search

ORNL / HydraGNN

Distributed PyTorch implementation of multi-headed graph convolutional neural networks
BSD 3-Clause "New" or "Revised" License
61 stars 27 forks source link

Testing hpo packages #218

Closed allaffa closed 6 months ago

allaffa commented 6 months ago

This PR introduces three main examples for hyperparameter optimization (HPO) using the QM9 dataset:

  1. qm9_optuna.py: HPO example using optuna
  2. qm9_deephyper.py: HPO example running in sequential using deephyper
  3. qm9_deephyper_multi.py: HPO example running in parallel using deephyper
jychoi-hpc commented 6 months ago

Can we save this qm9 as a separate directory (say, qm9_hpo)? We often use qm9 as a minimum case. It would be great if we make "qm9" as simple as possible. How do you think?

allaffa commented 6 months ago

Can we save this qm9 as a separate directory (say, qm9_hpo)? We often use qm9 as a minimum case. It would be great if we make "qm9" as simple as possible. How do you think?

@jychoi-hpc agree.

allaffa commented 6 months ago

@pzhanggit I also added num_headlayers and dim_headlayers as hyper-parameters