search

gstoica27 / ZipIt

A framework for merging models solving different tasks with different initializations into one multi-task model without any additional training
MIT License
280 stars 24 forks source link

Multidataset model training #22

Open ElliotStein opened 11 months ago

ElliotStein commented 11 months ago

Hi all,

Firstly, congrats on the excellent research!

I would like to reproduce Table 3 of the paper and need to first train the models to be merged. However, mmultids_trainer.py requires a checkpoint_path argument which doesn't make sense to me. I thought the idea was to train each model from scratch, not a pre-trained or shared initialisation? Why do we need this argument?

As a side note, if possible, it would be great if you could make pre-trained weights publicly available to make reproducibility more efficient.

Thanks!

ElliotStein commented 11 months ago

*This is in relation to the mmultids_trainer.py file in /non_imnet_training_scripts, rather than /training_scripts. Though I'm not clear on the difference between the two.