<div dir="ltr">I forgot to mention that the loaded voxel selction works fine.<div>It's just when I try to save it with h5py, even just after loading, it crashes.</div></div><div class="gmail_extra"><br><div class="gmail_quote">On Mon, Oct 12, 2015 at 1:02 PM, Swaroop Guntupalli <span dir="ltr"><<a href="mailto:swaroopgj@gmail.com" target="_blank">swaroopgj@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">Hi Nick,<div><br></div><div>I had a voxel selection pre-computed from a while ago (Oct. 2013).</div><div>When I try to save it now using h5py, it gives the following error (pasted below).</div><div><br></div><div>Is there a way to resolve this without re-computing the voxel selection?</div><div>I can point you to the voxel selection on hydra if that helps.</div><div><br></div><div>Thanks,</div><div>Swaroop</div><div><br></div><div><br></div><div><div>/PyMVPA/mvpa2/base/hdf5.py in obj2hdf(hdf, obj, name, memo, noid, **kwargs)</div><div> 758 # try disassembling the object</div><div> 759 try:</div><div>--> 760 pieces = obj.__reduce__()</div><div> 761 if __debug__:</div><div> 762 debug('HDF5', "Reduced '%s' (ref: %i) in [%s]"</div><div><br></div><div>/PyMVPA/mvpa2/misc/surfing/volume_mask_dict.pyc in __reduce__(self)</div><div> 568 return (self.__class__,</div><div> 569 (self._volgeom, self._source),</div><div>--> 570 self.__getstate__())</div><div> 571</div><div> 572 @deprecated("should be used for testing compatibility only - "</div><div><br></div><div>/PyMVPA/mvpa2/misc/surfing/volume_mask_dict.pyc in __getstate__(self)</div><div> 599 # here a new tuple is created</div><div> 600 s3 = _dict_with_arrays2array_tuple(s[3])</div><div>--> 601 s4 = _dict_with_arrays2array_tuple(s[4])</div><div> 602 ss = s[:3] + (s3, s4)</div><div> 603</div><div><br></div><div>/PyMVPA/mvpa2/misc/surfing/volume_mask_dict.pyc in _dict_with_arrays2array_tuple(d)</div><div> 938 # probably src2aux, so run recursively</div><div> 939 return dict((k, _dict_with_arrays2array_tuple(v))</div><div>--> 940 for k, v in d.iteritems())</div><div> 941</div><div> 942 keys = np.asarray(d.keys())</div><div><br></div><div>/PyMVPA/mvpa2/misc/surfing/volume_mask_dict.pyc in <genexpr>((k, v))</div><div> 938 # probably src2aux, so run recursively</div><div> 939 return dict((k, _dict_with_arrays2array_tuple(v))</div><div>--> 940 for k, v in d.iteritems())</div><div> 941</div><div> 942 keys = np.asarray(d.keys())</div><div><br></div><div>/PyMVPA/mvpa2/misc/surfing/volume_mask_dict.pyc in _dict_with_arrays2array_tuple(d)</div><div> 965 # ensure all values in the dict have the same datatype</div><div> 966 raise ValueError('Type mismatch for keys %s and %s: %s != %s' %</div><div>--> 967 (keys[0], key, data.dtype, v.dtype))</div><div> 968</div><div> 969 idxs = np.arange(length) + pos</div><div><br></div><div>ValueError: Type mismatch for keys 0 and 41: float64 != float32</div></div><div><br></div></div>
</blockquote></div><br></div>