ValueError when trying to convert Dataframe Column into float

Question

I am trying to convert all the data under the Column Yield_pct of my dataframe df below into float types, but am facing issues doing this. I get the Error ValueError: could not convert string to float: ' na' and hence I added the line if row1['Yield_pct'] != 'na': to my code below, yet I get the same error after adding this line.

              Date      Maturity      Yield_pct Currency
0       1986-01-01          0.25             na      CAD
1       1986-01-02          0.25   0.0948511020      CAD
2       1986-01-03          0.25   0.0972953210      CAD
3       1986-01-06          0.25   0.0965403640      CAD
4       1986-01-07          0.25   0.0953292440      CAD


for (i1, row1) in (df.iterrows()):
    if row1['Yield_pct'] != 'na':
        row1['Yield_pct'] = float(row1['Yield_pct'])
        if isinstance(row1['Yield_pct'], float)==1:
            print('SUCCESS')
        else:
            print('FAILURE')

Thank You

Edit: This is the lower part of the dataframe df:

920538  2015-01-19  empty string                     CAD
920539  2015-01-20  empty string                     CAD
920540  2015-01-21  empty string                     CAD
920541  2015-01-22  empty string                     CAD
920542  2015-01-23  empty string                     CAD
920543  2015-01-26  empty string                     CAD

Code that I am now using:

df = update('CAD')[0]
for (i1, row1) in (df.iterrows()):
    df = df.convert_objects(convert_numeric=True)
    if isinstance(row1['Yield_pct'], float)==1:
        print('SUCCESS')
    else:
        print('FAILURE')

What's your question? Notice that your error says you cannot convert string ' na' to a float. You're missing the space before the 'na' in your code. That's a starting place. — khol
– khol, Commented Jun 29, 2015 at 15:17
@EdChum Thank You. However, when I try this FAILURE still keeps getting printed out. — user131983
– user131983, Commented Jun 29, 2015 at 15:59
Please edit you question with the code you tried that now doesn't work — EdChum
– EdChum, Commented Jun 29, 2015 at 16:33
I don't know why you repeatedly call convert_objects, you just need to call it once, also when you claim it fails, after calling convert_objects what does df.info() show? You may be getting what you think is an error because the dtype is actually np.float64 — EdChum
– EdChum, Commented Jun 29, 2015 at 16:37

EdChum · Accepted Answer · 2015-06-29 15:16:28Z

3

Just use convert_objects, it will coerce any duff values into NaN:

In [75]:
df = df.convert_objects(convert_numeric=True)
df

Out[75]:
         Date  Maturity  Yield_pct Currency
0  1986-01-01      0.25        NaN      CAD
1  1986-01-02      0.25   0.094851      CAD
2  1986-01-03      0.25   0.097295      CAD
3  1986-01-06      0.25   0.096540      CAD
4  1986-01-07      0.25   0.095329      CAD

answered Jun 29, 2015 at 15:16

EdChum

397k204 gold badges836 silver badges583 bronze badges

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

6 Comments

Wesley Bowman Over a year ago

I didn't know this was a thing. Beautiful.

user131983 Over a year ago

@EdChum Thank You. But when I try the lines df = df.convert_objects(convert_numeric=True) row1['Yield_pct'] = float(row1['Yield_pct']) I don't know why this is.

EdChum Over a year ago

I don't understand your comment, using convert_objects you don't need any further casting

user131983 Over a year ago

@EdChum Unfortunately all the data in the Column are Strings, so I still need to convert those into float and so I use row1['Yield_pct'] = float(row1['Yield_pct']). If I use df = df.convert_objects(convert_numeric=True) alone, I won't be able to satisfy this requirement.

EdChum Over a year ago

Does df['Yield_pct'] = df['Yield_pct'].astype(np.float64) work? If it doesn't work it's because you have some strange values that you've not shown in your sample data, can you post this

|

Collectives™ on Stack Overflow

ValueError when trying to convert Dataframe Column into float

1 Answer 1

6 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

6 Comments

Your Answer

Sign up or log in

Post as a guest

Related