CODEWITHSUNDEEP
pythonpandasnumpy
kittygirl
kittygirl

Reputation: 2443

How to find the index of the not-null item which not in a list?

My system is python3.6, with numpy 1.16.2,scipy 1.2.1,matplotlib 3.0.3 

import pandas as pd
import numpy
df=pd.DataFrame({'col1':['a','b','c'],'col2':['d',numpy.NaN,'c'],'col3':['c','b','b']})
df = df.astype({"col2": 'category'})
print(df)

output of above script is: 

  col1 col2 col3
0    a    d    c
1    b  NaN    b
2    c    c    b

I want to find index of the not-null item in series col2 whose category is not in ['a','b','c']

In this case, d is not null and is not in ['a','b','c'],then the expect result should be the index of d,which is 0

My solution as blow: 

getindex=numpy.where(~df['col2'].isin(['a','b','c']) & df['col2'].notna())
#if getindex is not empty, print it
if not all(getindex):
    print(getindex)

The output of my solution script is: 

ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()

Upvotes: 1

Views: 384

Answers (2)

jezrael
jezrael

Reputation: 863226

Use:

getindex=df.index[(~df['col2'].isin(['a','b','c']) & df['col2'].notna())]
print (getindex)
Int64Index([0], dtype='int64')

If want select first value with no error if value not exist:

print (next(iter(getindex), 'no match'))
0

If want if empty statement use Index.empty for testing:

if not getindex.empty:
    print (getindex)

Your solution should working if add [0] for select first array from list:

getindex=np.where(~df['col2'].isin(['a','b','c']) & df['col2'].notna())[0]
print (getindex)
[0]

Upvotes: 2

BENY
BENY

Reputation: 323326

Do modify in you if condition 

getindex=np.where(~df['col2'].isin(['a','b','c']) & df['col2'].notna())
if any(~df['col2'].isin(['a','b','c']) & df['col2'].notna()): # change here to any 
    print(getindex)

(array([0], dtype=int64),)

Also base on your word #if getindex is not empty, print it

if len(getindex)!=0:
    print(getindex)

(array([0], dtype=int64),)

Upvotes: 1

Related Questions