Mailman 3 January 2016 - Kwant-discuss

How strict should we take backwards compatibility?
by Christoph Groth 07 Jan '16

07 Jan '16

Dear all, With Kwant, we strive to keep backwards compatibility across major versions: a script that only uses the documented API and works with Kwant 1.x should continue to work unchanged with Kwant 1.y, for any y >= x. Still, we have a bunch of ideas that require breaking strict compatibility. That’s why they have been going under “Kwant 2”: • A new low-level system format that supports more general symmetries and vectorization of value functions. • Composite systems that combine systems with various degrees of commensurate symmetries in a general way. (That’s a generalization of quasi-1-d leads attached to quasi-0-d scattering regions that we have now.) • Various small changes and fixes to the interface. These things pull in other changes (e.g. in the solvers) which means that “Kwant 2” is quite a big project. Recently, I asked myself whether we really have to interpret backwards compatibility in such a strict way. Perhaps a better way would be “if it does not hurt >98% of the users, it’s OK to change things in a backwards-incompatible (but well documented) way”. I believe that there is virtually no one who uses the low level system interface [1] directly in non-trivial ways (e.g. doing something with system.graph and not just calling system.hamiltonian_submatrix() for a finite system). Is this actually true? Can people who do things with low-level systems directly please chime in and tell us what their most involved usage cases are? If it turns out that a weak interpretation of backwards compatibility is indeed OK, we could tackle the new low-level system format already for Kwant 1.3. This would make the development of Kwant more evolutionary and should make it more dynamic. What do you think? Christoph [1] http://kwant-project.org/doc/1/reference/kwant.system

3 5

Installing Kwant on Windows
by Jerzy Wrobel 04 Jan '16

04 Jan '16

Hi All, The command, which is suggested on Kwant webpage pip3 install --no-deps *.whl generates (python 3.4 and win7) the following error message: Requirement '*.whl' looks like a filename, but the file does not exist *.whl is not a valid filename. Are there any other possibilities to install a bunch of packages with a single command on Windows? Thanks in advance, Jerzy

3 7

How resource dependent is Kwant?
by siddheshwar chopra 03 Jan '16

03 Jan '16

Dear Users, Can anybody tell me HOW resource intensive is KWant? I mean, as compared to DFT calculations. For DFT, computer has to be quite powerful. Any suggestion or explanation is welcomed. Especially in terms of number of atoms. Regards, -- *Dr. Siddheshwar chopra,* *M.Sc., Ph.D (Physics)Assistant Professor (Physics),* *Amity University, Noida, India.*

2 1

Hall effect (4-contacts measurement)
by Cabosart Damien 03 Jan '16

03 Jan '16

Dear all, I read some advices about the Hall effect (4-leads measurement) in previous posts referring to the code on the website: http://nbviewer.ipython.org/github/topocm/topocm_content/blob /master/w3_pump_QHE/Laughlinargument.ipynb However, I'm wondering if the following code is well correct in order to get the potential of each lead when a current I=1 is "imposed" between the contacts 4 and 5. r = np.linalg.inv(G) V = r.dot(np.array([0, 0, 0, 1, -1])) where G is the conductance matrix ( G[i,j] gives the conductance between lead i and j) and hence, r is the resistance matrix. V[0] is the potential of the lead 0 V[1] is the potential of the lead 1 V[2] is the potential of the lead 3 This would mean that, for example, V[0] = r[3,0] - r[4,0] However, is it more correct to write V[0] = r[3,0] - r[0,4] ? Indeed, r[0,4] is not equal to r[4,0] when we apply a magnetic field (the time reversal symmetry is broken). I think it is better to use r[0,4] in order to set a current flowing from lead 4 to lead 5. Am I right? Thank you in advance for your help. Best regards, D. Cabosart

2 1

Re: [Kwant] How resource dependent is Kwant?
by siddheshwar chopra 03 Jan '16

03 Jan '16

Dear Anton, Thank you for the information. As I have worked with only DFT till now, it was my concern to know how intensive configuration Kwant needs. Ok, so it seems that CPU time is not a concern...because in DFT it takes many hours.. Here it looks like RAM (as u said) is the bottleneck. In DFT we talk about say less than 500 atoms, but here as you mentioned, it is in thousands.. Thank you again, Kind Regards, On 03-Jan-2016 8:31 pm, "Anton Akhmerov" <anton.akhmerov+kd(a)gmail.com> wrote: > Dear Siddheshwar Chopra, > > It strongly depends on the specific system you're solving. You can > check out some benchmarks in the Kwant paper or run your own. > Typically RAM is the bottleneck, and in 2D a system with size of > around 2000x2000 should fit into a few GB, while in 3D it's maybe > 30x30x30. The runtime for systems of such sizes will tens of minutes, > probably. > > Best, > Anton > > On Tue, Dec 29, 2015 at 7:32 AM, siddheshwar chopra <sidhusai(a)gmail.com> > wrote: > > Dear Users, > > Can anybody tell me HOW resource intensive is KWant? I mean, as compared > to > > DFT calculations. For DFT, computer has to be quite powerful. Any > suggestion > > or explanation is welcomed. Especially in terms of number of atoms. > > > > Regards, > > > > -- > > Dr. Siddheshwar chopra, > > M.Sc., Ph.D (Physics) > > Assistant Professor (Physics), > > Amity University, Noida, India. > > >

1 0