-
Hello,
I saw you your UCCA parsing model can be trained only with UCCA parsings of German or English sentences.
I would like to train it also with UCCA parsings of French and Russian as well and was…
-
Hello,
Have you seen substantial improvements when using the large_ucca configurations vs. the base_ucca configurations?
I have gone over your paper and haven't found any mention of the comparison b…
-
Hello,
I am working on a project involving an conversion between mrp files representing ucca graphs and xml files.
I am trying to run the conversion script (`python3 main.py --read mrp --write ucca …
-
Hello,
I am using the package in order to convert from mrp to ucca graphs (in xml), and for some mrp files, with top nodes that have remote edges going out of them, the code crashes and prompts the f…
-
we still need to provide a validator for MRP graphs to participants. beyond making sure that a file can be parsed in the MRP file format, how far should we push notions of wellformedness? for exampl…
oepen updated
3 years ago
-
Hey,
When Trying to train the model for UCCA, I keep getting this error:
```
OSError: Model name 'xlm-roberta-base' was not found in model name list
(bert-base-uncased, bert-large-uncased, bert-b…
-
I want to compute SAMSA score of corpus.. Can I use EASSE Technique to find out Perfect SAMSA score to Evaluate corpus?
-
To train PERIN on a new dataset (not from MRP 2020), a companion file currently needs to be specified for the new text. Is this a real requirement, or is it just a result of the implementation? Does P…
-
See, for example, https://travis-ci.org/github/UniversalConceptualCognitiveAnnotation/UCCA_English-EWT/jobs/749979825
Many new failures are detected by the validation, but looking at the data, they…
-
Hello,
I would like to use your model to parse UCCA trees of passages in Russian.
Do you happen to know if your model supports other languages?
Thank you!