look for nan in a python dictionary

Asked Nov 28 '14 at 08:13

Active Jan 10 '15 at 17:33

Viewed 2,233 times

I have a pandas dataframe converted into a dictionary (OrderedDict) and now I need to look for real NaN values (not the 'nan' string values) without pandas/numpy methods (as it should be independent from these packages). The dictionary looks like this:

OrderedDict([('concentration', ['5', '5', '5', '5', '5', nan, 'nan', 'nan', 'nan', 'nan', 'nan']),
 ('val', [6.0, 6.0, 6.0, 7.0, 8.0, 9.0, nan, nan, 0.0, 3.0, nan])])

I would like to have something like this:

keys = odict.keys()
count = len(odict[keys[0]])
for current in range(count):
        row = []
        for item in table:
            column = table[item]
            value = column[current]
            if >>>value is not nan<<<:
                row.append("MISSING")
            elif value is not None:
                row.append("value")
            else:
                row.append("something else")
        print(row)

Can anybody help?

EDIT: The full solution for me is not so easy to find within the answers/comments from the duplicate question. That's why I add it here as a quote:

... Better to use a try/except construction: def is_nan(x): try: return math.isnan(x) except: return False – Rob

edited Jan 10 '15 at 17:33

Mike McKerns

33,715
8
119
139

asked Nov 28 '14 at 08:13

Antje Janosch

1,154
5
19
37

1

Use `math.isnan()` to see whether the value is nan or not. – RemcoGerlich Nov 28 '14 at 08:14
but math.isnan() expects float values. I then get an error: TypeError: a float is required – Antje Janosch Nov 28 '14 at 08:29
Well, then you know it's not nan. Catch the exception and move on? – RemcoGerlich Nov 28 '14 at 08:31
good idea. thanks. I'm not yet very deep into python. sorry for question with rather simple answers. – Antje Janosch Nov 28 '14 at 08:34

look for nan in a python dictionary

0 Answers0