Good morning everyone,
How is it possible to remove punctuation from user messages within a pipeline? I was removing them as a first step before doing tokenization within the tokenizer but it seems that if I do so entities get misaligned and stop working.
All the best,