Parity #74
Parity #74
Conversation
There was a problem hiding this comment.
I left some inline comments. Would be better to follow the general template in https://github.com/qutip/qutip-tutorials/blob/main/tutorials-v5/template.md a bit more too, and add some simple tests at the end.
Main feedback points are the latex at the beginning needs fixing, and the spectral function stuff can be tidied up and explained a bit more. I can have a go at this if it helps.
edit: forgot to mention, better to remove the dephasing example changes, I guess these were just added by mistake
|
|
||
| $H_{T}=H_{S}+H_{f}+H_{ef}$ | ||
|
|
||
| Where $H_{S}$ is the system Hamiltonian, which is divided into three contributions. The electron system hamiltonian ($H_{e}$), the single-mode cavity ($H_{c}$) and the light matter coupling ($H_{ef}$): |
There was a problem hiding this comment.
there is no cavity in this example, better to remove mention of it here and elsewhere.
|
|
||
| $H_{e}= \sum_{n=g,e}\sum_{\sigma=\uparrow,\downarrow} \epsilon_{n} \hat{n}^{n \sigma} + U_{n} \hat{n}_{n \uparrow} \hat{n}_{n \downarrow} $ | ||
|
|
||
| $H_{ec} = \sum_{\sigma=\uparrow,\downarrow} g_{ct} (d_{g\sigma}^{\dagger} d_{e\sigma} + d_{e\sigma}^{\dagger} d_{g\sigma} )(a^{\dagger}+a)$ |
There was a problem hiding this comment.
same here, no cavity coupling etc
|
|
||
| and | ||
|
|
||
| $H_{ef}= \sum_{k} \sum_{\alpha=L,R} \sum_{\sigma=\uparrow,\downarrow} g_{\alpha,k} (c_{\alpha,k}^{\dagger} d_{g\sigma}+c_{\alpha,k} d_{g\sigma}^{\dagger})$ |
There was a problem hiding this comment.
To simplify things i would remove the left/right label here. While one can derive this model from a two-lead one, in practise below we only one ''lead'' split up into two spin labelled baths.
| def _f(N, n): return _oprd([_sz(N, j) for j in range(n)])*_sm(N,n) | ||
| def liouvillian(H): | ||
| return - 1j * (spre(H) - spost(H)) | ||
| ``` |
There was a problem hiding this comment.
we can remove this and use internal functions, fcreate, fdestroy? also probably we dont need liouvillian
| Cw32 = (I_vec3 @ Cw31[:sup_dim]) | ||
| dddag.append(Cw32) | ||
|
|
||
| return -2*(np.array(ddagd).flatten()+np.array(dddag).flatten()) |
There was a problem hiding this comment.
Ideally these would be internal functions, but for now lets leave them here and just make them a bit more generic and documented a little better. also, we can probably make them data-layer agnostic with _data functions, like kron and solve? is it bad to expose users to these in a notebook? Worst case, leave as is, since HEOM mostly uses CSR internally anyway.
| plt.yticks([0.,1,2],[0,1,2]) | ||
| plt.xlabel(r"$\omega/\Gamma$",fontsize=28,labelpad=-10) | ||
| plt.ylabel(r"$2\pi \Gamma A(\omega)$ ",fontsize=28) | ||
|
|
There was a problem hiding this comment.
i tested this output and checked against the old data and it looks good.
This adds a notebook tutorial for the parity feature of HEOM from my pull reques qutip/qutip#2261