README.md

metadata

language:
  - en
license: mit
size_categories:
  - 10K<n<100K
task_categories:
  - translation
  - summarization
  - text-generation
dataset_info:
  - config_name: translation
    features:
      - name: x
        dtype: string
      - name: 'y'
        dtype: string
    splits:
      - name: train
        num_bytes: 98465532
        num_examples: 50000
      - name: valid
        num_bytes: 10120970
        num_examples: 5000
      - name: test
        num_bytes: 200120
        num_examples: 100
    download_size: 66841138
    dataset_size: 108786622
  - config_name: dialogue
    features:
      - name: x
        dtype: string
      - name: 'y'
        dtype: string
    splits:
      - name: train
        num_bytes: 6089377
        num_examples: 50000
      - name: validation
        num_bytes: 613073
        num_examples: 5000
      - name: test
        num_bytes: 10959
        num_examples: 100
    download_size: 4347985
    dataset_size: 6713409
  - config_name: summarization
    features:
      - name: x
        dtype: string
      - name: 'y'
        dtype: string
    splits:
      - name: train
        num_bytes: 98465532
        num_examples: 50000
      - name: validation
        num_bytes: 10120970
        num_examples: 5000
      - name: test
        num_bytes: 200120
        num_examples: 100
    download_size: 66841138
    dataset_size: 108786622
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: valid
        path: data/valid-*
      - split: test
        path: data/test-*
  - config_name: dialogue
    data_files:
      - split: train
        path: dialogue/train-*
      - split: validation
        path: dialogue/validation-*
      - split: test
        path: dialogue/test-*
  - config_name: summarization
    data_files:
      - split: train
        path: summarization/train-*
      - split: validation
        path: summarization/validation-*
      - split: test
        path: summarization/test-*
  - config_name: translation
    data_files:
      - split: train
        path: translation/train-*
      - split: validation
        path: translation/validation-*
      - split: test
        path: translation/test-*

Dataset Card for "GenMix50k"

Dataset Summary

Data Field

• x: Input Sequence
• y: Label Sequence

Data Split

• train: 50000
• val: 5000
• test: 100

Language

English

Source Data

• Translation: WMT-14 En-De
• Dialogue: Daily Dialogue
• Summarizaiton: CNN Daily Mail