Eugene Yang

Unable to resume CoCondenser pretraining

7

The model checkpoints seem to be hard-coded as the BertForMaskedLM and are unable to load but to the CoCondensor class. Adding the following attributes in the initialization can surpass the...

Specific paper to cite?

2

Is there any paper that you recommend us to cite if we use Tevatron in a research paper?

Lock for writing to the cache files

2

I sometimes have multiple processes or multiple machines accessing the same storage cluster that hosts the cache directory of `ir_datasets`. If multiple processes decide to download the same dataset at...

Enable pip install?

5

I'm wondering are you planning to add a `setup.py` to allow `pip install`? I could draft one if you want :)

https://github.com/eugene-yang/tarexp/blob/main/tarexp/dataset.py#L254 It really should be using `self.__class__(...)`