ersilia-os / zaira-chem

Automated QSAR based on multiple small molecule descriptors
GNU General Public License v3.0
27 stars 10 forks source link

Memory error in grover embedding #20

Open marcostorrework opened 1 year ago

marcostorrework commented 1 year ago

Describe the bug ZairaChem fails when trying to fit a model with an input file of 37637 rows. It fails because of running out of memory when calculating the Grover Embedding (model eos7w6n). This happens on a laptop with 16 GB RAM.

To Reproduce Steps to reproduce the behavior:

  1. Use a computer with 16 GB RAM
  2. Download the attached example file "train.csv"
  3. Create an empty directory "model"
  4. conda activate zairachem
  5. zairachem fit -i train.csv -m model

Expected behavior ZairaChem is supposed to fit a model and end without errors

Screenshots Log with error attached

Desktop (please complete the following information):

Additional context Running on a laptop with 16 GB RAM

Log with error: fit_20221220_0431.log Example input file to reproduce: train.csv