-
Notifications
You must be signed in to change notification settings - Fork 12
Brief Summary of development objectives
bnebgen-LANL edited this page Apr 3, 2023
·
1 revision
h5 data module -permenant train/test splits HIPNN interface
Facilitate dynamic loading+ error handling
- Better h5 management 1a) Data reducing process 1b) Parsl data integration
- pySCF + PyPI
- Multiple QM workers on a block
- Fix training: each model is its own task
- make molecule custom data time
- automatically held out test set
- make simple bootstrap dataset:
- Write a parsl configuration for a dedicated nodelist or a local system.
- get cpu and GPU tasks to run simultaniously on a node.