Hello, I am wondering if anyone has tried to integrate Spark NLP into rasa’s pipeline and if it is worth it to do so ? Thank you.
This should be possible with custom component if you are using for NLU pipeline. If you are looking at handling the log stream you can also write a custom event broker or use one of the existing ones.
Thank you for your reply, I know It is possible with custom components, I just wanted to check if anyone has tried it and/or can share an implementation example (some custom component like SparkFeaturizer or SparkTokenizer) and if the rasa team plans to have Spark based components supported in further rasa releases.