metadata

configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*
dataset_info:
  features:
    - name: sentenceA
      dtype: string
    - name: sentenceB
      dtype: string
    - name: label
      dtype: float64
  splits:
    - name: train
      num_bytes: 209357089
      num_examples: 71582
    - name: test
      num_bytes: 69736578
      num_examples: 23861
  download_size: 162843573
  dataset_size: 279093667
license: apache-2.0
task_categories:
  - sentence-similarity
language:
  - en
tags:
  - economics
  - finance
  - politics
size_categories:
  - 10K<n<100K

Dataset Card for "econo-pairs"

Econo-pairs is a dataset made of pairs of sentences extracted from worldwide central banks speeches and other public financial institutions. Each pair is labelled as a positive (1) or negative (0) one. Positive pairs are made of sentences extracted from the same speech. Negative pairs are made of sentences extracted from a random other speech.

Credits

Full credits to the Bank of International Settlements for making the resource available through the internet. Credits to Samuel Chaineau for making it available on HF Hub.

More Information needed