Batching of determinants #69
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…unk of the connected. selection_step is now an mpi function
…on of the connected space
* pt2 selection now iterates over the full internal space and only a chunk of the connected. selection_step is now an mpi function * fixed doctest * for real this time * for real this time * some initial changes before adding parameter to chunk connected space by input size * spelling * doctest fix * Added functionality for user to specify the size of the working portion of the connected space * Local/global sorts seperate in the selection step + added few more tests * Changes from recent review
* Add constraint-based excitation functions (no testing yet) * small name change * now constrained excitation handles opposite spin ones as well; this is a starting point for tests * no way this is working first try? * first test passing * Minimal test added and passing * Constraint-based generation of connected dets from wf + new test * Now constraints working for beta electrons as well * black * black
* External chunk into clean (Exa-sCI#61) * Implementation naive algo * pt2 selection now iterates over the full internal space and only a chunk of the connected. selection_step is now an mpi function * fixed doctest * for real this time * for real this time * some initial changes before adding parameter to chunk connected space by input size * spelling * doctest fix * Added functionality for user to specify the size of the working portion of the connected space * Local/global sorts seperate in the selection step + added few more tests * Changes from recent review * Initial PR for chunking the external space (Exa-sCI#53) * pt2 selection now iterates over the full internal space and only a chunk of the connected. selection_step is now an mpi function * fixed doctest * for real this time * for real this time * some initial changes before adding parameter to chunk connected space by input size * spelling * doctest fix * Added functionality for user to specify the size of the working portion of the connected space * Local/global sorts seperate in the selection step + added few more tests * Changes from recent review * Constraint based triplet excitations (#7) * Add constraint-based excitation functions (no testing yet) * small name change * now constrained excitation handles opposite spin ones as well; this is a starting point for tests * no way this is working first try? * first test passing * Minimal test added and passing * Constraint-based generation of connected dets from wf + new test * Now constraints working for beta electrons as well * black * black * changes * build fix attempt * attempt 2 * Update test_python.yml * update requirements.txt * Update test_python.yml * remove enumerate * Update test_python.yml --------- Co-authored-by: Thomas Applencourt <[email protected]> * only master loads wave function and integrals (Exa-sCI#62) * only master loads wave function and integrals * added # because it was missing --------- Co-authored-by: Thomas Applencourt <[email protected]> Co-authored-by: Nathan W Johnson <[email protected]>
* remove useless decorator * added two electron parent class * Not all tests will pass (intdriven not implemented yet); detdriven on the fly pt2 is done * remove old tests for chunked connected space (no longer necessary since all pt2 done on the fly) * forgot to remove something * selection step modified to work with on the fly pt2 * remove leftover print statements * very naive integral driven on the fly pt2 * load balancing * Static load balancing, currently very slow. Optimize later * update comment * Integral-driven constraints for category C; before re-factoring * Integral-driven constrained PT2 done for category D * Category E singles, onto doubles * Integral-driven constraints done for PT2, all categories; onto re-formatting * doctest * Some comments to make integral-driven code more readable * load balanced constraints; need to test for multiple ranks still * remove print * commit before testing on other branch * commit * parallel working * merge nathan's changes (#10) * External chunk into clean (Exa-sCI#61) * Implementation naive algo * pt2 selection now iterates over the full internal space and only a chunk of the connected. selection_step is now an mpi function * fixed doctest * for real this time * for real this time * some initial changes before adding parameter to chunk connected space by input size * spelling * doctest fix * Added functionality for user to specify the size of the working portion of the connected space * Local/global sorts seperate in the selection step + added few more tests * Changes from recent review * Initial PR for chunking the external space (Exa-sCI#53) * pt2 selection now iterates over the full internal space and only a chunk of the connected. selection_step is now an mpi function * fixed doctest * for real this time * for real this time * some initial changes before adding parameter to chunk connected space by input size * spelling * doctest fix * Added functionality for user to specify the size of the working portion of the connected space * Local/global sorts seperate in the selection step + added few more tests * Changes from recent review * Constraint based triplet excitations (#7) * Add constraint-based excitation functions (no testing yet) * small name change * now constrained excitation handles opposite spin ones as well; this is a starting point for tests * no way this is working first try? * first test passing * Minimal test added and passing * Constraint-based generation of connected dets from wf + new test * Now constraints working for beta electrons as well * black * black * changes * build fix attempt * attempt 2 * Update test_python.yml * update requirements.txt * Update test_python.yml * remove enumerate * Update test_python.yml --------- Co-authored-by: Thomas Applencourt <[email protected]> * only master loads wave function and integrals (Exa-sCI#62) * only master loads wave function and integrals * added # because it was missing * nathan PR (#9) * External chunk into clean (Exa-sCI#61) * Implementation naive algo * pt2 selection now iterates over the full internal space and only a chunk of the connected. selection_step is now an mpi function * fixed doctest * for real this time * for real this time * some initial changes before adding parameter to chunk connected space by input size * spelling * doctest fix * Added functionality for user to specify the size of the working portion of the connected space * Local/global sorts seperate in the selection step + added few more tests * Changes from recent review * Initial PR for chunking the external space (Exa-sCI#53) * pt2 selection now iterates over the full internal space and only a chunk of the connected. selection_step is now an mpi function * fixed doctest * for real this time * for real this time * some initial changes before adding parameter to chunk connected space by input size * spelling * doctest fix * Added functionality for user to specify the size of the working portion of the connected space * Local/global sorts seperate in the selection step + added few more tests * Changes from recent review * Constraint based triplet excitations (#7) * Add constraint-based excitation functions (no testing yet) * small name change * now constrained excitation handles opposite spin ones as well; this is a starting point for tests * no way this is working first try? * first test passing * Minimal test added and passing * Constraint-based generation of connected dets from wf + new test * Now constraints working for beta electrons as well * black * black * changes * build fix attempt * attempt 2 * Update test_python.yml * update requirements.txt * Update test_python.yml * remove enumerate * Update test_python.yml --------- Co-authored-by: Thomas Applencourt <[email protected]> * only master loads wave function and integrals (Exa-sCI#62) * only master loads wave function and integrals * added # because it was missing --------- Co-authored-by: Thomas Applencourt <[email protected]> Co-authored-by: Nathan W Johnson <[email protected]> --------- Co-authored-by: Thomas Applencourt <[email protected]> Co-authored-by: Nathan W Johnson <[email protected]> * comments * before pr --------- Co-authored-by: Thomas Applencourt <[email protected]> Co-authored-by: Nathan W Johnson <[email protected]>
* initial flexible typing * basic excitation drivers for tuple dets * fix build * move over constraint + phase drivers * mid re-factoring; commit before changing test files to new types * tests passing for tuple determinant * black * Excitation and phase classes fully moved to determinant * doctest * doctest * no more doctests please * Some re-factoring; generic methods moved to generic determinant class * black * Update test_python.yml Need 3.10 for .bit_count() * new io load_wf to handle bitstring representation * Overloaded spindet logical operations, Determinant NamedTuple class re=implemented to be more flexible * doctest + kwargs option for Determinant type * finish updating __new__() for Determinant * TIL about reverse operators * overloaded operators for bitstring type * re-factored types fully working for tuples * docs * io
Loopless phase algorithm. Now agnostic to spin determinant type
|
Thx! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.