Upgrading readme and PDAW weights analysis #59
Conversation
README.md updated with possibility to use polarization of the field (prompted by colleagues from the group). PDAW analysis of the weights was not very consistent. I've now modified it such that it makes more sense.
GitHub's markdown is unable to show vectors and absolute values normally; it turns them into columns somehow. This stupid way of splitting equations into parts is the only way I could force GitHub to show the correct equation. There are other solutions, but they are typically not compatible with other renderers.
|
Could you just have a quick look @danielhollas ? There is only a small change to the analysis part of PDAW, not changing the data in the output |
danielhollas
left a comment
danielhollas
left a comment
There was a problem hiding this comment.
Makes sense to me, thanks!
| We will use this input file in the following examples and refer to it as `input_file.dat`. | ||
|
|
||
| > [!NOTE] | ||
| > **Laser field polarization:** Both PDA and PDAW were derived to account for a linear laser field polarization. Although **PROMDENS** does not take the polarization vector $`\vec{E}_0`$ as a parameter, it can still account for the laser field polarization implicitly through the user input: the user just needs to provide the projection of the transition dipole moments to the laser field polarization |$`\vec{\mu}_{0i}`$ $\cdot$ $`\vec{E}_0`$| in the input file instead of the magnitude |$`\vec{\mu}_{0i}`$| as suggested above. In this manner, the code will account for a linearly polarized laser field. Note that if you provide |$`\vec{\mu}_{0i}`$ $\cdot$ $`\vec{E}_0`$|, the absorption spectrum calculated by **PROMDENS** corresponds to an effective absorption spectrum seen by the polarized laser pulse, not to the standard absorption spectrum measured in experiments. |
There was a problem hiding this comment.
Interesting. I guess in the future it would be nice to extend the code to handle this automatically?
There was a problem hiding this comment.
Yes, I didn't do it now because it would require all the components of TDM and setting up the field polarization. This is a bit simpler for now. But I agree it should be added in the future.
| self.weights[state] = self.tdm[state]**2*np.interp(self.de[state], self.field_ft_omega, self.field_ft)**2 | ||
|
|
||
| # normalization of all the weights | ||
| print(" - Normalization of the weights (sum of all weights equals 1)") |
There was a problem hiding this comment.
This is perhaps not needed? Not sure
| print(" - Normalization of the weights (sum of all weights equals 1)") |
There was a problem hiding this comment.
I would rather keep it. The normalization does not come from the equations. However, the weights can be tiny numbers, and you would anyway need to normalize them when processing the trajectories. Hence, I do it directly in the code, but I believe it is safer to mention that.
README.md updated with possibility to use polarization of the field (prompted by colleagues from the group).
PDAW analysis of the weights was not very consistent. I've now modified it such that it makes more sense.