Sean MacAvaney

Yeah, all the model weights an so on are stored in `~/data/onir/models/default/{ranker}/{vocab}/{trainer}/{train_dataset}/weights/` -- there you'll find 3 files-- one corresponding to the initial weights, one for the optimal epoch from...

Ah, sorry I misunderstood what you meant by export. OpenNIR was designed primarily with a CLI in mind. If you have other queries you want to run on the same...

The `sledge-med.p` file (and all weight files from OpenNIR, for that matter) is just from `torch.save` -- a pickle-encoded dict of pytorch tensors, if I recall properly. To load into...

You're trying to load up a `transformers` version of it, right? If so, this should do the trick! https://colab.research.google.com/drive/1t5UdW2Jebue1php888ldDll6yG5jQXQQ?usp=sharing (based off starting point from link above.) I have not tested...

Does this meet your needs too, @thigm85?

Hi @timbmg, Yeah, the 0th class corresponds to the relevance score (using the convention from Nogueira et al) \- sean

> However, the output seems not over-perform the classic BM25's result. Could you please verify if the uploaded sledge-med.p is effective or the instructions in the shared google doc are...

Weird, maybe there's some problem with the Colab example I tried putting together. But I also suspect that the model isn't so robust to adversarial text like "dog cat" --...

This should do the trick then! Here's a colab link: https://colab.research.google.com/drive/12EdgWMKbMJxmR8XrLUr74PbASsfI8g6N?usp=sharing And the code: ```python import pandas as pd import pyterrier as pt if not pt.started(): pt.init() import onir_pt sledgez...

Hi- can you clarify whether you're interested in using a different model initialisation (e.g., changing `bert-base-uncased` to something else) or using a model that's already been fully tuned for ranking?