如何从数据框中删除所有列值为零或布尔值为false的行

Question

How to drop the rows from dataframe that has all column values as zero using pandas? In the dataframe below I use (df.ix[:,'a':] == 0) but how do I then vectorize testing all Boolean values for each row are False?

df = pd.DataFrame({'a':[1,0,1,0], 'b':[1,0,0,0], 'c':[1,0,1,0], 'd':[1,0,0,0]}, index=['aa','bb','cc','dd'])
df.index.name = 'name'

      a  b  c  d
name            
aa    1  1  1  1
bb    0  0  0  0
cc    1  0  1  0
dd    0  0  0  0

I want to drop bb and dd rows from dataframe, as all the column values are empty.Theres is way to drop columns in pandas but nothing for rows.

Answer 1

You can use boolean indexing with inverted mask comparing all columns from a with 0 , where all values in a row are True by all(axis=1) :

mask = ~(df.ix[:,'a':] == 0).all(axis=1)
print (mask)
0     True
1    False
2     True
3    False
dtype: bool

print (df[mask])
  name  a  b  c  d
0   aa  1  1  1  1
2   cc  1  0  1  0

Another solution with checking at least one 1 return same output:

print ((df.ix[:,'a':]).any(1))
0     True
1    False
2     True
3    False
dtype: bool

print (df[(df.ix[:,'a':]).any(1)])
  name  a  b  c  d
0   aa  1  1  1  1
2   cc  1  0  1  0

If name is index name and first column is index:

print (df[~(df == 0).all(1)])
      a  b  c  d
name            
aa    1  1  1  1
cc    1  0  1  0

Or:

print (df[df.any(1)])
      a  b  c  d
name            
aa    1  1  1  1
cc    1  0  1  0

Answer 2

Use any() , which returns True iff there's any True value (ie a value different that 0).

df = df[df.any(axis=1)]

As for your example, before:

In[1]: df
Out[1]:       a  b  c  d
name            
aa    1  1  1  1
bb    0  0  0  0
cc    1  0  1  0
dd    0  0  0  0

And after:

      a  b  c  d
name            
aa    1  1  1  1
cc    1  0  1  0