
1   Apple        3.0    5.0       3.0
4  Grapes        7.0    7.0       7.0
6   Mango        2.0    8.0       2.0

And,

df.mask(df == 0).query('YESTERDAY == TODAY')

Output:

    FRUIT  YESTERDAY  TODAY  TOMORROW
0   Apple        5.0    5.0       3.0
4  Grapes        7.0    7.0       7.0
7   Mango        4.0    4.0       6.0

Upvotes: 1

Related Questions