Fenia Christopoulou

Hello, could you check previous issues for this? I think the tokenizer wasn't run properly, so might need to check what happened there.

Hey there! May I ask what is the task you target exactly? Also, what kind of annotations you have? For instance, you have only entities (mentions) or concept annotations as...

Hi again! Yes, sure, you can use the model that way, but you need to discard the concept (E) nodes completely as well as edges between them and other nodes....

Hello there! Could you check if Genia-tagger was properly installed? I added some additional instruction in the readme file, as the original tagger link was broken.

It should be fine, could you make sure you have ruby installed and perhaps run the pre-processing again?

Hi, have you installed GeniaTagger as described in the readme?

The data_processing script produces a specific data format used for this paper. You can convert the data to a format of your choice by editing the script, if you want...

Hello, I translated your comment to "Can you provide the data set I'm reproducing this project?". As mentioned, you can edit the `process.py` script (https://github.com/fenchri/edge-oriented-graph/blob/master/data_processing/process.py) in order to convert the...

Thanks @stas00 for having a look and apologies for the late reply. Indeed, the fix resolves the issue! :tada: On a related note, the computation happening [here](https://github.com/huggingface/transformers/blob/main/src/transformers/trainer.py#L1654) seems to chop...

Hello, please make sure `ruby` is installed. Do a fresh install of the packages and try again? In case you are running this on Windows, I am not sure I...