Struct rust_bert::m2m_100::M2M100Model

pub struct M2M100Model { /* fields omitted */ }

M2M100 Base model

Base architecture for M2M100 model. Usually complemented with a task-specific head, such as a language model head. It is made of the following blocks:

• encoder: M2M100Encoder (transformer) made of a vector of encoding layers
• decoder: M2M100Decoder (transformer) made of a vector of decoding layers with self attention and encoder cross-attention. caching is implemented for the decoder to avoid recalculating static states (encoder key/values and previously calculated decoder key/values)
• pad_token_id: padding token id

Implementations

impl M2M100Model

pub fn new<'p, P>(p: P, config: &M2M100Config) -> M2M100Model where
    P: Borrow<Path<'p>>, 

Build a new M2M100Model

Arguments

• p - Variable store path for the root of the M2M100 model
• config - M2M100Config object defining the model architecture

Example

use rust_bert::m2m_100::{M2M100Config, M2M100Model};
use rust_bert::Config;
use std::path::Path;
use tch::{nn, Device};

let config_path = Path::new("path/to/config.json");
let device = Device::Cpu;
let p = nn::VarStore::new(device);
let config = M2M100Config::from_file(config_path);
let m2m100: M2M100Model = M2M100Model::new(&p.root() / "m2m100", &config);

pub fn forward_t(
    &self,
    input_ids: Option<&Tensor>,
    attention_mask: Option<&Tensor>,
    decoder_input_ids: Option<&Tensor>,
    encoder_output: Option<&Tensor>,
    decoder_attention_mask: Option<&Tensor>,
    layer_states: Option<Vec<(Option<LayerState>, Option<LayerState>)>>,
    train: bool
) -> MBartModelOutput

Forward pass through the model

Arguments

• input_ids - Optional input tensor of shape (batch size, source_sequence_length). Must be provided when not running in generation mode
• attention_mask - Optional attention mask of shape (batch size, source_sequence_length) for the encoder positions. Positions with a mask with value 0 will be masked.
• decoder_input_ids - Optional input tensor of shape (batch size, target_sequence_length). Must be provided when running in generation mode (e.g. initialized with a BOS token)
• encoder_outputs - Optional tuple made of a tensor of shape (batch size, source_sequence_length, encoder_hidden_dim) and optional vectors of tensors of length num_encoder_layers with shape (batch size, source_sequence_length, hidden_size). These correspond to the encoder last hidden state and optional hidden states/attention weights for encoder layers. When provided, the encoder hidden state will not be recalculated. Useful for generation tasks.
• decoder_attention_mask - Optional attention mask of shape (batch size, target_sequence_length) for the decoder positions. Positions with a mask with value 0 will be masked.
• train - boolean flag to turn on/off the dropout layers in the model. Should be set to false for inference.

Returns

• M2M100ModelOutput containing:
  • decoder_output - Tensor of shape (batch size, target_sequence_length, hidden_size) representing the activations of the last decoder hidden state
  • encoder_hidden_states - Option<Tensor> of shape (batch size, source_sequence_length, hidden_size) representing the activations of the last encoder hidden state if it was not provided, otherwise None
  • cache - (Option<Tensor>, Option<Vec<&LayerState, &LayerState>>) of length n_layer containing the encoder padding mask and past keys and values for both the self attention and the encoder cross attention of each layer of the decoder.
  • all_encoder_hidden_states - Option<Vec<Tensor>> of length num_encoder_layers with shape (batch size, source_sequence_length, hidden_size)
  • all_encoder_attentions - Option<Vec<Tensor>> of length num_encoder_layers with shape (batch size, source_sequence_length, hidden_size)
  • all_decoder_hidden_states - Option<Vec<Tensor>> of length num_decoder_layers with shape (batch size, target_sequence_length, hidden_size)
  • all_decoder_attentions - Option<Vec<Tensor>> of length num_decoder_layers with shape (batch size, target_sequence_length, hidden_size)

Example

use rust_bert::m2m_100::{M2M100Config, M2M100Model};
let (batch_size, source_sequence_length, target_sequence_length) = (64, 128, 56);
let input_tensor = Tensor::rand(&[batch_size, source_sequence_length], (Int64, device));
let target_tensor = Tensor::rand(&[batch_size, target_sequence_length], (Int64, device));
let encoder_attention_mask =
    Tensor::ones(&[batch_size, source_sequence_length], (Int64, device));
let decoder_attention_mask =
    Tensor::ones(&[batch_size, source_sequence_length], (Int64, device));

let model_output = no_grad(|| {
    m2m100_model.forward_t(
        Some(&input_tensor),
        Some(&encoder_attention_mask),
        Some(&target_tensor),
        None,
        Some(&decoder_attention_mask),
        None,
        false,
    )
});