Try upgrading packages for dbias

[amyheather]

Dbias currently has error when loading the model.

I tried a solution from maniospas from dreji18/Fairness-in-AI#7 whichd did not work, but I should try again upgrading packages and using their function, as they got things working using recent packages

TypeError                                 Traceback (most recent call last)
Cell In[3], [line 1](vscode-notebook-cell:?execution_count=3&line=1)
----> [1](vscode-notebook-cell:?execution_count=3&line=1) from Dbias.text_debiasing import * 
      [3](vscode-notebook-cell:?execution_count=3&line=3) # returns unbiased recommendations for a given sentence fragment.
      [4](vscode-notebook-cell:?execution_count=3&line=4) run("Billie Eilish issues apology for mouthing an anti-Asian derogatory term in a resurfaced video.", show_plot = True)

File ~/mambaforge/envs/bias_dbias/lib/python3.8/site-packages/Dbias/text_debiasing.py:34
     [30](https://file+.vscode-resource.vscode-cdn.net/home/amy/Documents/stars/identify_bias/dbias/~/mambaforge/envs/bias_dbias/lib/python3.8/site-packages/Dbias/text_debiasing.py:30) classifier = pipeline('text-classification', model=model, tokenizer=tokenizer) # cuda = 0,1 based on gpu availability
     [32](https://file+.vscode-resource.vscode-cdn.net/home/amy/Documents/stars/identify_bias/dbias/~/mambaforge/envs/bias_dbias/lib/python3.8/site-packages/Dbias/text_debiasing.py:32) # ner model
     [33](https://file+.vscode-resource.vscode-cdn.net/home/amy/Documents/stars/identify_bias/dbias/~/mambaforge/envs/bias_dbias/lib/python3.8/site-packages/Dbias/text_debiasing.py:33) #pip install https://huggingface.co/d4data/en_pipeline/resolve/main/en_pipeline-any-py3-none-any.whl
---> [34](https://file+.vscode-resource.vscode-cdn.net/home/amy/Documents/stars/identify_bias/dbias/~/mambaforge/envs/bias_dbias/lib/python3.8/site-packages/Dbias/text_debiasing.py:34) nlp = spacy.load("en_pipeline")
     [36](https://file+.vscode-resource.vscode-cdn.net/home/amy/Documents/stars/identify_bias/dbias/~/mambaforge/envs/bias_dbias/lib/python3.8/site-packages/Dbias/text_debiasing.py:36) # masked language model
     [37](https://file+.vscode-resource.vscode-cdn.net/home/amy/Documents/stars/identify_bias/dbias/~/mambaforge/envs/bias_dbias/lib/python3.8/site-packages/Dbias/text_debiasing.py:37) tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")

File ~/mambaforge/envs/bias_dbias/lib/python3.8/site-packages/spacy/__init__.py:51, in load(name, vocab, disable, exclude, config)
     [30](https://file+.vscode-resource.vscode-cdn.net/home/amy/Documents/stars/identify_bias/dbias/~/mambaforge/envs/bias_dbias/lib/python3.8/site-packages/spacy/__init__.py:30) def load(
     [31](https://file+.vscode-resource.vscode-cdn.net/home/amy/Documents/stars/identify_bias/dbias/~/mambaforge/envs/bias_dbias/lib/python3.8/site-packages/spacy/__init__.py:31)     name: Union[str, Path],
     [32](https://file+.vscode-resource.vscode-cdn.net/home/amy/Documents/stars/identify_bias/dbias/~/mambaforge/envs/bias_dbias/lib/python3.8/site-packages/spacy/__init__.py:32)     *,
   (...)
     [36](https://file+.vscode-resource.vscode-cdn.net/home/amy/Documents/stars/identify_bias/dbias/~/mambaforge/envs/bias_dbias/lib/python3.8/site-packages/spacy/__init__.py:36)     config: Union[Dict[str, Any], Config] = util.SimpleFrozenDict(),
     [37](https://file+.vscode-resource.vscode-cdn.net/home/amy/Documents/stars/identify_bias/dbias/~/mambaforge/envs/bias_dbias/lib/python3.8/site-packages/spacy/__init__.py:37) ) -> Language:
     [38](https://file+.vscode-resource.vscode-cdn.net/home/amy/Documents/stars/identify_bias/dbias/~/mambaforge/envs/bias_dbias/lib/python3.8/site-packages/spacy/__init__.py:38)     """Load a spaCy model from an installed package or a local path.
     [39](https://file+.vscode-resource.vscode-cdn.net/home/amy/Documents/stars/identify_bias/dbias/~/mambaforge/envs/bias_dbias/lib/python3.8/site-packages/spacy/__init__.py:39) 
     [40](https://file+.vscode-resource.vscode-cdn.net/home/amy/Documents/stars/identify_bias/dbias/~/mambaforge/envs/bias_dbias/lib/python3.8/site-packages/spacy/__init__.py:40)     name (str): Package name or model path.
...
--> [774](https://file+.vscode-resource.vscode-cdn.net/home/amy/Documents/stars/identify_bias/dbias/~/mambaforge/envs/bias_dbias/lib/python3.8/typing.py:774)         return issubclass(cls, self.__origin__)
    [775](https://file+.vscode-resource.vscode-cdn.net/home/amy/Documents/stars/identify_bias/dbias/~/mambaforge/envs/bias_dbias/lib/python3.8/typing.py:775)     if cls._special:
    [776](https://file+.vscode-resource.vscode-cdn.net/home/amy/Documents/stars/identify_bias/dbias/~/mambaforge/envs/bias_dbias/lib/python3.8/typing.py:776)         return issubclass(cls.__origin__, self.__origin__)

TypeError: issubclass() arg 1 must be a class

[maniospas]

Hi, accidentally noticed this issue. This particular error is because you are using an older version of Python (I think the issubclass definition became looser later on). You will probably also have issues if you use any version other than exactly Python 3.11 with the hack I described. I get some issues in Python 3.13 due some of the tensorflow project's complications in releasing that cascaded through the ecosystem, and I believe I could not get 3.12 to work because it needs to be able to build spacy 3.2.0. Some dependency versions that were useful to set:

protobuf==4.25.8
spacy==3.2.0
spacy[transformers]

Edit: added dependencies.

[amyheather]

Thank you @maniospas ! That is very helpful.