Transformers pipelines allow everyone to easily add functionality.
It is as simple as just applying the pipeline once it is instantiated.
This allows users to:
Store metadata about the pipeline automatically
Automatically apply a number of HuggingFace pipelines with ease
Running it on a dataset:
from transformers import pipeline pipeline = pipeline("automatic-speech-recognition", model="facebook/wav2vec2-base-960h", device=0) ds.apply_transformers_pipeline( text_fields, pipeline )