quarel_tagger.json

{
  "dataset_reader": {
    "type": "quarel",
    "lazy": false,
    "gold_world_extractions": true,
    "sample": -1,
    "entity_types": ["world"],
    "lf_syntax": "quarel_v1",
    "tagger_only": true,
    "question_token_indexers": {
      "tokens": {
        "type": "single_id",
        "lowercase_tokens": true
      }
    }
  },
  "vocabulary": {
    "non_padded_namespaces": ["actions"],
    "pretrained_files": {"tokens": "/inputs/glove/glove.6B.100d.txt.gz"},
    "min_pretrained_embeddings": {"tokens": 30000}
  },
  "train_data_path": "https://s3-us-west-2.amazonaws.com/allennlp/datasets/quarel/quarel-v1-train.jsonl",
  "validation_data_path": "https://s3-us-west-2.amazonaws.com/allennlp/datasets/quarel/quarel-v1-dev.jsonl",
  "model": {
    "type": "crf_tagger",
    "label_encoding": "BIO",
    "dropout": 0.2,
    "include_start_end_transitions": false,
    "text_field_embedder": {
      "tokens": {
        "type": "embedding",
        "pretrained_file": "/inputs/glove/glove.6B.100d.txt.gz",
        "embedding_dim": 100,
        "trainable": false
      }
    },
    "encoder": {
      "type": "lstm",
      "input_size": 100,
      "hidden_size": 50,
      "bidirectional": true,
      "num_layers": 1
    }
  },
  "iterator": {
    "type": "basic",
    "batch_size": 64
  },
  "trainer": {
    "optimizer": {
      "type": "adam"
    },
    "validation_metric": "+f1-measure-overall",
    "num_serialized_models_to_keep": 3,
    "num_epochs": 150,
    "grad_norm": 5.0,
    "patience": 25,
    "cuda_device": -1
  }
}