LSTM Language Model

LSTM Language Model

class openspeech.models.lstm_lm.model.LSTMLanguageModel(configs: omegaconf.dictconfig.DictConfig, tokenizer: openspeech.tokenizers.tokenizer.Tokenizer)[source]

LSTM language model. Paper: http://www-i6.informatik.rwth-aachen.de/publications/download/820/Sundermeyer-2012.pdf

Parameters

  • configs (DictConfig) – configuration set.

  • tokenizer (Tokenizer) – tokenizer is in charge of preparing the inputs for a model.

Inputs:
  • inputs (torch.FloatTensor): A input sequence passed to encoders. Typically for inputs this will be

    a padded FloatTensor of size (batch, seq_length, dimension).

Returns

Result of model predictions.

Return type

outputs (dict)

LSTM Language Model Configuration

class openspeech.models.lstm_lm.configurations.LSTMLanguageModelConfigs(model_name: str = 'lstm_lm', num_layers: int = 3, hidden_state_dim: int = 512, dropout_p: float = 0.3, rnn_type: str = 'lstm', max_length: int = 128, teacher_forcing_ratio: float = 1.0, optimizer: str = 'adam')[source]

This is the configuration class to store the configuration of a LSTMLanguageModel.

It is used to initiated an LSTMLanguageModel model.

Configuration objects inherit from :class: ~openspeech.dataclass.configs.OpenspeechDataclass.

Parameters

  • model_name (str) – Model name (default: lstm_lm)

  • num_layers (int) – The number of lstm layers. (default: 3)

  • hidden_state_dim (int) – The hidden state dimension of model. (default: 512)

  • dropout_p (float) – The dropout probability of encoder. (default: 0.3)

  • rnn_type (str) – Type of rnn cell (rnn, lstm, gru) (default: lstm)

  • max_length (int) – Max decoding length. (default: 128)

  • teacher_forcing_ratio (float) – The ratio of teacher forcing. (default: 1.0)

  • optimizer (str) – Optimizer for training. (default: adam)