I'm trying to use PythonCreator to process and Analyze some data, and my results are in a Pandas DataFrame. There's any way to convert that table to features to use in FME? Thanks for your help

If each record in the DataFrame corresponds to an FME feature, then it should be a simple matter of iterating over the DataFrame, create an instance of FMEFeature and set the attributes (and perhaps even the geometry) before outputting the feature. For example:

# Assuming df is your DataFrame object
for index, row in df.iterrows():
    f = fmeobjects.FMEFeature()
    f.setAttribute('my_attribute_name1', row['attribute1'])
    f.setAttribute('my_attribute_name2', row['attribute2'])
    self.pyoutput(f)

# Assuming df is your DataFrame object
for index, row in df.iterrows():
    f = fmeobjects.FMEFeature()
    f.setAttribute('my_attribute_name1', row['attribute1'])
    f.setAttribute('my_attribute_name2', row['attribute2'])
    self.pyoutput(f)

Thanks for your reply.

I've iterated over the dataframe in every way possible, and I've tried yours, and still have the same error: "TypeError: Could not convert attribute value to a supported attribute type", all my records from the DF are floats. Any Idea what I do wrong?

You can try casting the attribute value to string to see if that works.

I normally encounter that error when the data value is a list.

Yes, that worked.

Thank you all for your replies!

Are you sure it is a standard float and not e.g. numpy float32? The fmeobjects API only knows how to work with the basic Python datatypes such as str/unicode, int and float (and Python lists if you specify an FME list attribute).

You could try to simply cast the value to a regular float, I'm suspecting it would work.

Tip: you can find the exact type of any object by doing

print(type(my_object))

Example:

>>> npfloat = numpy.float32(123.0)
>>> print(type(npfloat))
<class 'numpy.float32'>
 
>>> stdfloat = 123.0
>>> print(type(stdfloat))
<class 'float'>

See also https://note.nkmk.me/en/python-type-isinstance/

You are right, it's was numpy.float32

Converting it to standard float worked as well.

Thanks