Datasets:

Mardiyyah
/

TAPT_data_V2_split

language:
  - en
license: apache-2.0
size_categories:
  - 1K<n<100K
pretty_name: Biomedical-Articles
dataset_info:
  features:
    - name: PMCID
      dtype: string
    - name: Title
      dtype: string
    - name: Sentences
      dtype: string
  splits:
    - name: train
      num_bytes: 2636198
      num_examples: 9733
    - name: validation
      num_bytes: 560363
      num_examples: 1945
  download_size: 902746
  dataset_size: 3196561
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: validation
        path: data/validation-*

Dataset Card for "TAPT_data_V2_split"

Dataset Description

Generated using EPMC API PMCIDs:

Train-Val split ratio 80:20. Data has been split on PMCIDs for traceability.

More Information Needed