Is there anything wrong the tokenizer?
The rnafm can work well, but the mrnafm cannot. Is there anything wrong the tokenizer?
Traceback (most recent call last):
File "/home/.../try.py", line 4, in <module>
tokenizer = RnaTokenizer.from_pretrained('multimolecule/mrnafm')
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/../python3.11/site-packages/transformers/tokenization_utils_base.py", line 2271, in from_pretrained
return cls._from_pretrained(
^^^^^^^^^^^^^^^^^^^^^
File "/home/.../python3.11/site-packages/transformers/tokenization_utils_base.py", line 2505, in _from_pretrained
tokenizer = cls(*init_inputs, **init_kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/.../python3.11/site-packages/multimolecule/tokenisers/rna/tokenization_rna.py", line 85, in __init__
raise ValueError("Codon and nmers cannot be used together.")
ValueError: Codon and nmers cannot be used together.
Sorry, we have updated the model file with our new codebase, but they are not yet available on PyPi.
Could you try install the latest version on GitHub via pip install git+https://github.com/DLS5-Omics/MultiMolecule and see if it works?
We are finalising the v0.0.5, and should update PyPi in the coming days.
It works. Thank you!
No worries. We do apologise for the inconvenience it has caused.
We tried to release v0.0.5 in the past week, but certain test cases keep failing.
We need extra time to ensure that it is not an issue with the MultiMolecule, and it seems everything is working now.
We just need to run a few more tests before we can release.
We have just released v0.0.5!
This should fix the issue.