Re: [ojAlgo-user] Changing a few values to a BasicMatrix

[Anders P. <an...@op...>]

To clarify when/if you’re working with immutable matrices you have to this:

tmpTensor[9] = tmpTensor[9].add(56, 0, tmpDifference);



> On 23 okt. 2015, at 17:03, Anders Peterson <an...@op...> wrote:
> 
> You could do something like this:
> 
>        BasicMatrix[] tmpTensor = new BasicMatrix[10] {….};
> 
>        BasicMatrix tmpDifference = PrimitiveMatrix.FACTORY.getBuilder(3, 1600).fillAll(3.14).build();
> 
>        tmpTensor[9].add(56, 0, tmpDifference);
> 
> That will change rows 56, 57 and 58 of matrix 9.
> 
> It doesn’t cost you (CPU) anything now. Depending on what you do it may cost you later.
> 
> 
> I’m guessing BasicMatrix is not a good choice for you. Doesn’t sound like immutable objects suit your needs. Try working with PrimitiveDenseStore instead.
> 
> 	PrimitiveDenseStore[] tmpTensor = new PrimitiveDenseStore[10] {….};
> 
> 
> Did you read the "Getting Started: Linear Algebra” wiki page?
> 
> https://github.com/optimatika/ojAlgo/wiki/Getting-Started
> 
> 
> In the org.ojalgo.array package there are classes supporting 1, 2 or “any” dimensional arrays of double, BigDecimal, ComplexNumber, Quaternion and RationalNumber. The arrays can be sparse or dense and arbitrarily large. You can directly do this:
> 
> ArrayAnyD<Double> tmpTensor3D = ArrayAnyD.PRIMITIVE.makeZero(1000, 1600, 10);
> 
> 
> I must admit that the code supporting the any-dimensional features is underdeveloped and not very well tested. 99% seems to only need 1 or 2 dimensional arrays (vectors and matrices). If the ArrayAnyD class doesn’t have what you need it can be added.
> 
> 
> /Anders
> 
> 
> 
>> On 23 okt. 2015, at 15:53, Stathis Maroulis <sta...@gm...> wrote:
>> 
>> Hello,
>> I'm currently trying to  implement a 3D Tensor library, using Ojalgo. Every time i want to change some elements of the matrices I am using a builder. But i think I'm losing a lot of (CPU) time doing this.
>> For example when if have a 1000x1600x10 tensor (an array of ten 1000x16000 BasicMatrix matrices) and what to change some rows, i have to rebuilt the whole tensor. Am I missing something or this is the way we do this?
>> Thanks for you time.
>> 
>> 
>> Stathis Maroulis
>> ------------------------------------------------------------------------------
>> _______________________________________________
>> ojAlgo-user mailing list
>> ojA...@li...
>> https://lists.sourceforge.net/lists/listinfo/ojalgo-user
> 
> 
> ------------------------------------------------------------------------------
> _______________________________________________
> ojAlgo-user mailing list
> ojA...@li...
> https://lists.sourceforge.net/lists/listinfo/ojalgo-user