Developing task parallelization #331
Draft
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
* add multiple branches to Base model * adjust based on Max's new datasets with dataset_name * add json for compute grad energy physics-informed force prediction * revert time change * reverse prior change * Add check to ensure correct output dim for energy in physics-informed force prediction * Implement the check after setting var_config * revert change * Shorten comment --------- Co-authored-by: Rylie Weaver <[email protected]> Co-authored-by: Rylie Weaver <[email protected]>
* data attributes updated for consistency across datasets * non-normalized chemical composition added as data attribute * download_dataset.sh added for transition1x example * download dataset flag updated * scripts updated * development of tranistion1x scripts continues * transiton1x scripts completed * black formatting fixed * printouts removed * parallelizatin of data reading introduced * blsck formatting fixed * detach().clone() used to defined normalized energy per atom and black formatting * add compute_grad_energy=False as explicit argument * add data name as attributed to each data object * compute_grad_energy is parsed as input argument with default value set to False * edge_index, edge_attr, and edge_shifts explicitly itnroduced in the definition of the Data object * changed data.force into data.forces for ani1x and qm7x examples * smiles_string added as data attribute * remove redundant logic on energy normalization from omat24 example * force threshold value increased to 1000 for ani-1x * Reverted smiles_utils.py to version from commit 3c3c434 * xyz2mol functionalities put in a separate file * download dataset script added for qm7x * renamed data.force as data.forces in ani1x * natoms converted into a tensor * verbosity level ntroduced for ani1x * Z corrected into atomic_numbers for qm7x example * bug fixed for data attributes in transition1x * try-except in transition1x rescoped * transform coordinates fixed in transition1x * iterate_tqdm used in utils.create_graph_data for transition1x example * total_enerfgy replaced with energy --------- Co-authored-by: Massimiliano Lupo Pasini <[email protected]> Co-authored-by: Massimiliano Lupo Pasini <[email protected]>
…ion of linear regression term (ORNL#324) * normalization of chemical composition histogram removed from calculation of linear regression term * energy and forces added as attrbutes to check in the graph samples
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.
This is for developing task parallelization to enable parallel execution of multi-task heads:
We use DeviceMesh to manage subgroup DDP communications.
It is still a work in progress.