| { | |
| "dataset_reader": { | |
| "type": "squad", | |
| "token_indexers": { | |
| "elmo": { | |
| "type": "elmo_characters" | |
| }, | |
| "token_characters": { | |
| "type": "characters", | |
| "character_tokenizer": { | |
| "byte_encoding": "utf-8", | |
| "end_tokens": [ | |
| 260 | |
| ], | |
| "start_tokens": [ | |
| 259 | |
| ] | |
| }, | |
| "min_padding_length": 5 | |
| }, | |
| "tokens": { | |
| "type": "single_id", | |
| "lowercase_tokens": true | |
| } | |
| } | |
| }, | |
| "model": { | |
| "type": "bidaf", | |
| "dropout": 0.2, | |
| "matrix_attention": { | |
| "type": "linear", | |
| "combination": "x,y,x*y", | |
| "tensor_1_dim": 200, | |
| "tensor_2_dim": 200 | |
| }, | |
| "modeling_layer": { | |
| "type": "lstm", | |
| "bidirectional": true, | |
| "dropout": 0.2, | |
| "hidden_size": 100, | |
| "input_size": 800, | |
| "num_layers": 2 | |
| }, | |
| "num_highway_layers": 2, | |
| "phrase_layer": { | |
| "type": "lstm", | |
| "bidirectional": true, | |
| "hidden_size": 100, | |
| "input_size": 1224, | |
| "num_layers": 1 | |
| }, | |
| "span_end_encoder": { | |
| "type": "lstm", | |
| "bidirectional": true, | |
| "hidden_size": 100, | |
| "input_size": 1400, | |
| "num_layers": 1 | |
| }, | |
| "text_field_embedder": { | |
| "token_embedders": { | |
| "elmo": { | |
| "type": "elmo_token_embedder", | |
| "do_layer_norm": false, | |
| "dropout": 0, | |
| "options_file": "hf://lysandre/elmo-2x4096_512_2048cnn_2xhighway/options.json", | |
| "weight_file": "hf://lysandre/elmo-2x4096_512_2048cnn_2xhighway/weights.hdf5" | |
| }, | |
| "token_characters": { | |
| "type": "character_encoding", | |
| "dropout": 0.2, | |
| "embedding": { | |
| "embedding_dim": 16, | |
| "num_embeddings": 262 | |
| }, | |
| "encoder": { | |
| "type": "cnn", | |
| "embedding_dim": 16, | |
| "ngram_filter_sizes": [ | |
| 5 | |
| ], | |
| "num_filters": 100 | |
| } | |
| }, | |
| "tokens": { | |
| "type": "embedding", | |
| "embedding_dim": 100, | |
| "pretrained_file": "https://allennlp.s3.amazonaws.com/datasets/glove/glove.6B.100d.txt.gz", | |
| "trainable": false | |
| } | |
| } | |
| } | |
| }, | |
| "train_data_path": "https://allennlp.s3.amazonaws.com/datasets/squad/squad-train-v1.1.json", | |
| "validation_data_path": "https://allennlp.s3.amazonaws.com/datasets/squad/squad-dev-v1.1.json", | |
| "trainer": { | |
| "cuda_device": 0, | |
| "grad_norm": 5, | |
| "learning_rate_scheduler": { | |
| "type": "reduce_on_plateau", | |
| "factor": 0.5, | |
| "mode": "max", | |
| "patience": 2 | |
| }, | |
| "num_epochs": 20, | |
| "optimizer": { | |
| "type": "adam", | |
| "betas": [ | |
| 0.9, | |
| 0.9 | |
| ] | |
| }, | |
| "patience": 10, | |
| "validation_metric": "+em" | |
| }, | |
| "data_loader": { | |
| "batch_sampler": { | |
| "type": "bucket", | |
| "batch_size": 40 | |
| } | |
| } | |
| } | |