Seq2SeqModel.Config

Component: Seq2SeqModel

class Seq2SeqModel.Config

Bases: Model.Config

All Attributes (including base classes)

inputs: ModelInput = ModelInput()

encoder_decoder: RNNModel.Config = RNNModel.Config()

source_embedding: WordEmbedding.Config = WordEmbedding.Config()

target_embedding: WordEmbedding.Config = WordEmbedding.Config()

dict_embedding: Optional[DictEmbedding.Config] = None

contextual_token_embedding: Optional[ContextualTokenEmbedding.Config] = None

output_layer: Seq2SeqOutputLayer.Config = Seq2SeqOutputLayer.Config()

sequence_generator: ScriptedSequenceGenerator.Config = ScriptedSequenceGenerator.Config()

Default JSON

{
    "inputs": {
        "src_seq_tokens": {
            "is_input": true,
            "column": "text",
            "tokenizer": {
                "Tokenizer": {
                    "split_regex": "\\s+",
                    "lowercase": true,
                    "use_byte_offsets": false
                }
            },
            "add_bos_token": false,
            "add_eos_token": false,
            "use_eos_token_for_bos": false,
            "max_seq_len": null,
            "vocab": {
                "build_from_data": true,
                "size_from_data": 0,
                "min_counts": 0,
                "vocab_files": []
            },
            "vocab_file_delimiter": " "
        },
        "trg_seq_tokens": {
            "is_input": true,
            "column": "text",
            "tokenizer": {
                "Tokenizer": {
                    "split_regex": "\\s+",
                    "lowercase": true,
                    "use_byte_offsets": false
                }
            },
            "add_bos_token": false,
            "add_eos_token": false,
            "use_eos_token_for_bos": false,
            "max_seq_len": null,
            "vocab": {
                "build_from_data": true,
                "size_from_data": 0,
                "min_counts": 0,
                "vocab_files": []
            },
            "vocab_file_delimiter": " "
        },
        "dict_feat": null,
        "contextual_token_embedding": null
    },
    "encoder_decoder": {
        "encoder": {
            "embed_dim": 512,
            "hidden_dim": 512,
            "num_layers": 1,
            "dropout_in": 0.1,
            "dropout_out": 0.1,
            "bidirectional": false
        },
        "decoder": {
            "encoder_hidden_dim": 512,
            "embed_dim": 512,
            "hidden_dim": 512,
            "out_embed_dim": 512,
            "cell_type": "lstm",
            "num_layers": 1,
            "dropout_in": 0.1,
            "dropout_out": 0.1,
            "attention_type": "dot",
            "attention_heads": 8,
            "first_layer_attention": false,
            "averaging_encoder": false
        }
    },
    "source_embedding": {
        "load_path": null,
        "save_path": null,
        "freeze": false,
        "shared_module_key": null,
        "embed_dim": 100,
        "embedding_init_strategy": "random",
        "embedding_init_range": null,
        "embeddding_init_std": 0.02,
        "export_input_names": [
            "tokens_vals"
        ],
        "pretrained_embeddings_path": "",
        "vocab_file": "",
        "vocab_size": 0,
        "vocab_from_train_data": true,
        "vocab_from_all_data": false,
        "vocab_from_pretrained_embeddings": false,
        "lowercase_tokens": true,
        "min_freq": 1,
        "mlp_layer_dims": [],
        "padding_idx": null,
        "cpu_only": false,
        "skip_header": true,
        "delimiter": " "
    },
    "target_embedding": {
        "load_path": null,
        "save_path": null,
        "freeze": false,
        "shared_module_key": null,
        "embed_dim": 100,
        "embedding_init_strategy": "random",
        "embedding_init_range": null,
        "embeddding_init_std": 0.02,
        "export_input_names": [
            "tokens_vals"
        ],
        "pretrained_embeddings_path": "",
        "vocab_file": "",
        "vocab_size": 0,
        "vocab_from_train_data": true,
        "vocab_from_all_data": false,
        "vocab_from_pretrained_embeddings": false,
        "lowercase_tokens": true,
        "min_freq": 1,
        "mlp_layer_dims": [],
        "padding_idx": null,
        "cpu_only": false,
        "skip_header": true,
        "delimiter": " "
    },
    "dict_embedding": null,
    "contextual_token_embedding": null,
    "output_layer": {
        "loss": {
            "CrossEntropyLoss": {}
        }
    },
    "sequence_generator": {
        "beam_size": 2,
        "targetlen_cap": 100,
        "targetlen_a": 0,
        "targetlen_b": 2,
        "targetlen_c": 2,
        "quantize": true,
        "length_penalty": 0.25,
        "nbest": 2,
        "stop_at_eos": true,
        "record_attention": false
    }
}