Dipanjan (DJ) Sarkar

That's because you are creating the prediction labels in the wrong way. You can follow the process from here https://github.com/dipanjanS/adv_nlp_workshop_odsc_europe22/blob/main/02_NLP_Applications_Text_Classification_Deep_Learning_CNN_Models.ipynb and then pass the outputs to the label encoder.

Yes unfortunately TensorFlow has removed that function. You can use the Model Evaluation code here and reuse the same in the other model https://github.com/dipanjanS/adv_nlp_workshop_odsc_europe22/blob/main/02_NLP_Applications_Text_Classification_Deep_Learning_CNN_Models.ipynb Hope this helps. On Wed, Jul...

The reason for that is two-fold 1. Spacy's CLI wasn't matured yet at the time of the book's release 2. Sometimes for proxy and other internal environments the CLI download...

Yes sadly this is a known issue even with the contractions package on PyPI. The whole regex pattern framework used for this will need to be modified I am guessing....

It's not a publicly available library, there is a `model_evaluation_utils.py` file which is imported in the code and used

Could you please mention more details like which notebook, which folder \ chapter and files you are using, and even show the code causing the error. It is difficult to...

That would be more of a spacy issue I think, as far as I know typically any lemmatizer returns text in lower case. You might consider checking `nltk` but I...

yes definitely, that would be a good point to mention I think

Thanks, yes we are aware of this happening due to the deprecation of the `Merge` API, I will be pushing this change soon in my 2nd edition of the text...

Would be great to know if this might support Python 3 anytime soon?