A Friday 10 December 2010 21:24:35 Dominik Szczerba escrigué:
> Hi,
>
> When calling:
>
> f = tables.openFile(fname)
> points = array(f.getNode("/points").read())
> tets = array(f.getNode("/tetrahedrons").read())
> domain = array(f.getNode("/domain").read())
> f.close()
>
> I am getting the following error:
>
> /usr/lib/python2.6/dist-packages/tables/leaf.py:415:
> PerformanceWarning: The Leaf ``/tetrahedrons`` is exceeding the
> maximum recommended rowsize (13107200 bytes);
> be ready to see PyTables asking for *lots* of memory and possibly
> slow I/O. You may want to reduce the rowsize by trimming the value
> of dimensions that are orthogonal (and preferably close) to the main
> dimension of this leave. Alternatively, in case you have specified
> a very small/large chunksize, you may want to increase/decrease it.
> PerformanceWarning)
>
> I only found one similar thread in the archives, unfortunately, never
> concluded.
>
> My file is:
> > h5ls Obese-00000.h5
>
> domain Dataset {4622544}
> points Dataset {3, 793418}
> tetrahedrons Dataset {4, 4622544}
>
> No error is reported for the other two arrays.
>
> Can it be that pytables silently assumes row-major ordering for
> matrices? I need to store my data in the fortran order.
Yes, it assumes row-major order (the default for NumPy). The above
warning is to prevent people about the potentially high memory
consumption of iterating the dataset like this:
for row in tetrahedrons:
# do things with row
But, while warning inexpert people about this is generally a good thing,
I recognize that these warning can be rather annoying for 'expert'
people. Hmmm, I'm thinking that adding an 'EXPERT_MODE' parameter would
be nice. Added a ticket:
http://pytables.org/trac/ticket/327
Meanwhile, you can get rid of such a warning using the Python machinery:
http://docs.python.org/library/warnings.html
Hope this helps,
--
Francesc Alted
------------------------------------------------------------------------------
Oracle to DB2 Conversion Guide: Learn learn about native support for PL/SQL,
new data types, scalar functions, improved concurrency, built-in packages,
OCI, SQL*Plus, data movement tools, best practices and more.
http://p.sf.net/sfu/oracle-sfdev2dev
_______________________________________________
Pytables-users mailing list
Pytables-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/pytables-users
