This PR addressed #14 by introducing the feature of utilizing unique neural networks for each neighbor type in the model, along with several enhancements and bug fixes:
Implement the usage of unique neural networks (NNs) for each neighbor type, allowing for more nuanced interactions in the model: 36e95c5, 3fcaa94, 1bdd8e8, 6dba467, 3c01dff
Create and update SLURM scripts for managing job submissions under different computational scenarios: 25072dd, 706f1c2
Enable dynamic specification of target lists via script arguments to enhance flexibility in experiments: bc06fd9
Refactor key utility functions for better maintenance and readability, adding new wrappers for adjacency and incidence matrices: 3ed96f6
Update utility file naming: 8f2a5e5, a5cfaa7
Add documentation and tests to the sparse_to_dense() function for better understanding and reliability: bc3a45e
This effort has streamlined several aspects of our computational setup and enriched the model's capability to handle complex network structures. We are monitoring ongoing runs to evaluate the impact of these changes on model performance.
This PR addressed #14 by introducing the feature of utilizing unique neural networks for each neighbor type in the model, along with several enhancements and bug fixes:
This effort has streamlined several aspects of our computational setup and enriched the model's capability to handle complex network structures. We are monitoring ongoing runs to evaluate the impact of these changes on model performance.