EncoderModelInput

class pytext.models.bert_classification_models.EncoderModelInput

Bases: ModelInput

All Attributes (including base classes)

tokens: Tensorizer.Config = Tensorizer.Config()

dense: Optional[FloatListTensorizer.Config] = None

labels: LabelTensorizer.Config = LabelTensorizer.Config()

num_tokens: NtokensTensorizer.Config = NtokensTensorizer.Config(names=['tokens'], indexes=[2])

Default JSON

{
    "tokens": {
        "Tensorizer": {
            "is_input": true
        }
    },
    "dense": null,
    "labels": {
        "LabelTensorizer": {
            "is_input": false,
            "column": "label",
            "allow_unknown": false,
            "pad_in_vocab": false,
            "label_vocab": null,
            "label_vocab_file": null,
            "add_labels": null
        }
    },
    "num_tokens": {
        "is_input": false,
        "names": [
            "tokens"
        ],
        "indexes": [
            2
        ]
    }
}