roberta-base + DAPT + Domain-Specific QA
Objective:
This is Roberta Base with Domain Adaptive Pretraining on Movie Corpora --> Then a changed head to do the SQuAD Task. This makes a QA model capable of answering questions in the movie domain.
https://huggingface.co/thatdramebaazguy/movie-roberta-base was used as the MovieRoberta.
model_name = "thatdramebaazguy/movie-roberta-squad"
pipeline(model=model_name, tokenizer=model_name, revision="v1.0", task="question-answering")
Overview
Language model: roberta-base
Language: English
Downstream-task: QA
Training data: imdb, polarity movie data, cornell_movie_dialogue, 25mlens movie names, SQuADv1
Eval data: MoviesQA (From https://github.com/ibm-aur-nlp/domain-specific-QA)
Infrastructure: 1x Tesla v100
Code: See example
Hyperparameters
Num examples = 88567
Num Epochs = 10
Instantaneous batch size per device = 32
Total train batch size (w. parallel, distributed & accumulation) = 32
Performance
Eval on MoviesQA
- eval_samples = 5032
- exact_match = 51.64944
- f1 = 65.53983
Eval on SQuADv1
- exact_match = 81.23936
- f1 = 89.27827
Github Repo:
