Attend the Women in ML Symposium on December 7 Register now

persian_ner

References:

fold1

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:persian_ner/fold1')
  • Description:
The dataset includes 250,015 tokens and 7,682 Persian sentences in total. It is available in 3 folds to be used in turn as training and test sets. The NER tags are in IOB format.
  • License: Creative Commons Attribution 4.0 International License
  • Version: 1.1.0
  • Splits:
Split Examples
'test' 2560
'train' 5121
  • Features:
{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 13,
            "names": [
                "O",
                "I-event",
                "I-fac",
                "I-loc",
                "I-org",
                "I-pers",
                "I-pro",
                "B-event",
                "B-fac",
                "B-loc",
                "B-org",
                "B-pers",
                "B-pro"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

fold2

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:persian_ner/fold2')
  • Description:
The dataset includes 250,015 tokens and 7,682 Persian sentences in total. It is available in 3 folds to be used in turn as training and test sets. The NER tags are in IOB format.
  • License: Creative Commons Attribution 4.0 International License
  • Version: 1.1.0
  • Splits:
Split Examples
'test' 2561
'train' 5120
  • Features:
{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 13,
            "names": [
                "O",
                "I-event",
                "I-fac",
                "I-loc",
                "I-org",
                "I-pers",
                "I-pro",
                "B-event",
                "B-fac",
                "B-loc",
                "B-org",
                "B-pers",
                "B-pro"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

fold3

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:persian_ner/fold3')
  • Description:
The dataset includes 250,015 tokens and 7,682 Persian sentences in total. It is available in 3 folds to be used in turn as training and test sets. The NER tags are in IOB format.
  • License: Creative Commons Attribution 4.0 International License
  • Version: 1.1.0
  • Splits:
Split Examples
'test' 2560
'train' 5121
  • Features:
{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 13,
            "names": [
                "O",
                "I-event",
                "I-fac",
                "I-loc",
                "I-org",
                "I-pers",
                "I-pro",
                "B-event",
                "B-fac",
                "B-loc",
                "B-org",
                "B-pers",
                "B-pro"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}