Open RasmusOrsoe opened 4 months ago
I got this to run locally on my machine by passing the config dict variables into the GraphNetDataModule arguments and calling the train_dataloader and val_dataloader properties. This implementation works with both .db files and parquet paths containing separate directories for pulse maps and truth tables. I will submit a pull request soon.
@OscarBarreraGithub Sounds great! I assigned you to the issue.
Since #642 we've gotten a much more elegant way of producing dataloaders for training and inference. We should update our training scripts to use this new method instead, and strongly consider to remove the old method, as discussed in #467.