eole-model/config.json

{
  "src_vocab_size": 32000,
  "valid_metrics": [
    "BLEU"
  ],
  "tensorboard_log_dir_dated": "tensorboard/Oct-04_21-52-07",
  "n_sample": 0,
  "tgt_vocab_size": 32000,
  "transforms": [
    "sentencepiece",
    "filtertoolong"
  ],
  "overwrite": true,
  "vocab_size_multiple": 8,
  "tensorboard_log_dir": "tensorboard",
  "report_every": 100,
  "share_vocab": false,
  "src_vocab": "cs.eole.vocab",
  "save_data": "data",
  "seed": 1234,
  "tgt_vocab": "en.eole.vocab",
  "tensorboard": true,
  "training": {
    "batch_type": "tokens",
    "world_size": 1,
    "valid_batch_size": 2048,
    "label_smoothing": 0.1,
    "accum_count": [
      15
    ],
    "num_workers": 0,
    "model_path": "quickmt-cs-en-eole-model",
    "gpu_ranks": [
      0
    ],
    "valid_steps": 5000,
    "accum_steps": [
      0
    ],
    "average_decay": 0.0001,
    "batch_size": 6000,
    "optim": "adamw",
    "max_grad_norm": 0.0,
    "param_init_method": "xavier_uniform",
    "batch_size_multiple": 8,
    "bucket_size": 128000,
    "dropout": [
      0.1
    ],
    "decay_method": "noam",
    "normalization": "tokens",
    "dropout_steps": [
      0
    ],
    "learning_rate": 2.0,
    "train_steps": 100000,
    "adam_beta2": 0.998,
    "save_checkpoint_steps": 5000,
    "prefetch_factor": 32,
    "compute_dtype": "torch.float16",
    "keep_checkpoint": 4,
    "warmup_steps": 2000,
    "attention_dropout": [
      0.1
    ]
  },
  "model": {
    "heads": 8,
    "position_encoding_type": "SinusoidalInterleaved",
    "hidden_size": 1024,
    "share_embeddings": false,
    "share_decoder_embeddings": true,
    "transformer_ff": 4096,
    "architecture": "transformer",
    "embeddings": {
      "src_word_vec_size": 1024,
      "tgt_word_vec_size": 1024,
      "position_encoding_type": "SinusoidalInterleaved",
      "word_vec_size": 1024
    },
    "encoder": {
      "layers": 8,
      "heads": 8,
      "position_encoding_type": "SinusoidalInterleaved",
      "hidden_size": 1024,
      "encoder_type": "transformer",
      "src_word_vec_size": 1024,
      "transformer_ff": 4096,
      "n_positions": null
    },
    "decoder": {
      "decoder_type": "transformer",
      "layers": 2,
      "heads": 8,
      "position_encoding_type": "SinusoidalInterleaved",
      "hidden_size": 1024,
      "n_positions": null,
      "transformer_ff": 4096,
      "tgt_word_vec_size": 1024
    }
  },
  "transforms_configs": {
    "sentencepiece": {
      "tgt_subword_model": "${MODEL_PATH}/en.spm.model",
      "src_subword_model": "${MODEL_PATH}/cs.spm.model"
    },
    "filtertoolong": {
      "tgt_seq_length": 256,
      "src_seq_length": 256
    }
  },
  "data": {
    "corpus_1": {
      "path_align": null,
      "path_tgt": "train.en",
      "path_src": "train.cs",
      "transforms": [
        "sentencepiece",
        "filtertoolong"
      ]
    },
    "valid": {
      "path_align": null,
      "path_tgt": "valid.en",
      "path_src": "valid.cs",
      "transforms": [
        "sentencepiece",
        "filtertoolong"
      ]
    }
  }
}