architectures.py

import torch
from torch import nn
from typing import Optional, Union, Tuple
import random

from transformers.models.bart.modeling_bart import (
    BartPretrainedModel,
    BartLearnedPositionalEmbedding,
    BartAttention,
    _expand_mask
)
from transformers.modeling_outputs import (
    BaseModelOutput,
)
from transformers.activations import ACT2FN
from transformers import BartConfig, BartForConditionalGeneration

"""
The Autoencoder Network
"""
class Autoencoder(nn.Module):
    def __init__(self, bart_encoder_emb_size, first_proj, second_proj, third_proj, max_len):
        super(Autoencoder, self).__init__()
        self.encoder = nn.Sequential(
            nn.Linear(bart_encoder_emb_size, first_proj, bias=False),
            nn.BatchNorm1d(max_len),
            nn.ReLU(True),
            nn.Linear(first_proj, second_proj, bias=False),
            nn.BatchNorm1d(max_len),
            nn.ReLU(True),
            nn.Linear(second_proj, third_proj, bias=False),
            nn.BatchNorm1d(max_len),
            nn.Tanh())
        self.decoder = nn.Sequential(
            nn.Linear(third_proj, second_proj, bias=False),
            nn.BatchNorm1d(max_len),
            nn.ReLU(True),
            nn.Linear(second_proj, first_proj, bias=False),
            nn.BatchNorm1d(max_len),
            nn.ReLU(True),
            nn.Linear(first_proj, bart_encoder_emb_size, bias=False),
            nn.BatchNorm1d(max_len))

    def forward(self, x):
        x = self.encoder(x)
        x = self.decoder(x)
        return x

"""
The code adopted from the Huggingface library.
! Changing the parameter that handles encoder embedding size. (d_model)
"""
class BartEncoderLayer(nn.Module):
    def __init__(self, config: BartConfig):
        super().__init__()
        self.embed_dim = config.enc_d_model
        self.self_attn = BartAttention(
            embed_dim=self.embed_dim,
            num_heads=config.encoder_attention_heads,
            dropout=config.attention_dropout,
        )
        self.self_attn_layer_norm = nn.LayerNorm(self.embed_dim)
        self.dropout = config.dropout
        self.activation_fn = ACT2FN[config.activation_function]
        self.activation_dropout = config.activation_dropout
        self.fc1 = nn.Linear(self.embed_dim, config.encoder_ffn_dim)
        self.fc2 = nn.Linear(config.encoder_ffn_dim, self.embed_dim)
        self.final_layer_norm = nn.LayerNorm(self.embed_dim)

    def forward(
        self,
        hidden_states: torch.FloatTensor,
        attention_mask: torch.FloatTensor,
        layer_head_mask: torch.FloatTensor,
        output_attentions: Optional[bool] = False,
    ) -> Tuple[torch.FloatTensor, Optional[torch.FloatTensor]]:

        residual = hidden_states
        hidden_states, attn_weights, _ = self.self_attn(
            hidden_states=hidden_states,
            attention_mask=attention_mask,
            layer_head_mask=layer_head_mask,
            output_attentions=output_attentions,
        )
        hidden_states = nn.functional.dropout(hidden_states, p=self.dropout, training=self.training)
        hidden_states = residual + hidden_states
        hidden_states = self.self_attn_layer_norm(hidden_states)

        residual = hidden_states
        hidden_states = self.activation_fn(self.fc1(hidden_states))
        hidden_states = nn.functional.dropout(hidden_states, p=self.activation_dropout, training=self.training)
        hidden_states = self.fc2(hidden_states)
        hidden_states = nn.functional.dropout(hidden_states, p=self.dropout, training=self.training)
        hidden_states = residual + hidden_states
        hidden_states = self.final_layer_norm(hidden_states)

        if hidden_states.dtype == torch.float16 and (
            torch.isinf(hidden_states).any() or torch.isnan(hidden_states).any()
        ):
            clamp_value = torch.finfo(hidden_states.dtype).max - 1000
            hidden_states = torch.clamp(hidden_states, min=-clamp_value, max=clamp_value)

        outputs = (hidden_states,)

        if output_attentions:
            outputs += (attn_weights,)

        return outputs
    
"""
Adding the Autoencoder at the end of the BART's encoder to compress its output.
The code adopted from the Huggingface library.
"""
class BartAeEncoder(BartPretrainedModel):

    def __init__(self, config: BartConfig, ae, embed_tokens: Optional[nn.Embedding] = None):
        super().__init__(config)

        self.dropout = config.dropout
        self.layerdrop = config.encoder_layerdrop

        embed_dim = config.enc_d_model
        self.padding_idx = config.pad_token_id
        self.max_source_positions = config.max_position_embeddings
        self.embed_scale = math.sqrt(embed_dim) if config.scale_embedding else 1.0
        
        self.ae = ae

        if embed_tokens is not None:
            self.embed_tokens = embed_tokens
        else:
            self.embed_tokens = nn.Embedding(config.vocab_size, embed_dim, self.padding_idx)

        self.embed_positions = BartLearnedPositionalEmbedding(
            config.max_position_embeddings,
            embed_dim,
        )
        self.layers = nn.ModuleList([BartEncoderLayer(config) for _ in range(config.encoder_layers)])
        self.layernorm_embedding = nn.LayerNorm(embed_dim)

        self.gradient_checkpointing = False
        # Initialize weights and apply final processing
        self.post_init()

    def get_input_embeddings(self):
        return self.embed_tokens

    def set_input_embeddings(self, value):
        self.embed_tokens = value

    def forward(
        self,
        input_ids: torch.LongTensor = None,
        attention_mask: Optional[torch.Tensor] = None,
        head_mask: Optional[torch.Tensor] = None,
        inputs_embeds: Optional[torch.FloatTensor] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        return_dict: Optional[bool] = None,
    ) -> Union[Tuple, BaseModelOutput]:

        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
        output_hidden_states = (
            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
        )
        return_dict = return_dict if return_dict is not None else self.config.use_return_dict

        # retrieve input_ids and inputs_embeds
        if input_ids is not None and inputs_embeds is not None:
            raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
        elif input_ids is not None:
            input_shape = input_ids.size()
            input_ids = input_ids.view(-1, input_shape[-1])
        elif inputs_embeds is not None:
            input_shape = inputs_embeds.size()[:-1]
        else:
            raise ValueError("You have to specify either input_ids or inputs_embeds")

        if inputs_embeds is None:
            inputs_embeds = self.embed_tokens(input_ids) * self.embed_scale

        embed_pos = self.embed_positions(input_shape)

        hidden_states = inputs_embeds + embed_pos
        hidden_states = self.layernorm_embedding(hidden_states)
        hidden_states = nn.functional.dropout(hidden_states, p=self.dropout, training=self.training)

        # expand attention_mask
        if attention_mask is not None:
            # [bsz, seq_len] -> [bsz, 1, tgt_seq_len, src_seq_len]
            attention_mask = _expand_mask(attention_mask, inputs_embeds.dtype)

        encoder_states = () if output_hidden_states else None
        all_attentions = () if output_attentions else None

        # check if head_mask has a correct number of layers specified if desired
        if head_mask is not None:
            if head_mask.size()[0] != (len(self.layers)):
                raise ValueError(
                    f"The head_mask should be specified for {len(self.layers)} layers, but it is for {head_mask.size()[0]}."
                )

        for idx, encoder_layer in enumerate(self.layers):
            if output_hidden_states:
                encoder_states = encoder_states + (hidden_states,)
            # add LayerDrop (see https://arxiv.org/abs/1909.11556 for description)
            dropout_probability = random.uniform(0, 1)
            if self.training and (dropout_probability < self.layerdrop):  # skip the layer
                layer_outputs = (None, None)
            else:
                if self.gradient_checkpointing and self.training:

                    def create_custom_forward(module):
                        def custom_forward(*inputs):
                            return module(*inputs, output_attentions)

                        return custom_forward

                    layer_outputs = torch.utils.checkpoint.checkpoint(
                        create_custom_forward(encoder_layer),
                        hidden_states,
                        attention_mask,
                        (head_mask[idx] if head_mask is not None else None),
                    )
                else:
                    layer_outputs = encoder_layer(
                        hidden_states,
                        attention_mask,
                        layer_head_mask=(head_mask[idx] if head_mask is not None else None),
                        output_attentions=output_attentions,
                    )

                hidden_states = layer_outputs[0]

            if output_attentions:
                all_attentions = all_attentions + (layer_outputs[1],)

        if output_hidden_states:
            encoder_states = encoder_states + (hidden_states,)
        
        hidden_states = self.ae.encoder(hidden_states)
        
        if not return_dict:
            return tuple(v for v in [hidden_states, encoder_states, all_attentions] if v is not None)
        return BaseModelOutput(
            last_hidden_state=hidden_states, hidden_states=encoder_states, attentions=all_attentions
        )

"""
Initialize the compressedBART model.
"""
class CBartForConditionalGeneration(BartForConditionalGeneration):
    def __init__(self, config, ae):
        
        super().__init__(config)
        
        # Convert the Encoder
        self.model.encoder = BartAeEncoder(config, ae)