model.py

# -*- coding: utf-8 -*-

# Most code for this implementation is borrowed from transformers

import math
import random

import torch
import torch.nn as nn
import torch.nn.functional as F
from transformers.utils import logging

from transformers.models.mbart.modeling_mbart import (
    MBartEncoderLayer, _expand_mask, MBartEncoder, MBartDecoder,
    MBartLearnedPositionalEmbedding, MBartDecoderLayer)
from transformers.modeling_outputs import (
    BaseModelOutput, BaseModelOutputWithPastAndCrossAttentions)
from transformers import MBartModel, MBartForConditionalGeneration

logger = logging.get_logger(__name__)


def shift_tokens_right(input_ids, pad_token_id, decoder_start_token_id):
    """
    Shift input ids one token to the right.
    """
    shifted_input_ids = input_ids.new_zeros(input_ids.shape)
    shifted_input_ids[:, 1:] = input_ids[:, :-1].clone()
    shifted_input_ids[:, 0] = decoder_start_token_id

    assert pad_token_id is not None, "self.model.config.pad_token_id has to be defined."
    # replace possible -100 values in labels by `pad_token_id`
    shifted_input_ids.masked_fill_(shifted_input_ids == -100, pad_token_id)

    return shifted_input_ids


class AdapterForMBart(nn.Module):
    def __init__(self, dim=512, bias=True):
        super().__init__()
        self.activation_fn = F.relu
        self.layer_norm = nn.LayerNorm(dim)
        self.d_proj = nn.Linear(dim, dim, bias=bias)
        self.u_proj = nn.Linear(dim, dim, bias=bias)

    def forward(self, inps):
        residual = inps
        inps = self.layer_norm(inps)
        inps = self.activation_fn(self.d_proj(inps))
        inps = self.u_proj(inps)
        outs = residual + inps

        return outs


class MBartEncoderWithAdapter(MBartEncoder):
    """
    Transformer encoder consisting of *config.encoder_layers* self attention layers. Each layer is a
    :class:`MBartEncoderLayer`.

    Args:
        config: MBartConfig
        embed_tokens (torch.nn.Embedding): output embedding
    """

    def __init__(self, config, embed_tokens=None):
        super().__init__(config)

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

        embed_dim = config.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

        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 = MBartLearnedPositionalEmbedding(
            config.max_position_embeddings,
            embed_dim,
        )
        self.layers = nn.ModuleList([MBartEncoderLayer(config) for _ in range(config.encoder_layers)])
        self.layernorm_embedding = nn.LayerNorm(embed_dim)
        self.layer_norm = nn.LayerNorm(config.d_model)

        # self.adapters = AdapterForMBart(config.d_model)
        self.adapters = nn.ModuleList([AdapterForMBart(config.d_model) for _ in range(config.encoder_layers)])

        self.init_weights()

    def forward(
        self,
        input_ids=None,
        attention_mask=None,
        head_mask=None,
        inputs_embeds=None,
        output_attentions=None,
        output_hidden_states=None,
        return_dict=None,
    ):
        r"""
        Args:
            input_ids (:obj:`torch.LongTensor` of shape :obj:`(batch_size, sequence_length)`):
                Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you
                provide it.

                Indices can be obtained using :class:`~transformers.MBartTokenizer`. See
                :meth:`transformers.PreTrainedTokenizer.encode` and :meth:`transformers.PreTrainedTokenizer.__call__`
                for details.

                `What are input IDs? <../glossary.html#input-ids>`__
            attention_mask (:obj:`torch.Tensor` of shape :obj:`(batch_size, sequence_length)`, `optional`):
                Mask to avoid performing attention on padding token indices. Mask values selected in ``[0, 1]``:

                - 1 for tokens that are **not masked**,
                - 0 for tokens that are **masked**.

                `What are attention masks? <../glossary.html#attention-mask>`__
            head_mask (:obj:`torch.Tensor` of shape :obj:`(encoder_layers, encoder_attention_heads)`, `optional`):
                Mask to nullify selected heads of the attention modules. Mask values selected in ``[0, 1]``:

                - 1 indicates the head is **not masked**,
                - 0 indicates the head is **masked**.

            inputs_embeds (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, sequence_length, hidden_size)`, `optional`):
                Optionally, instead of passing :obj:`input_ids` you can choose to directly pass an embedded
                representation. This is useful if you want more control over how to convert :obj:`input_ids` indices
                into associated vectors than the model's internal embedding lookup matrix.
            output_attentions (:obj:`bool`, `optional`):
                Whether or not to return the attentions tensors of all attention layers. See ``attentions`` under
                returned tensors for more detail.
            output_hidden_states (:obj:`bool`, `optional`):
                Whether or not to return the hidden states of all layers. See ``hidden_states`` under returned tensors
                for more detail.
            return_dict (:obj:`bool`, `optional`):
                Whether or not to return a :class:`~transformers.file_utils.ModelOutput` instead of a plain tuple.
        """
        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 = F.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:
            assert head_mask.size()[0] == (
                len(self.layers)
            ), 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):
        for idx, (encoder_layer, adapter) in enumerate(zip(self.layers, self.adapters)):
            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 getattr(self.config, "gradient_checkpointing", False) 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,
                    )
                # if idx<6:
                #    hidden_states = layer_outputs[0]
                # else:
                #    hidden_states = self.adapters[idx-6](layer_outputs[0])
                hidden_states = adapter(layer_outputs[0])

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

        hidden_states = self.layer_norm(hidden_states)

        if output_hidden_states:
            encoder_states = encoder_states + (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
        )


class MBartDecoderWithAdapter(MBartDecoder):
    """
    Transformer decoder consisting of *config.decoder_layers* layers. Each layer is a :class:`MBartDecoderLayer`

    Args:
        config: MBartConfig
        embed_tokens (torch.nn.Embedding): output embedding
    """

    def __init__(self, config, embed_tokens=None):
        super().__init__(config)

        self.dropout = config.dropout
        self.layerdrop = config.decoder_layerdrop
        self.padding_idx = config.pad_token_id
        self.max_target_positions = config.max_position_embeddings
        self.embed_scale = math.sqrt(config.d_model) if config.scale_embedding else 1.0

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

        self.embed_positions = MBartLearnedPositionalEmbedding(
            config.max_position_embeddings,
            config.d_model,
        )
        self.layers = nn.ModuleList([MBartDecoderLayer(config) for _ in range(config.decoder_layers)])
        self.layernorm_embedding = nn.LayerNorm(config.d_model)
        self.layer_norm = nn.LayerNorm(config.d_model)

        # self.adapters = AdapterForMBart(config.d_model)
        self.adapters = nn.ModuleList([AdapterForMBart(config.d_model) for _ in range(config.decoder_layers)])

        self.init_weights()

    def forward(
            self,
            input_ids=None,
            attention_mask=None,
            encoder_hidden_states=None,
            encoder_attention_mask=None,
            head_mask=None,
            cross_attn_head_mask=None,
            past_key_values=None,
            inputs_embeds=None,
            use_cache=None,
            output_attentions=None,
            output_hidden_states=None,
            return_dict=None,
    ):
        r"""
        Args:
            input_ids (:obj:`torch.LongTensor` of shape :obj:`(batch_size, sequence_length)`):
                Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you
                provide it.

                Indices can be obtained using :class:`~transformers.MBartTokenizer`. See
                :meth:`transformers.PreTrainedTokenizer.encode` and :meth:`transformers.PreTrainedTokenizer.__call__`
                for details.

                `What are input IDs? <../glossary.html#input-ids>`__
            attention_mask (:obj:`torch.Tensor` of shape :obj:`(batch_size, sequence_length)`, `optional`):
                Mask to avoid performing attention on padding token indices. Mask values selected in ``[0, 1]``:

                - 1 for tokens that are **not masked**,
                - 0 for tokens that are **masked**.

                `What are attention masks? <../glossary.html#attention-mask>`__
            encoder_hidden_states (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, encoder_sequence_length, hidden_size)`, `optional`):
                Sequence of hidden-states at the output of the last layer of the encoder. Used in the cross-attention
                of the decoder.
            encoder_attention_mask (:obj:`torch.LongTensor` of shape :obj:`(batch_size, encoder_sequence_length)`, `optional`):
                Mask to avoid performing cross-attention on padding tokens indices of encoder input_ids. Mask values
                selected in ``[0, 1]``:

                - 1 for tokens that are **not masked**,
                - 0 for tokens that are **masked**.

                `What are attention masks? <../glossary.html#attention-mask>`__
            head_mask (:obj:`torch.Tensor` of shape :obj:`(decoder_layers, decoder_attention_heads)`, `optional`):
                Mask to nullify selected heads of the attention modules. Mask values selected in ``[0, 1]``:

                - 1 indicates the head is **not masked**,
                - 0 indicates the head is **masked**.

            cross_attn_head_mask (:obj:`torch.Tensor` of shape :obj:`(decoder_layers, decoder_attention_heads)`, `optional`):
                Mask to nullify selected heads of the cross-attention modules in the decoder to avoid performing
                cross-attention on hidden heads. Mask values selected in ``[0, 1]``:

                - 1 indicates the head is **not masked**,
                - 0 indicates the head is **masked**.

            past_key_values (:obj:`Tuple[Tuple[torch.Tensor]]` of length :obj:`config.n_layers` with each tuple having 2 tuples each of which has 2 tensors of shape :obj:`(batch_size, num_heads, sequence_length - 1, embed_size_per_head)`):
                Contains precomputed key and value hidden-states of the attention blocks. Can be used to speed up
                decoding.

                If :obj:`past_key_values` are used, the user can optionally input only the last
                :obj:`decoder_input_ids` (those that don't have their past key value states given to this model) of
                shape :obj:`(batch_size, 1)` instead of all :obj:`decoder_input_ids`` of shape :obj:`(batch_size,
                sequence_length)`.
            inputs_embeds (:obj:`torch.FloatTensor` of shape :obj:`(batch_size, sequence_length, hidden_size)`, `optional`):
                Optionally, instead of passing :obj:`input_ids` you can choose to directly pass an embedded
                representation. This is useful if you want more control over how to convert :obj:`input_ids` indices
                into associated vectors than the model's internal embedding lookup matrix.
            output_attentions (:obj:`bool`, `optional`):
                Whether or not to return the attentions tensors of all attention layers. See ``attentions`` under
                returned tensors for more detail.
            output_hidden_states (:obj:`bool`, `optional`):
                Whether or not to return the hidden states of all layers. See ``hidden_states`` under returned tensors
                for more detail.
            return_dict (:obj:`bool`, `optional`):
                Whether or not to return a :class:`~transformers.file_utils.ModelOutput` instead of a plain tuple.
        """
        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
        )
        use_cache = use_cache if use_cache is not None else self.config.use_cache
        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 decoder_input_ids and decoder_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 decoder_input_ids or decoder_inputs_embeds")

        # past_key_values_length
        past_key_values_length = past_key_values[0][0].shape[2] if past_key_values is not None else 0

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

        attention_mask = self._prepare_decoder_attention_mask(
            attention_mask, input_shape, inputs_embeds, past_key_values_length
        )

        # expand encoder attention mask
        if encoder_hidden_states is not None and encoder_attention_mask is not None:
            # [bsz, seq_len] -> [bsz, 1, tgt_seq_len, src_seq_len]
            encoder_attention_mask = _expand_mask(encoder_attention_mask, inputs_embeds.dtype, tgt_len=input_shape[-1])

        # embed positions
        positions = self.embed_positions(input_shape, past_key_values_length)

        hidden_states = inputs_embeds + positions
        hidden_states = self.layernorm_embedding(hidden_states)

        hidden_states = F.dropout(hidden_states, p=self.dropout, training=self.training)

        # decoder layers
        all_hidden_states = () if output_hidden_states else None
        all_self_attns = () if output_attentions else None
        all_cross_attentions = () if (output_attentions and encoder_hidden_states is not None) else None
        next_decoder_cache = () if use_cache else None

        # check if head_mask/cross_attn_head_mask has a correct number of layers specified if desired
        for attn_mask, mask_name in zip([head_mask, cross_attn_head_mask], ["head_mask", "cross_attn_head_mask"]):
            if attn_mask is not None:
                assert attn_mask.size()[0] == (
                    len(self.layers)
                ), f"The `{mask_name}` should be specified for {len(self.layers)} layers, but it is for {head_mask.size()[0]}."
        # for idx, decoder_layer in enumerate(self.layers):
        for idx, (decoder_layer, adapter) in enumerate(zip(self.layers, self.adapters)):
            # add LayerDrop (see https://arxiv.org/abs/1909.11556 for description)
            if output_hidden_states:
                all_hidden_states += (hidden_states,)
            dropout_probability = random.uniform(0, 1)
            if self.training and (dropout_probability < self.layerdrop):
                continue

            past_key_value = past_key_values[idx] if past_key_values is not None else None

            if getattr(self.config, "gradient_checkpointing", False) and self.training:

                if use_cache:
                    logger.warning(
                        "`use_cache=True` is incompatible with `config.gradient_checkpointing=True`. Setting "
                        "`use_cache=False`..."
                    )
                    use_cache = False

                def create_custom_forward(module):
                    def custom_forward(*inputs):
                        # None for past_key_value
                        return module(*inputs, output_attentions, use_cache)

                    return custom_forward

                layer_outputs = torch.utils.checkpoint.checkpoint(
                    create_custom_forward(decoder_layer),
                    hidden_states,
                    attention_mask,
                    encoder_hidden_states,
                    encoder_attention_mask,
                    head_mask[idx] if head_mask is not None else None,
                    cross_attn_head_mask[idx] if cross_attn_head_mask is not None else None,
                    None,
                )
            else:

                layer_outputs = decoder_layer(
                    hidden_states,
                    attention_mask=attention_mask,
                    encoder_hidden_states=encoder_hidden_states,
                    encoder_attention_mask=encoder_attention_mask,
                    layer_head_mask=(head_mask[idx] if head_mask is not None else None),
                    cross_attn_layer_head_mask=(
                        cross_attn_head_mask[idx] if cross_attn_head_mask is not None else None
                    ),
                    past_key_value=past_key_value,
                    output_attentions=output_attentions,
                    use_cache=use_cache,
                )
            # if idx<6:
            #    hidden_states = layer_outputs[0]
            # else:
            #    hidden_states = self.adapters[idx-6](layer_outputs[0])
            hidden_states = adapter(layer_outputs[0])

            if use_cache:
                next_decoder_cache += (layer_outputs[3 if output_attentions else 1],)

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

                if encoder_hidden_states is not None:
                    all_cross_attentions += (layer_outputs[2],)

        hidden_states = self.layer_norm(hidden_states)

        # add hidden states from the last decoder layer
        if output_hidden_states:
            all_hidden_states += (hidden_states,)

        next_cache = next_decoder_cache if use_cache else None
        if not return_dict:
            return tuple(
                v
                for v in [hidden_states, next_cache, all_hidden_states, all_self_attns, all_cross_attentions]
                if v is not None
            )
        return BaseModelOutputWithPastAndCrossAttentions(
            last_hidden_state=hidden_states,
            past_key_values=next_cache,
            hidden_states=all_hidden_states,
            attentions=all_self_attns,
            cross_attentions=all_cross_attentions,
        )


class MBartModelWithAdapter(MBartModel):
    def __init__(self, config):
        super().__init__(config)

        padding_idx, vocab_size = config.pad_token_id, config.vocab_size
        self.shared = nn.Embedding(vocab_size, config.d_model, padding_idx)

        self.encoder = MBartEncoderWithAdapter(config, self.shared)
        self.decoder = MBartDecoderWithAdapter(config, self.shared)

        self.init_weights()


class MBartWithAdapterForMTST(MBartForConditionalGeneration):
    base_model_prefix = "model"
    _keys_to_ignore_on_load_missing = [
        r"final_logits_bias",
        r"encoder\.version",
        r"decoder\.version",
        r"lm_head\.weight",
    ]

    def __init__(self, config):
        super().__init__(config)
        self.config = config
        self.model = MBartModelWithAdapter(config)

        self.register_buffer("final_logits_bias", torch.zeros((1, self.model.shared.num_embeddings)))
        self.lm_head = nn.Linear(config.d_model, self.model.shared.num_embeddings, bias=False)

        self.init_weights()