RegressionModelInput

class pytext.models.roberta.RegressionModelInput

Bases: ConfigBase

All Attributes (including base classes)

tokens: RoBERTaTensorizer.Config = RoBERTaTensorizer.Config()

labels: NumericLabelTensorizer.Config = NumericLabelTensorizer.Config()

Default JSON

{
    "tokens": {
        "RoBERTaTensorizer": {
            "is_input": true,
            "columns": [
                "text"
            ],
            "tokenizer": {
                "GPT2BPETokenizer": {
                    "bpe_encoder_path": "manifold://pytext_training/tree/static/vocabs/bpe/gpt2/encoder.json",
                    "bpe_vocab_path": "manifold://pytext_training/tree/static/vocabs/bpe/gpt2/vocab.bpe",
                    "lowercase": false
                }
            },
            "base_tokenizer": null,
            "vocab_file": "gpt2_bpe_dict",
            "max_seq_len": 256,
            "add_selfie_token": false
        }
    },
    "labels": {
        "is_input": false,
        "column": "label",
        "rescale_range": null
    }
}