Psi 4 integration #170
Psi 4 integration #170
Conversation
# Conflicts: # moldesign/_tests/test_data_structures.py # moldesign/method.py
# Conflicts: # DockerMakefiles/Moldesign.yml # DockerMakefiles/NWChem.yml # moldesign/helpers/__init__.py # moldesign/molecules/notebook_display.py # moldesign/utils/__init__.py # moldesign/viewer/viewer3d.py # scripts/conda_deploy.sh
# Conflicts: # moldesign/_tests/test_mm.py # moldesign/interfaces/openbabel.py # moldesign/models/amber.py # moldesign/models/nwchem.py # moldesign/molecules/trajectory.py # moldesign/utils/__init__.py # moldesign/utils/classes.py # moldesign/utils/exportall.py # moldesign/utils/exports.py # moldesign/utils/exportutils.py
…"'dipole'" to be returned. Syntax for "'dipole'" collection will be changed.
…ith minimizing molecules using qm packages. The gradient will be assigned to the wavefunction using the constructor. A minimize module was added to the Psi4Potential. It calls the qsi4.opt function and saves the trajectory. Other appropriate changes were made to the Psi4Potential to facilitate execuation of this function. A function was added to the Psi4Model to save a Psi4 optimization history as a MDT minimization trajectory. The the module was added outside the Psi4Potential class to allow for a user-defined Psi4 optimization history to be added as a minimization trajectory. The Psi4interface module was also changed to allow for a user defined psi4history to be added.
…nstead of a string. The psi4_interface psi4_to_mdt function was written and now returns molecules. Changes were made to the initiation of the multiplicity in the molecule class. The Psi4 model was changed to accomodate charge and multiplicities from psi4 molecules. The interface was prepared to have a mdt wavefunction initiation function added later.
…ion was added to the psi4 model. The molecule class was prepared for addding molecular vibrations.
--testall
…calculations. These changes consisted of handling of units and data structuring of certain quantities.
…n a skeleton for saving normal mode displacements to atoms. The skeletal additions to the atoms will need to be reviewed and rendered functional.
|
Joseph O'Brien seems not to be a GitHub user. You need a GitHub account to be able to sign the CLA. If you have already a GitHub account, please add the email address used for this commit to your account. You have signed the CLA already but the status is still pending? Let us recheck it. |
|
Hey @obrien951 and @loriab - Huge apologies for not getting to this yet. @ktbolt and I are working through getting a working version of MDT 0.8 out into the wild, and this will be coming up shortly afterwards. |
|
I'm behind on this too. Good luck with MKL 0.8 prep. |
|
@obrien951 - I'm FINALLY taking a look at this. And it looks good! I'm making a few changes to update it for the most recent version of the code (mostly just changed import paths). One more thing before it gets merged - can you agree to the contributor agreement here? It gives us permission to integrate the code into the project and manage it as part of the codebase. |
|
Hi Aaron, glad to see how much this has come along since I last checked the repo! Sorry about the CLA. I watched Joe do the agreement process, but it must not have gone through. But I see now that you force pushed to master, perhaps, or something that has confounded this PR's history? Is the thing to do to start from a current master, cherry-pick over @obrien951's commits (should preserve author info), then submit a new PR, trying the CLA again? An update on the Psi side, All of Joe's changes are at the bottom of psi4/psi4#834, which additionally makes a lot more vibrational analysis info available for your taking. Look forward to seeing you next week! |
Cumulative work on the MDT PSI4 interface. Be advised to find commits from multiple branches.
Handles optimizations, energies, one-electron properties, and plotting of normal mode displacement vectors.