Pere Lluis
Hi, could you please provide the code/text to replicate the issue? I've tried adding a new line to the example in the README, but it still outputs the same triplets....
If I am not mistaken, that comes from a breaking change with an update of the datasets library, in which the path is transformed to a DataFilesList, probably in order...
Since that is the version on reqs, I am closing this.
Yes, you are right, it is a typo. I believe the confusion comes from the fact that it was evaluated on 220 relations (since 10 were missing from the set...
Hi there! Those are indeed very good examples of the shortcomings of REBEL. Negation is usually hard to handle by many models, and it seems REBEL is also vulnerable to...
Sorry, I was a bit AFK during August. I am afraid it would be very hard to deal with this at inference time. You could try to run an NLI...
Hi @Theosagie. I am sorry but I didn't understand how the data you shared can be framed as a Relation Extraction task, could you perhaps give some example? In any...
I am sorry about that, I think at some point there was a version incompatibility regarding the use of hydra/omegaconf with the checkpointing. Since you are already loading the module...
Working on [CNNs branch](https://github.com/CoVital-Project/Spo2_evaluation/tree/CNNs). Currently on Semcova dataset: Lowest MAE for Spo2 on test set, paper reported 1.1%: - Mobilenet_v2: 0.85% - Resnet18: 1.05% - VGG13: 0.95% - VG16: 0.87%...
Maybe add the NN models in the modelling folder?