Training a custom model with span categorizater instead of NER

[Oumayma68]

Since the NER component doesn't work with overlapping spans, I was wondering if it's possible to train a custom model with span categorizer instead of ner ? if it is possible how to do it?
For eg, I have this Sentence: "The launch of the new iPhone was a major event in the tech industry."
I want to extract the following entities/spans:
"the new iPhone" [Product]
"The launch of the new iPhone" [Event]

[rmitsch]

Hi @Oumayma68, spancat is a trainable component. Check out this example project (or some of the other publicly available resources).