{
  "PubMedAbstractsNER": {
    "description": "This dataset contains annotated biomedical texts with named entities from different categories. The dataset includes manually tokenized texts and a list of recognized named entities with their types and descriptions.",
    "citation": "",
    "homepage": "",
    "license": "",
    "features": {
      "tokenized_text": {
        "dtype": "string",
        "_type": "Sequence"
      },
      "ner": {
        "feature": {
          "start": {
            "dtype": "int32",
            "_type": "Value"
          },
          "end": {
            "dtype": "int32",
            "_type": "Value"
          },
          "label": {
            "dtype": "string",
            "_type": "Value"
          }
        },
        "_type": "Sequence"
      }
    },
    "splits": {
      "train": {
        "name": "train",
        "num_bytes": 151668527,
        "num_examples": 35000,
        "dataset_name": "PubMedAnstractsNER"
      }
    }
  }
}