search

Hyperparticle / udify

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.
https://arxiv.org/abs/1904.02099
MIT License
219 stars 56 forks source link

Training udify model for Russian. #21

Closed Aditi138 closed 3 years ago

Aditi138 commented 3 years ago

Hello, I am training a udify model only for Russian where I train only on the Russian data from UD2.3. However, I am running into the following issue. The same code runs fine on other languages from UD. Traceback (most recent call last): File "train.py", line 69, in train_model(train_params, serialization_dir, recover=bool(args.resume)) File "/usr1/home/user/anaconda2/envs/py36/lib/python3.6/site-packages/allennlp/commands/train.py", line 226, in train_model cache_prefix) File "/usr1/home/user/anaconda2/envs/py36/lib/python3.6/site-packages/allennlp/training/trainer_pieces.py", line 65, in from_params model = Model.from_params(vocab=vocab, params=params.pop('model')) File "/usr1/home/user/anaconda2/envs/py36/lib/python3.6/site-packages/allennlp/common/from_params.py", line 365, in from_params return subclass.from_params(params=params, extras) File "/usr1/home/user/anaconda2/envs/py36/lib/python3.6/site-packages/allennlp/common/from_params.py", line 386, in from_params kwargs = create_kwargs(cls, params, extras) File "/usr1/home/user/anaconda2/envs/py36/lib/python3.6/site-packages/allennlp/common/from_params.py", line 133, in create_kwargs kwargs[name] = construct_arg(cls, name, annotation, param.default, params, extras) File "/usr1/home/user/anaconda2/envs/py36/lib/python3.6/site-packages/allennlp/common/from_params.py", line 257, in construct_arg value_dict[key] = value_cls.from_params(params=value_params, subextras) File "/usr1/home/user/anaconda2/envs/py36/lib/python3.6/site-packages/allennlp/common/from_params.py", line 365, in from_params return subclass.from_params(params=params, extras) File "/usr1/home/user/anaconda2/envs/py36/lib/python3.6/site-packages/allennlp/common/from_params.py", line 388, in from_params return cls(kwargs) # type: ignore File "/usr1/home/user/udify/udify/models/tag_decoder.py", line 106, in init div_value=4.0) File "/usr1/home/user/anaconda2/envs/py36/lib/python3.6/site-packages/torch/nn/modules/adaptive.py", line 116, in init raise ValueError("cutoffs should be a sequence of unique, positive " ValueError: cutoffs should be a sequence of unique, positive integers sorted in an increasing order, where each value is between 1 and n_classes-1

Hyperparticle commented 3 years ago

Were you able to solve the issue?

Aditi138 commented 3 years ago

Yes, I was able to!

euhkim commented 2 years ago

Yes, I was able to!

How did you solve it?