[SciPy-user] Masked array question

Robert Ferrell ferrell@diablotech....
Mon Jun 8 11:59:41 CDT 2009


I have a tuple of strings that I want to convert to an array of  
floats.  Some of the strings are empty, so I thought I could use a  
masked array to mask out the empty strings.  (In my application, empty  
string means no data, so ignore.)

I tried:

np.ma.masked_array(('1.', ' '), mask=(False, True), dtype=(np.float32,  
np.float32))

but that doesn't work.  It returns

/Library/Frameworks/Python.framework/Versions/2.5.2001/lib/python2.5/ 
site-packages/numpy-1.3.0.dev5934-py2.5-macosx-10.3-i386.egg/numpy/ma/ 
core.py in __new__(cls, data, mask, dtype, copy, subok, ndmin,  
fill_value, keep_mask, hard_mask, flag, shrink, **options)
    1257             shrink = flag
    1258         # Process data............
-> 1259         _data = np.array(data, dtype=dtype, copy=copy,  
subok=True, ndmin=ndmin)
    1260         _baseclass = getattr(data, '_baseclass', type(_data))
    1261         # Check that we'ew not erasing the mask..........

<type 'exceptions.ValueError'>: setting an array element with a  
sequence.

Is there a way to do this (easily)?  What I can do is first convert  
the strings to floats, and set the empty strings to some fill value,  
and then pass that to masked_array.  Is that the best solution?  Or is  
there some method that I've missed?

thanks,
-robert


More information about the SciPy-user mailing list