Oooops. I overlooked one fundamental issue, related to my previous question
1): I already HAVE my data in a numpy array, but do not seem to be allowed
to pass it when creating the CArray. Is it not possible??
Regards,
Dominik
On Mon, Jan 3, 2011 at 9:39 PM, Dominik Szczerba <domi...@itis.ethz.ch>wrote:
> Many thanks for useful hints. Some questions to them:
>
> 1) Where is a list of constructor options for carray, I could not find it
> in the document you referenced.
> 2) Will the "decent value" when chunk size is not specified handle properly
> a case where my matrix is ordered like fortran (column-wise)?
>
> Thanks a lot,
> Dominik
>
>
> On Mon, Jan 3, 2011 at 5:48 PM, Francesc Alted <fal...@pytables.org>wrote:
>
>> Hi Dominik,
>>
>> 2011/1/3, Dominik Szczerba <domi...@itis.ethz.ch>:
>> > Hi,
>> >
>> > I could not find a clear example how to create a compressed dataset.
>> Could
>> > someone please point me in the right direction?
>> > Let's say I want to do something as simple as the below, only
>> compressing
>> > the data using zlib, when required defining 1e6 chunk size.
>> >
>> > f = tables.openFile(fname,"w")
>> > f.createArray(f.root,"u",u)
>> > f.createArray(f.root,"v",v)
>> > f.createArray(f.root,"w",w)
>> > f.close()
>>
>> Array objects are non-chunked. In order to use compression, you need
>> to use a CArray:
>>
>> http://www.pytables.org/docs/manual/ch04.html#CArrayClassDescr
>>
>> or an EArray (objects enlargeable on a certain dimension):
>>
>> http://www.pytables.org/docs/manual/ch04.html#EArrayClassDescr
>>
>> For choosing the desired chunksize, make use of the `chunkshape`
>> attribute in constructors (if not specified, a decent default will be
>> automatically chosen for you).
>>
>> HTH,
>>
>> --
>> Francesc Alted
>>
>>
>> ------------------------------------------------------------------------------
>> Learn how Oracle Real Application Clusters (RAC) One Node allows customers
>> to consolidate database storage, standardize their database environment,
>> and,
>> should the need arise, upgrade to a full multi-node Oracle RAC database
>> without downtime or disruption
>> http://p.sf.net/sfu/oracle-sfdevnl
>> _______________________________________________
>> Pytables-users mailing list
>> Pytables-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/pytables-users
>>
>>
>
------------------------------------------------------------------------------
Learn how Oracle Real Application Clusters (RAC) One Node allows customers
to consolidate database storage, standardize their database environment, and,
should the need arise, upgrade to a full multi-node Oracle RAC database
without downtime or disruption
http://p.sf.net/sfu/oracle-sfdevnl
_______________________________________________
Pytables-users mailing list
Pytables-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/pytables-users
