Double precision wavelet transform

[Bouchan__Thomas]

Hi, 

I'd like to know if it was possible to use the wavelet transform in double precision (64bits).

Thanks

Thomas

[Alice_H_Intel]

Hello,

Thanks for your question. Wavelet transform function support 16s, 32f and 64. You can refer this document for more detail information. 

https://software.intel.com/sites/default/files/ipps_0.pdf ;

Thanks,

Alice

[Bouchan__Thomas]

Hello, 

Sorry my question was not very precise, I was talking about a 2D wavelet transform, for images. 

In the documentation it is only mentioned 32bits but I wonder what would happen if my inputs are of double precision.

Thanks

Thomas

[Alice_H_Intel]

Hello Thomas,

Right. for 2D wavelet transform, it only supports 32f so far. I will file it as feature request for future review.  

Thanks,

Alice

[Bouchan__Thomas]

Thanks! That would be really useful indeed.

[Gennady_F_Intel]

here is the link to the intel online service center -  https://sfederation.intel.com/affwebservices/public/saml2sso?SPID=https://saml.salesforce.com.isvc.portal2&ProtocolBinding=urn:oasis:names:tc:SAML:2.0:bindings:HTTP-POST&RelayState=%2Fservicecenter%3Flang%3Den-US