RoBERTaR3F.ConfigΒΆ
Component: RoBERTaR3F
-
class
RoBERTaR3F.
Config
[source] Bases:
RoBERTa.Config
All Attributes (including base classes)
- inputs: InputConfig = InputConfig()
- encoder: RoBERTaEncoderBase.Config = RoBERTaEncoderJit.Config()
- decoder: MLPDecoder.Config = MLPDecoder.Config()
- output_layer: ClassificationOutputLayer.Config = ClassificationOutputLayer.Config()
- r3f_options: R3FConfigOptions = R3FConfigOptions()
Default JSON
{
"inputs": {
"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
}
},
"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
}
}
},
"encoder": {
"RoBERTaEncoderJit": {
"load_path": null,
"save_path": null,
"freeze": false,
"shared_module_key": null,
"output_dropout": 0.4,
"embedding_dim": 768,
"pooling": "cls_token",
"export": false,
"projection_dim": 0,
"normalize_output_rep": false,
"pretrained_encoder": {
"load_path": "public",
"save_path": null,
"freeze": false,
"shared_module_key": null
}
}
},
"decoder": {
"load_path": null,
"save_path": null,
"freeze": false,
"shared_module_key": null,
"hidden_dims": [],
"out_dim": null,
"layer_norm": false,
"dropout": 0.0,
"bias": true,
"activation": "relu",
"temperature": 1.0,
"spectral_normalization": false
},
"output_layer": {
"load_path": null,
"save_path": null,
"freeze": false,
"shared_module_key": null,
"loss": {
"CrossEntropyLoss": {}
},
"label_weights": null
},
"r3f_options": {
"r3f_lambda_by_loss": {},
"r3f_default_lambda": 0.5,
"eps": 1e-05,
"noise_type": "uniform"
}
}