WIP : Spin Operators, Tests. WD : +,-,ops added. #22

amitjamadagni · 2015-03-29T20:53:31Z

WIP :

Spin operators to be implemented
Tests

Implemented

+,- on QuArray, CTranspose
momentm_op, position_op added.

acroy · 2015-03-29T21:32:05Z

src/arrays/arraymath.jl

@@ -61,6 +61,9 @@ end

 *(dm1::DualMatrix, dm2::DualMatrix) = (dm1.qarr*dm2.qarr)'

+(qarr1::Union(QuArray,CTranspose), qarr2::Union(QuArray,CTranspose)) = bases(qarr1) == bases(qarr2) ? (return QuArray(coeffs(qarr1)+coeffs(qarr2), bases(qarr1))) : "Not compatible"


I think you want to use error("Bases not compatible.") here. Should we use type parameters (for the basis and N) to restrict the functions in the first place?

@acroy something like

function +{B<:AbstractBasis,N}(qarr1::Union(QuArray{B,N}, CTranspose{B,N}), qarr2::Union(QuArray{B,N},CTranspose{B,N}))

Yes, that way we get e MethodError "for free" if someone tries to add say a QuVector and a QuMatrix. Nevertheless we have to check that the bases are the same (I am not sure if == works though).

I think == is the right thing to do here, since it can be overloaded to work properly for each basis type - "value" equality is what we're looking for here, yes? Also, all we need to do to support mixed-basis type operations is to simply define some promotion rules between the various basis types, of which there aren't that many.

acroy · 2015-03-30T16:44:55Z

src/arrays/ladderops.jl

 after_eye(lens, pivot))
 end

+function positionop(n::Int)
+ cop = raiseop(n)
+ return (cop+cop')/sqrt(2.)


We should use scale! here since + already makes a copy and / would make another one (same for momentumop below).

@acroy Ah! Once again, I comment only to find that you've beat me to it by the time I'm done writing. I'm starting to believe that you are, in fact, a ninja...

Actually, I just realized we won't be able to perform an in-place scaling of the Float64 array by a complex number; see my edit on my comment below.

amitjamadagni · 2015-03-30T20:34:16Z

@jrevels I was thinking a bit off on it. I realize it is costlier. Anyways thanks for this :).

acroy · 2015-03-31T11:14:48Z

For spin operators we might want to keep an eye on FixedSizeArrays.jl (also JuliaLang/julia#7568)?

amitjamadagni · 2015-04-01T13:37:41Z

@acroy I was doing something like this based on QuTiP :

function spin_j(j)
          m = reverse([-j:j-1])
          N = length(m)+1
          m_coeffs = [sqrt(j*(j+1.0) - (x+1.0)*x) for x in m]
          return QuArray(spdiagm(m_coeffs,1,N,N))
end

As far as my understanding of FixedSizeArrays.jl goes we are trying to create the array in a more efficient way with focus on the size. I am still trying to further understand but please do let me know if this is going in the right direction.

acroy · 2015-04-01T16:42:07Z

@amitjamadagni : for now we probably want (have) to use sparse matrices as containers as you do. I just stumbled across FixedSizeArrays and thought those could be useful for stuff like Pauli spin matrices, which are quite common. AFAICT the FixedSizeArrays need Julia 0.4 anyways, so this is something for QuBase 2.0 :-)

amitjamadagni · 2015-04-02T17:18:48Z

@acroy @jrevels a review would be helpful. I have added the spin operators. Thanks

acroy · 2015-04-03T08:32:43Z

You have to rebase against the latest master and maybe we should put the spin operators into a separate file?

amitjamadagni · 2015-04-03T08:35:01Z

@acroy sure okay. But could we move the entire machinery into a single file with the name as operators.jl ??

acroy · 2015-04-03T08:36:43Z

src/arrays/ladderops.jl

+ return scale(im/sqrt(2.), cop-cop')
+end
+
+function spin_h1(j::Float64)


Maybe it is more elegant to use Rationals here (see docs)?

acroy · 2015-04-03T09:41:46Z

Let's see what Jarrett says, he has a better overview of the repo structure. I thought we could move the spin operators into a file spinops.jl or such ...

Conflicts: src/arrays/quarray.jl

amitjamadagni · 2015-04-03T13:51:32Z

@acroy I am trying to rebase against the latest master. In the attempt I created a bit of mess with the commits. Any pointers relating to rebase would be helpful. Thanks.

jrevels · 2015-04-03T16:40:36Z

I agree with using Rationals for the spin operators and moving them to their own file ("spinops.jl" sounds good to me).

@amitjamadagni You might've already seen this before, but I found this tutorial very helpful when learning to rebase. Is there a specific thing you're having trouble with, error message, etc.?

amitjamadagni · 2015-04-03T16:51:26Z

@jrevels Yeah I have been trying and playing around but the problem I end up is I am unable to see the previous commits when I do a rebase -i master, to squash. I will give it a try once again and let you know.

amitjamadagni · 2015-04-10T16:59:11Z

@acroy @jrevels I have made an attempt. A review would be helpful.

acroy · 2015-04-10T17:21:26Z

I think it is basically Ok. If you still have energy, you could add some comments to that spin function, in particular reminding us that it is not type stable. It would also be great if you could squash your commits.

amitjamadagni · 2015-04-10T17:47:04Z

@acroy I am unable to squash some of the previous commits.

acroy · 2015-04-10T17:53:16Z

Why?

amitjamadagni · 2015-04-10T17:57:41Z

I dont get the previous 8 commits before the last when i do a git rebase -i master.

acroy · 2015-04-10T18:00:47Z

It should probably be git rebase -i HEAD~9.

amitjamadagni · 2015-04-10T18:13:19Z

@acroy Thanks for the help. Done squashing.

acroy · 2015-04-10T18:16:12Z

Mhm, I can't merge it. Maybe you have to rebase again?

Position and momentum, Spin operators, Related Tests.

amitjamadagni · 2015-04-10T22:59:07Z

@acroy seems like I am stuck in a loop. Can I close this and open an another pull request ??

WIP : Operators, Docs, Tests. WD : +,-,ops added.

0fe8886

acroy reviewed Mar 29, 2015
View reviewed changes

Use of Type parameters.

98fec4b

acroy reviewed Mar 30, 2015
View reviewed changes

Use of scale!

f3b5be2

Use of scale, AbstractQuArray usage.

d9574c2

WIP : Spin operators.

722fbc5

amitjamadagni force-pushed the op_machinery branch from 375a9e9 to 722fbc5 Compare April 1, 2015 21:36

amitjamadagni added 2 commits April 2, 2015 15:32

Spin Operators.

0dc1e40

Check on input.

d9007d7

acroy reviewed Apr 3, 2015
View reviewed changes

Merge branch 'master' into op_machinery

62145d3

Conflicts: src/arrays/quarray.jl

amitjamadagni force-pushed the op_machinery branch 4 times, most recently from 62145d3 to 9d59b67 Compare April 3, 2015 13:06

amitjamadagni force-pushed the op_machinery branch from 8ba5471 to 285bbec Compare April 10, 2015 17:43

amitjamadagni force-pushed the op_machinery branch from 285bbec to 2734d63 Compare April 10, 2015 18:12

amitjamadagni force-pushed the op_machinery branch 14 times, most recently from f409ff7 to 77d22e1 Compare April 10, 2015 22:45

Added : Operators

879e9f6

Position and momentum, Spin operators, Related Tests.

amitjamadagni force-pushed the op_machinery branch from 53c541b to 879e9f6 Compare April 10, 2015 22:48

amitjamadagni closed this Apr 11, 2015

amitjamadagni deleted the op_machinery branch April 11, 2015 05:43

amitjamadagni mentioned this pull request Apr 11, 2015

Operators, Tests. #24

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

WIP : Spin Operators, Tests. WD : +,-,ops added. #22

WIP : Spin Operators, Tests. WD : +,-,ops added. #22

amitjamadagni commented Mar 29, 2015

acroy Mar 29, 2015

amitjamadagni Mar 30, 2015

acroy Mar 30, 2015

jrevels Mar 30, 2015

acroy Mar 30, 2015

jrevels Mar 30, 2015

jrevels Mar 30, 2015

amitjamadagni commented Mar 30, 2015

acroy commented Mar 31, 2015

amitjamadagni commented Apr 1, 2015

acroy commented Apr 1, 2015

amitjamadagni commented Apr 2, 2015

acroy commented Apr 3, 2015

amitjamadagni commented Apr 3, 2015

acroy Apr 3, 2015

acroy commented Apr 3, 2015

amitjamadagni commented Apr 3, 2015

jrevels commented Apr 3, 2015

amitjamadagni commented Apr 3, 2015

amitjamadagni commented Apr 10, 2015

acroy commented Apr 10, 2015

amitjamadagni commented Apr 10, 2015

acroy commented Apr 10, 2015

amitjamadagni commented Apr 10, 2015

acroy commented Apr 10, 2015

amitjamadagni commented Apr 10, 2015

acroy commented Apr 10, 2015

amitjamadagni commented Apr 10, 2015

		@@ -61,6 +61,9 @@ end

		(dm1::DualMatrix, dm2::DualMatrix) = (dm1.qarrdm2.qarr)'

		+(qarr1::Union(QuArray,CTranspose), qarr2::Union(QuArray,CTranspose)) = bases(qarr1) == bases(qarr2) ? (return QuArray(coeffs(qarr1)+coeffs(qarr2), bases(qarr1))) : "Not compatible"

WIP : Spin Operators, Tests. WD : +,-,ops added. #22

WIP : Spin Operators, Tests. WD : +,-,ops added. #22

Conversation

amitjamadagni commented Mar 29, 2015

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

amitjamadagni commented Mar 30, 2015

acroy commented Mar 31, 2015

amitjamadagni commented Apr 1, 2015

acroy commented Apr 1, 2015

amitjamadagni commented Apr 2, 2015

acroy commented Apr 3, 2015

amitjamadagni commented Apr 3, 2015

Choose a reason for hiding this comment

acroy commented Apr 3, 2015

amitjamadagni commented Apr 3, 2015

jrevels commented Apr 3, 2015

amitjamadagni commented Apr 3, 2015

amitjamadagni commented Apr 10, 2015

acroy commented Apr 10, 2015

amitjamadagni commented Apr 10, 2015

acroy commented Apr 10, 2015

amitjamadagni commented Apr 10, 2015

acroy commented Apr 10, 2015

amitjamadagni commented Apr 10, 2015

acroy commented Apr 10, 2015

amitjamadagni commented Apr 10, 2015