17

i have a table in my pandas dataframe. df

id count price
1    2     100
2    7      25
3    3     720
4    7     221
5    8     212
6    2     200

i want to create a new dataframe(df2) from this, selecting rows where count is 2 and price is 100,and count is 7 and price is 221

my output should be df2 =

id count price
1    2     100
4    7     221

i am trying using df[df['count'] == '2' & df['price'] == '100']

but getting error

TypeError: cannot compare a dtyped [object] array with a scalar of type [bool]
Shubham R
  • 7,382
  • 18
  • 53
  • 119

1 Answers1

19

You nedd add () because & has higher precedence than ==:

df3 = df[(df['count'] == '2') & (df['price'] == '100')]
print (df3)
  id count price
0  1     2   100

If need check multiple values use isin:

df4 = df[(df['count'].isin(['2','7'])) & (df['price'].isin(['100', '221']))]
print (df4)
  id count price
0  1     2   100
3  4     7   221

But if check numeric, use:

df3 = df[(df['count'] == 2) & (df['price'] == 100)]
print (df3)

df4 = df[(df['count'].isin([2,7])) & (df['price'].isin([100, 221]))]
print (df4)
jezrael
  • 822,522
  • 95
  • 1,334
  • 1,252
  • but how do i join two results in 1 df.? i was thinking of creating two different dataframe for my two rows that i want and then append them? is there any other way to do it? – Shubham R Nov 30 '16 at 10:05
  • Do you think assign to `df3` or `df4` ? – jezrael Nov 30 '16 at 10:09