| { |
| "dataset_reader": { |
| "type": "text_classification_json_utf8", |
| "token_indexers": { |
| "tokens": { |
| "type": "single_id", |
| "namespace": "tokens", |
| "lowercase_tokens": true, |
| "token_min_padding_length": 5 |
| } |
| }, |
| "tokenizer": { |
| "type": "pretrained_transformer", |
| "model_name": "/data/tianxing/PycharmProjects/AllenNLP/pretrained_models/huggingface/google-bert/bert-base-multilingual-uncased" |
| }, |
| "max_sequence_length": 256 |
| }, |
| "train_data_path": "train.json", |
| "validation_data_path": "valid.json", |
| "vocabulary": { |
| "directory_path": "vocabulary" |
| }, |
| "model": { |
| "type": "basic_classifier", |
| "text_field_embedder": { |
| "token_embedders": { |
| "tokens": { |
| "type": "embedding", |
| "num_embeddings": 105880, |
| "embedding_dim": 64 |
| } |
| } |
| }, |
| "seq2seq_encoder": { |
| "type": "pass_through", |
| "input_dim": 64 |
| }, |
| "seq2vec_encoder": { |
| "type": "bag_of_embeddings", |
| "embedding_dim": 64 |
| } |
| }, |
| "data_loader": { |
| "type": "multiprocess", |
| "batch_size": 64, |
| "shuffle": true |
| }, |
| "trainer": { |
| "type": "gradient_descent", |
| "cuda_device": -1, |
| "optimizer": { |
| "type": "bert_adam", |
| "lr": 5e-05, |
| "warmup": 0.1, |
| "t_total": 50000, |
| "schedule": "warmup_linear" |
| }, |
| "checkpointer": { |
| "serialization_dir": "serialization_dir", |
| "keep_most_recent_by_count": 10 |
| }, |
| "patience": 5, |
| "validation_metric": "+accuracy", |
| "num_epochs": 100 |
| } |
| } |
