Struct rust_bert::marian::MarianForConditionalGeneration

pub struct MarianForConditionalGeneration { /* fields omitted */ }

Marian Model for conditional generation

Marian model with a vocabulary decoding head It is made of the following blocks:

• base_model: BartModel Base BART model
• linear: Linear layer with bias tied to the weights of the token id embeddings

Implementations

impl MarianForConditionalGeneration

pub fn new(
    p: &Path,
    config: &BartConfig,
    generation_mode: bool
) -> MarianForConditionalGeneration

Build a new MarianForConditionalGeneration

Arguments

• p - Variable store path for the root of the BART model
• config - BartConfig object defining the model architecture
• generation_mode - flag indicating if the model should run in generation mode (a decoder start token must then be provided)

Example

use tch::{nn, Device};
use rust_bert::Config;
use std::path::Path;
use rust_bert::bart::{BartConfig, BartForConditionalGeneration};

let config_path = Path::new("path/to/config.json");
let device = Device::Cpu;
let p = nn::VarStore::new(device);
let config = BartConfig::from_file(config_path);
let generation_mode = true;
let bart: BartForConditionalGeneration = BartForConditionalGeneration::new(&(&p.root() / "bart"), &config, generation_mode);

pub fn forward_t(
    &mut self,
    input_ids: Option<&Tensor>,
    attention_mask: Option<&Tensor>,
    encoder_outputs: Option<(Tensor, Option<Vec<Tensor>>, Option<Vec<Tensor>>)>,
    decoder_input_ids: Option<&Tensor>,
    decoder_attention_mask: Option<&Tensor>,
    train: bool
) -> (Tensor, Tensor, Option<Vec<Tensor>>, Option<Vec<Tensor>>, Option<Vec<Tensor>>, Option<Vec<Tensor>>)

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.
• 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_input_ids - Optional input tensor of shape (batch size, target_sequence_length). Must be provided when running in generation mode (e.g. initialiazed with a BOS token)
• 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

• lm_logits - Tensor of shape (batch size, target_sequence_length, vocab_size) representing the logits for each vocab item and position
• encoder_hidden_states - Tensor of shape (batch size, source_sequence_length, hidden_size) representing the activations of the last encoder hidden state
• 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::bart::{BartConfig, BartForConditionalGeneration};
 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 (decoder_output, encoder_hidden_states,
      all_encoder_hidden_states, all_encoder_attentions,
      all_decoder_hidden_states, all_decoder_attentions) = no_grad(|| {
   bart_model
        .forward_t(Some(&input_tensor),
                   Some(&encoder_attention_mask),
                   None,
                   Some(&target_tensor),
                   Some(&decoder_attention_mask),
                   false)
   });

pub fn encode(
    &mut self,
    input_ids: &Tensor,
    attention_mask: Option<&Tensor>
) -> Tensor

pub fn reset_cache(&mut self)

Resets the decoder cached keys and values. Should be run for every new generation using the model.

Trait Implementations

impl LMHeadModel for MarianForConditionalGeneration

fn forward_t(
    &mut self,
    input_ids: &Option<Tensor>,
    _layer_past: &Option<Vec<Tensor>>,
    attention_mask: &Option<Tensor>,
    _token_type_ids: &Option<Tensor>,
    _position_ids: &Option<Tensor>,
    _input_embeds: &Option<Tensor>,
    encoder_outputs: Option<&Tensor>,
    decoder_input_ids: &Option<Tensor>,
    train: bool
) -> Result<(Tensor, Option<Tensor>, Option<Vec<Tensor>>, Option<Vec<Tensor>>, Option<Vec<Tensor>>), &'static str>

Forward pass through the model

Arguments

• input_ids - Optional input tensor of shape (batch size, sequence_length). If None, pre-computed embeddings must be provided (see input_embeds)
• layer_past - Unused for BART
• attention_mask - Optional mask of shape (batch size, sequence_length). Masked position have value 0, non-masked value 1. If None set to 1
• input_embeds - Unused for BART
• token_type_ids - Unused for BART
• position_ids - Unused for BART
• 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_input_ids - Optional input tensor of shape (batch size, target_sequence_length). Must be provided when running in generation mode (e.g. initialiazed with a BOS token)
• train - boolean flag to turn on/off the dropout layers in the model. Should be set to false for inference.

Returns

• lm_logits - Tensor of shape (batch size, sequence_length, vocab_size) representing the logits for each vocab item and position
• past - None
• encoder_hidden_states - Option<Tensor> Hidden states for the encoder
• hidden_states - None
• attentions - None

Example

use rust_bert::gpt2::{Gpt2Config, GPT2LMHeadModel};
use rust_bert::pipelines::generation::LMHeadModel;
 let (batch_size, sequence_length, past_sequence_length) = (64, 128, 56);
 let input_tensor = Tensor::rand(&[batch_size, sequence_length], (Int64, device));
 let mut past: Vec<Tensor> = Vec::with_capacity(config.n_layer as usize);
 for _ in 0..config.n_layer as usize {
   past.push(Tensor::rand(&[2, batch_size, config.n_head, past_sequence_length, config.n_embd / config.n_head], (Double, device)))
}
 let attention_mask = Tensor::zeros(&[batch_size, sequence_length], (Int64, device));
 let token_type_ids = Tensor::ones(&[batch_size, sequence_length], (Int64, device));
 let position_ids = Tensor::arange(sequence_length, (Int64, device)).expand(&[batch_size, sequence_length], true);

 let (output, encoder_hidden_states, _, hidden_states, attentions) = no_grad(|| {
   gpt2_model
        .forward_t(&Some(input_tensor),
                   &Some(past),
                   &Some(attention_mask),
                   &Some(token_type_ids),
                   &Some(position_ids),
                   &None,
                   None,
                   &None,
                   false).unwrap()
   });

impl LanguageGenerator<MarianForConditionalGeneration, MarianVocab, MarianTokenizer> for MarianGenerator

fn generate(
    &mut self,
    prompt_texts: Option<Vec<&str>>,
    attention_mask: Option<Tensor>
) -> Vec<String>

Generate text based on a vector of promp texts.