stanojevic / ccgtools Goto Github PK

I attempted to run the cells in notebooks/demo.ipynb sequentially using Google Colab.
However, I encountered an issue when loading the pretrained RoBERTa parser.
I was able to use the same command without any issues before.
The error message I received is as follows:

RuntimeError: Error(s) in loading state_dict for Model:
Unexpected key(s) in state_dict: "embedder.model.embeddings.position_ids".

I would greatly appreciate it if you could fix this! Thank you!

I met a problem with loading pre-trained models (English-RoBERTa.ckpt)

File "/Conda/envs/ccgparser/bin/ccg-parser", line 8, in
sys.exit(parse())
File "/Conda/envs/ccgparser/lib/python3.7/site-packages/ccg/supertagger/main.py", line 87, in parse
parser = _load_and_setup_model(args, parsing=True)
File "/Conda/envs/ccgparser/lib/python3.7/site-packages/ccg/supertagger/main.py", line 28, in _load_and_setup_model
num_cpus=num_cpus)
File "/Conda/envs/ccgparser/lib/python3.7/site-packages/ccg/supertagger/parser.py", line 21, in init
self.model = Model.load_model(model_file)
File "/Conda/envs/ccgparser/lib/python3.7/site-packages/ccg/supertagger/model.py", line 238, in load_model
model = Model.load_from_checkpoint(checkpoint_path=model_file)
File "/Conda/envs/ccgparser/lib/python3.7/site-packages/pytorch_lightning/core/saving.py", line 143, in load_from_checkpoint
**kwargs,
File "/Conda/envs/ccgparser/lib/python3.7/site-packages/pytorch_lightning/core/saving.py", line 180, in _load_from_checkpoint
return _load_state(cls, checkpoint, strict=strict, kwargs)
File "/Conda/envs/ccgparser/lib/python3.7/site-packages/pytorch_lightning/core/saving.py", line 225, in _load_state
obj = cls(_cls_kwargs)
File "/Conda/envs/ccgparser/lib/python3.7/site-packages/ccg/supertagger/model.py", line 42, in init
self.lr = trial.suggest_loguniform("lr", 1e-5, 0.01)
File "/Conda/envs/ccgparser/lib/python3.7/site-packages/optuna/_deprecated.py", line 113, in wrapper
return func(*args, **kwargs)
File "/Conda/envs/ccgparser/lib/python3.7/site-packages/optuna/trial/_fixed.py", line 89, in suggest_loguniform
return self.suggest_float(name, low, high, log=True)
File "/Conda/envs/ccgparser/lib/python3.7/site-packages/optuna/trial/_fixed.py", line 79, in suggest_float
return self._suggest(name, FloatDistribution(low, high, log=log, step=step))
File "/Conda/envs/ccgparser/lib/python3.7/site-packages/optuna/trial/_fixed.py", line 138, in _suggest
distributions.check_distribution_compatibility(self._distributions[name], distribution)
File "/Conda/envs/ccgparser/lib/python3.7/site-packages/optuna/distributions.py", line 639, in check_distribution_compatibility
raise ValueError("Cannot set different distribution kind to the same parameter name.")
ValueError: Cannot set different distribution kind to the same parameter name.