Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Autodiff #107

Closed
antoine-levitt opened this issue Jan 19, 2020 · 5 comments
Closed

Autodiff #107

antoine-levitt opened this issue Jan 19, 2020 · 5 comments
Labels
discussion Discussion thread of broader scope

Comments

@antoine-levitt
Copy link
Member

Just opening this issue to collect thoughts on how to exploit AD. A good first step to explore how feasible this is is energy_ewald(), which is completely self-contained and should be very AD-friendly, both for forward and backward.
@mfherbst
Copy link
Member

Same is true (with probably easier analytic expressions on top) for the correction due to the PSP energy_nuclear_psp_correction.

@antoine-levitt antoine-levitt mentioned this issue Feb 4, 2020
Merged
@antoine-levitt antoine-levitt added the feature New feature or request label May 24, 2020
@mfherbst mfherbst added discussion Discussion thread of broader scope and removed feature New feature or request labels Jun 3, 2021
@mfherbst
Copy link
Member

mfherbst commented Jun 3, 2021

@niklasschmitz just wanted to point you at this thread ;)

@niklasschmitz niklasschmitz mentioned this issue Jun 7, 2021
Closed
2 tasks
@antoine-levitt
Copy link
Member Author

antoine-levitt commented Jul 8, 2021 •
edited
Loading

Roadmap, ordered by increased difficulty:

  • merge the HF PR
  • make an example out of the polarizability thing
  • replace compute_deltarho function
  • (Antoine) temperature
  • (Michael/Antoine) make an API for stresses
  • upstream as much as possible
  • reverse
  • benchmarks (when we have reverse)
  • higher derivatives

@antoine-levitt
Copy link
Member Author

Closing this in favor of a new one

@mfherbst
Copy link
Member

mfherbst commented Mar 4, 2022

see #614

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
discussion Discussion thread of broader scope
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@antoine-levitt @mfherbst