udify
udify copied to clipboard
Hyperparticle
A single model that parses Universal Dependencies across 75 languages. Given a sentence, jointly predicts part-of-speech tags, morphology tags, lemmas, and dependency trees.
When I try to run a clean checkout of UDify, I get the following error: ``` (udify-venv) fran@tlazolteotl /var/lib/home/fran/source/udify $ python predict.py udify-model.tar.gz data/UD_Kiche-IU/quc_iu-ud-test.conllu logs/pred.conllu --eval_file logs/pred.json Traceback (most recent...
Hello, I am training a udify model only for Korean where I train only on the Korean data from UD2.3. However, I am running into the following issue. The same...
Hi, has anybody looked into training a version of udify with XLM-RoBERTa? Seems like it could help with the low-resource languages in multilingual BERT so I'm planning on giving it...
I'm trying to run udify on some data and have followed the instructions, e.g. ```bash $ git clone https://github.com/Hyperparticle/udify $ pip install -r ./requirements.txt $ curl --remote-name-all https://lindat.mff.cuni.cz/repository/xmlui/bitstream/handle/11234/1-3042{/udify-model.tar.gz,/udify-bert.tar.gz} ``` I...
Hello, I am trying to use the XLMRoberta model instead of BERT and I made the following changes to the `bert_pretrained.py`: ``` from transformers import XLMRobertaTokenizer from transformers import XLMRobertaModel,...
I'm feeding this raw input to the predict.py - "Il est assez sûr de lui pour danser et chanter en public ." by setting --raw_text flag and since I want...
Is it possible to predict multi-word expression (MWE) from raw text? I run `predict.py` with option `--raw_text` to find that MWE cannot be predicted. For example, in Italy, "della" is...
This is for future reference, for anyone working off of this model's edge score outputs. There is a semantic naming bug in the dependency decoder that actually switches the `head_arc_representation`...
Hi! Thanks for releasing the library. I'm encountering "nan" loss on training with the following commit, which I think is the most recent version: `60f35edc52862109555f4acf66236becc29705ad` Here are instructions to reproduce:...
It would be useful to integrate the [UDify model](https://github.com/Hyperparticle/udify) directly into AllenNLP as a PR, as the code merely extends the library to handle a few extra features. Since the...
Metadata
Owner
Metadata
A single model that parses Universal Dependencies across 75 languages. Given a sentence, jointly predicts part-of-speech tags, morphology tags, lemmas, and dependency trees.