dataset_abstracts / README.md
davanstrien's picture
davanstrien HF staff
Upload README.md with huggingface_hub
1864268
|
raw
history blame
2.43 kB
metadata
language:
  - en
size_categories:
  - n<1K
task_categories:
  - text-classification
dataset_info:
  - config_name: annotated
    features:
      - name: text
        dtype: string
      - name: inputs
        struct:
          - name: abstract
            dtype: string
          - name: title
            dtype: string
          - name: url
            dtype: string
      - name: prediction
        dtype: 'null'
      - name: prediction_agent
        dtype: 'null'
      - name: annotation
        dtype: string
      - name: annotation_agent
        dtype: string
      - name: vectors
        dtype: 'null'
      - name: multi_label
        dtype: bool
      - name: explanation
        dtype: 'null'
      - name: id
        dtype: string
      - name: metadata
        dtype: 'null'
      - name: status
        dtype: string
      - name: event_timestamp
        dtype: string
      - name: metrics
        struct:
          - name: text_length
            dtype: int64
      - name: label
        dtype:
          class_label:
            names:
              '0': new_dataset
              '1': no_new_dataset
    splits:
      - name: train
        num_bytes: 1066485.0276679841
        num_examples: 379
      - name: test
        num_bytes: 357370.9723320158
        num_examples: 127
    download_size: 841229
    dataset_size: 1423856
  - config_name: unlabelled
    features:
      - name: text
        dtype: string
      - name: inputs
        struct:
          - name: abstract
            dtype: string
          - name: title
            dtype: string
          - name: url
            dtype: string
      - name: prediction
        dtype: 'null'
      - name: prediction_agent
        dtype: 'null'
      - name: annotation
        dtype: string
      - name: annotation_agent
        dtype: string
      - name: vectors
        dtype: 'null'
      - name: multi_label
        dtype: bool
      - name: explanation
        dtype: 'null'
      - name: id
        dtype: string
      - name: metadata
        dtype: 'null'
      - name: status
        dtype: string
      - name: event_timestamp
        dtype: timestamp[us]
      - name: metrics
        struct:
          - name: text_length
            dtype: int64
      - name: label
        dtype: string
    splits:
      - name: train
        num_bytes: 1372259.876
        num_examples: 494
    download_size: 792644
    dataset_size: 1372259.876
configs:
  - config_name: annotated
    data_files:
      - split: train
        path: annotated/train-*
      - split: test
        path: annotated/test-*
  - config_name: unlabelled
    data_files:
      - split: train
        path: unlabelled/train-*
tags:
  - 'arxiv '

Dataset Card for "dataset_abstracts"

More Information needed