modularStarEncoder.py

from transformers import  Starcoder2Model
import sys
from .config import ModularStarEncoderConfig
import os
from dataclasses import dataclass
from typing import Optional, Tuple, Union
import sys
import torch
import torch.utils.checkpoint
from torch import nn
from torch.nn import  CrossEntropyLoss
from transformers.activations import ACT2FN
from transformers.modeling_utils import PreTrainedModel
from transformers.utils import (
    ModelOutput,
    logging,

)

logger = logging.get_logger(__name__)

class StarEncoder2PreTrainedModel(PreTrainedModel):
    """
    An abstract class to handle weights initialization and a simple interface for downloading and loading pretrained
    models.
    """

    config_class = ModularStarEncoderConfig
    base_model_prefix = "ModularStarEncoder"
    model_type = "ModularStarEncoder"
    supports_gradient_checkpointing = True
    _supports_flash_attn_2 = True
    _supports_sdpa = True
    _supports_cache_class = True



    def _init_weights(self, module):
        """Initialize the weights"""
        if isinstance(module, nn.Linear):
            # Slightly different from the TF version which uses truncated_normal for initialization
            # cf https://github.com/pytorch/pytorch/pull/5617
            module.weight.data.normal_(mean=0.0, std=self.config.initializer_range)
            if module.bias is not None:
                module.bias.data.zero_()
        elif isinstance(module, nn.Embedding):
            module.weight.data.normal_(mean=0.0, std=self.config.initializer_range)
            if module.padding_idx is not None:
                module.weight.data[module.padding_idx].zero_()
        elif isinstance(module, nn.LayerNorm):
            module.bias.data.zero_()
            module.weight.data.fill_(1.0)

class StarEncoder2Pooler(nn.Module):
    def __init__(self, config):
        super().__init__()
        self.dense = nn.Linear(config.hidden_size, config.hidden_size)
        self.activation = nn.Tanh()

    def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
        # We "pool" the model by simply taking the hidden state corresponding
        # to the last token.
        last_token_tensor = hidden_states[:, -1]
        pooled_output = self.dense(last_token_tensor)
        pooled_output = self.activation(pooled_output)
        return pooled_output

@dataclass
class ModularStarEncoderOutput(ModelOutput):
    """
    Output type of [`BertForPreTraining`].

    Args:
        loss (*optional*, returned when `labels` is provided, `torch.FloatTensor` of shape `(1,)`):
            Total loss as the sum of the masked language modeling loss and the next sequence prediction
            (classification) loss.
        prediction_logits (`torch.FloatTensor` of shape `(batch_size, sequence_length, config.vocab_size)`):
            Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).
        seq_relationship_logits (`torch.FloatTensor` of shape `(batch_size, 2)`):
            Prediction scores of the in context classification (classification) head (scores of True/False continuation
            before SoftMax).
        hidden_states (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`):
            Tuple of `torch.FloatTensor` (one for the output of the embeddings + one for the output of each layer) of
            shape `(batch_size, sequence_length, hidden_size)`.

            Hidden-states of the model at the output of each layer plus the initial embedding outputs.
        attentions (`tuple(torch.FloatTensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`):
            Tuple of `torch.FloatTensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length,
            sequence_length)`.

            Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
            heads.
    """

    last_hidden_state: Optional[torch.FloatTensor] = None
    hidden_states: Optional[Tuple[torch.FloatTensor]] = None
    loss: Optional[torch.FloatTensor] = None
    prediction_logits: torch.FloatTensor = None
    seq_relationship_logits: torch.FloatTensor = None
    attentions: Optional[Tuple[torch.FloatTensor]] = None




class StarEncoder2PredictionHeadTransform(nn.Module):
    def __init__(self, config):
        super().__init__()
        self.is_matryoshka = config.layer_matryoshka_loss

        if self.is_matryoshka:
            self.dense = nn.Linear(config.hidden_size + config.conditional_size, config.hidden_size + config.conditional_size)
            self.LayerNorm = nn.LayerNorm(config.hidden_size + config.conditional_size, eps=config.layer_norm_eps)

        else:
            self.dense = nn.Linear(config.hidden_size, config.hidden_size)
            self.LayerNorm = nn.LayerNorm(config.hidden_size, eps=config.layer_norm_eps)

        if isinstance(config.hidden_act, str):
            self.transform_act_fn = ACT2FN[config.hidden_act]
        else:
            self.transform_act_fn = config.hidden_act

    def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
        hidden_states = self.dense(hidden_states)
        hidden_states = self.transform_act_fn(hidden_states)
        hidden_states = self.LayerNorm(hidden_states)
        return hidden_states



class StarEncoder2LMPredictionHead(nn.Module):
    def __init__(self, config):
        super().__init__()
        for element in dir(config):
            value = getattr(config, element)  # Get the attribute value
            if (isinstance(value, tuple) or isinstance(value, list)) and len(value)>0:
                setattr(config, element, value[0])
        self.transform = StarEncoder2PredictionHeadTransform(config)

        # The output weights are the same as the input embeddings, but there is
        # an output-only bias for each token.
        self.is_matryoshka = config.layer_matryoshka_loss

        if self.is_matryoshka:
            self.decoder = nn.Linear(config.hidden_size + config.conditional_size, config.vocab_size, bias=False)
        else:
            self.decoder = nn.Linear(config.hidden_size, config.vocab_size, bias=False)


        self.bias = nn.Parameter(torch.zeros(config.vocab_size))

        # Need a link between the two variables so that the bias is correctly resized with `resize_token_embeddings`
        self.decoder.bias = self.bias

    def forward(self, hidden_states):
        hidden_states = self.transform(hidden_states)
        hidden_states = self.decoder(hidden_states)
        return hidden_states

class StarEncoder2PreTrainingHeads(nn.Module):
    def __init__(self, config):
        super().__init__()
        self.predictions = StarEncoder2LMPredictionHead(config)
        self.is_matryoshka = config.layer_matryoshka_loss
        if self.is_matryoshka:
            self.seq_relationship = nn.Linear(config.hidden_size + config.conditional_size, 2)
            self.conditional_embeddings = nn.Embedding(len(config.matryoshka_layers),config.conditional_size)
        else:
            self.seq_relationship = nn.Linear(config.hidden_size, 2)



    def forward(self, sequence_output, pooled_output,idx_layer: Optional[torch.Tensor] = None):
        if self.is_matryoshka:
            device_sequence = sequence_output.get_device()
            if device_sequence<0:
                device_sequence = "cpu"
            prediction_scores = self.predictions(torch.cat([sequence_output , self.conditional_embeddings(torch.tensor(idx_layer,device=device_sequence).int()).expand(sequence_output.size()[0],sequence_output.size()[1],-1)],dim=-1))
            seq_relationship_score = self.seq_relationship(torch.cat([pooled_output , self.conditional_embeddings(torch.tensor(idx_layer,device=device_sequence).int()).expand(pooled_output.size()[0],-1)],dim=-1))
        else:
            prediction_scores = self.predictions(sequence_output)
            seq_relationship_score = self.seq_relationship(pooled_output)
        return prediction_scores, seq_relationship_score





class ModularStarEncoder(StarEncoder2PreTrainedModel):
    _tied_weights_keys = ["predictions.decoder.bias", "cls.predictions.decoder.weight"]
    config_class = ModularStarEncoderConfig
    def __init__(self, config):
        super().__init__(config)
        self.model_type = "ModularStarEncoder"
        self.cls = StarEncoder2PreTrainingHeads(config)
        self.layer_matryoshka_loss = config.layer_matryoshka_loss
        self.matryoshka_layers = config.matryoshka_layers

        if  self.layer_matryoshka_loss:
            config.sliding_window = None
            logger.warning_once(
                "The matryoshka loss is implemented without sliding_window, if you want to use the sliding window set sliding_window to True"
            )
            if self.matryoshka_layers[-1] != config.num_hidden_layers:
                logger.warning_once(
                    f"To get optimal results, the last layer on matryoshka layers, which now is {self.matryoshka_layers[-1]} "
                     "must be set as the overall number of hidden layers."
                    f"The overall number of hidden layers is now set to {config.num_hidden_layers}"
                )
                sys.exit()



        self.starEncoder2 = Starcoder2Model(config)


        self.pooler = StarEncoder2Pooler(config)

        #setting off causal masking
        for layer in self.starEncoder2.layers:
            layer.self_attn.is_causal=False



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





    def forward(
        self,
        input_ids: Optional[torch.Tensor] = None,
        attention_mask: Optional[torch.Tensor] = None,
        #token_type_ids: Optional[torch.Tensor] = None,
        position_ids: Optional[torch.Tensor] = None,
        head_mask: Optional[torch.Tensor] = None,
        inputs_embeds: Optional[torch.Tensor] = None,
        labels: Optional[torch.Tensor] = None,
        next_sentence_label: Optional[torch.Tensor] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        return_dict: Optional[bool] = None,
    ) -> Union[Tuple[torch.Tensor], ModularStarEncoderOutput]:
        r"""
            labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
                Labels for computing the masked language modeling loss. Indices should be in `[-100, 0, ...,
                config.vocab_size]` (see `input_ids` docstring) Tokens with indices set to `-100` are ignored (masked),
                the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`
            next_sentence_label (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
                This label is assigned to the in context loss:
                - 0 indicates sequence B belongs to the same repository of A,
                - 1 indicates sequence B is a random repository.
            kwargs (`Dict[str, any]`, optional, defaults to *{}*):
                Used to hide legacy arguments that have been deprecated.


        """
        return_dict = return_dict if return_dict is not None else self.config.use_return_dict

        outputs = self.starEncoder2(
                input_ids,
                attention_mask=attention_mask,
                position_ids=position_ids,
                inputs_embeds=inputs_embeds,
                output_attentions=output_attentions,
                output_hidden_states=True,
                return_dict=return_dict,
            )

        #if layer matryoshka on, compute the scores for all the heads
        if self.layer_matryoshka_loss:
          prediction_scores = []
          seq_relationship_score = []
          #for layer in outputs.hidden_states:
          for counter,idx_layer in enumerate(self.matryoshka_layers):

            #pooling head to project last hidden states as CLS token is in the last position
            pooled_output = self.pooler(outputs.hidden_states[idx_layer])
            #all the hidden states related to the last layer
            sequence_output = outputs.hidden_states[idx_layer]
            temp_prediction_scores, temp_seq_relationship_score = self.cls(sequence_output, pooled_output,counter)
            prediction_scores.append(temp_prediction_scores)
            seq_relationship_score.append(temp_seq_relationship_score)
        else:
          #pooling head to project last hidden states as CLS token is in the last position
          pooled_output = self.pooler(outputs.last_hidden_state)
          #all the hidden states related to the last layer
          sequence_output = outputs.last_hidden_state
          prediction_scores, seq_relationship_score = self.cls(sequence_output, pooled_output)

        total_loss = None
        if labels is not None and next_sentence_label is not None and not self.layer_matryoshka_loss:
            loss_fct = CrossEntropyLoss()
            masked_lm_loss = loss_fct(prediction_scores.view(-1, self.config.vocab_size), labels.view(-1))
            next_sentence_loss = loss_fct(seq_relationship_score.view(-1, 2), next_sentence_label.view(-1))
            total_loss = masked_lm_loss + next_sentence_loss

        elif labels is not None and next_sentence_label is not None and self.layer_matryoshka_loss:
            loss_fct = CrossEntropyLoss()
            num_layers = len(prediction_scores)

            #for layer in self.matryoshka_layers: seq_relationship_score
            for index in range(num_layers):
                masked_lm_loss = loss_fct(prediction_scores[index].view(-1, self.config.vocab_size), labels.view(-1))
                next_sentence_loss = loss_fct(seq_relationship_score[index].view(-1, 2), next_sentence_label.view(-1))
                if total_loss:
                    total_loss += (masked_lm_loss + next_sentence_loss) * ((index+1)/num_layers)
                else:
                    total_loss = (masked_lm_loss + next_sentence_loss) * ((index+1)/num_layers)

              


        if not return_dict:
            output = (prediction_scores, seq_relationship_score) + outputs[2:]
            return ((total_loss,) + output) if total_loss is not None else output


        last_hidden_state= outputs.hidden_states[-1]

        return ModularStarEncoderOutput(
            last_hidden_state = last_hidden_state,
            hidden_states = outputs.hidden_states,
            loss = total_loss,
            prediction_logits = prediction_scores,
            seq_relationship_logits = seq_relationship_score,
            attentions = outputs.attentions,
        )