Thanks for the link to Gaël's study (that shows well the strengths of pytables/blosc).
Here' s the issue I created related to dataset size estimation linked for reference of people landing from web searches on this thread: https://github.com/PyTables/PyTables/issues/141 (there are pointers there to everything I know about current approaches). -á. On Wed, Mar 28, 2012 at 22:43, Francesc Alted <fal...@pytables.org> wrote: > In case you want more feedback on compression filters, this study might > be interesting for you: > > http://gael-varoquaux.info/blog/?p=159 > > Francesc > > On 3/28/12 12:33 PM, Alvaro Tejero Cantero wrote: >> Hi, >> >> Trying to evaluate compression filters, I was looking for a call in >> PyTables to get the size of a dataset (in bytes). As I didn't find it >> I remembered the many benchmarks and found instead [1] that the way to >> do it is to create single-dataset files and interrogate the >> filesystem. Curiously enough, I didn't find that feature either in >> h5ls, hdfview or ViTables. Is there a structural reason why this size >> cannot be computed from library calls? >> >> -á. >> >> [1] https://github.com/PyTables/PyTables/blob/master/bench/blosc.py >> >> ------------------------------------------------------------------------------ >> This SF email is sponsosred by: >> Try Windows Azure free for 90 days Click Here >> http://p.sf.net/sfu/sfd2d-msazure >> _______________________________________________ >> Pytables-users mailing list >> Pytables-users@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/pytables-users > > > -- > Francesc Alted > > > ------------------------------------------------------------------------------ > This SF email is sponsosred by: > Try Windows Azure free for 90 days Click Here > http://p.sf.net/sfu/sfd2d-msazure > _______________________________________________ > Pytables-users mailing list > Pytables-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/pytables-users ------------------------------------------------------------------------------ This SF email is sponsosred by: Try Windows Azure free for 90 days Click Here http://p.sf.net/sfu/sfd2d-msazure _______________________________________________ Pytables-users mailing list Pytables-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/pytables-users
