Transformer(huggingface)
B) Pipeline
AutoModelForSequenceClassification 은 뭐하는 거지..?
C) Tranining Model from Scratch
This is a good approach to take if you have a lot of data and it is very different from the pretraining data used for the available models. However, it also requires considerably more compute resources to pretrain a language model than just to fine-tune an existing one.