metadata
configs:
- config_name: default
data_files:
- split: train
path: pubmed_qa_train.csv
- split: validation
path: pubmed_qa_validation.csv
- config_name: truncated
data_files:
- split: train
path: train_truncated.csv
- split: validation
path: validation_truncated.csv
- config_name: truncated_512
data_files:
- split: train
path: train_truncated_512.csv
- split: validation
path: validation_truncated_512.csv
- config_name: truncated_512_augmented
data_files:
- split: validation
path: validation_truncated_512-augmented.csv