Iterating a data-frame column: TypeError: 'float' object is not subscriptable

Question

I have a data-frame (df) with a column called Id which looks like

The column has a type: dtype: object I have worked out the maximum Id value and assigned to a variable called maxId (which is 678 and am looking to apply a sequentially increasing maxId to the empty elements so in this example my output would be:

Where element 3 and 53 are assigned values of 679 and 680 respectively.

I have tried the following code where i loop through the column looking for null elements and then applythuing the maxId to these:

for item, frame in df['Id'].iteritems():
        if pd.isnull(frame):
            maxId = maxId + 1
            frame['Id'] = maxId

But I get an error:

TypeError: 'float' object is not subscriptable

What do I need to do for a fix?

In general with Pandas, you should look to avoid row-wise for loops. Vectorised column-wise operations are possible. — jpp
– jpp, Commented Jan 10, 2019 at 11:33
Please provide a minimal reproducible example, as well as the entire error message. — AMC
– AMC, Commented Apr 24, 2020 at 0:28

jpp · Accepted Answer · 2019-01-10 11:14:38Z

2

Using pd.Series.isnull and np.arange:

# calculate maximum value
maxId = int(pd.to_numeric(df['Id'], errors='coerce').max())

# calculate Boolean series of nulls
nulls = df['Id'].isnull()

# assign range starting from one above maxId
df.loc[nulls, 'Id'] = np.arange(maxId + 1, maxId + 1 + nulls.sum())

print(df)

#      Id
# 0     3
# 1    67
# 2   356
# 3   679
# 50   P4
# 51   P5
# 52  678
# 53  680
# 54    2

answered Jan 10, 2019 at 11:14

jpp

166k37 gold badges301 silver badges362 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

anky · Accepted Answer · 2019-01-10 11:12:28Z

1

As you say you have already figured the maxId you can try this vectorized solution :

>>df

    Id
0   3
1   67
2   356
3   NaN
5   P4
6   P5
7   678
8   NaN
9   2

n = 678
n=n+1
df.loc[df.Id.isnull(), 'Id'] = list(np.arange(n,n+len(df.Id[df.Id.isna()].values)))
>>df

Output:

edited Jan 10, 2019 at 11:12

answered Jan 10, 2019 at 11:07

anky

75.3k11 gold badges46 silver badges76 bronze badges

Comments

adan · Accepted Answer · 2019-01-10 10:39:50Z

0

Do you need the values like 'P4' and 'P5' ? I tried to reproduce a similar DataFrame to yours, but without those values and it just works :

df = pd.DataFrame({'A' : [20,4, np.nan, np.nan, 12, np.nan, 6, 10]})

maxID = df['A'].max()

for i in range (len(df['A'])):
    if pd.isnull(df['A'].loc[i]):
        maxID +=1
        df['A'].loc[i] = maxID

I think your error occurs because you are trying to access an element of a float such as you would do with a list.

Exemple :

my_float = 3.0 
my_float[0]

TypeError: 'float' object is not subscriptable

answered Jan 10, 2019 at 10:39

adan

1311 silver badge11 bronze badges

Collectives™ on Stack Overflow

Iterating a data-frame column: TypeError: 'float' object is not subscriptable

3 Answers 3

Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related