modeling_meralion.py

# coding=utf-8
# Copyright 2024 the HuggingFace Inc. team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""PyTorch MERaLiON model."""

import math
from dataclasses import dataclass
from typing import List, Optional, Tuple, Union

import torch
import torch.utils.checkpoint
from torch import nn

from transformers.activations import ACT2FN
from transformers.cache_utils import EncoderDecoderCache, StaticCache, HybridCache
from transformers.generation import GenerationMixin
from transformers.modeling_outputs import ModelOutput, BaseModelOutput
from transformers.modeling_utils import PreTrainedModel
from transformers.utils import (
    add_start_docstrings,
    add_start_docstrings_to_model_forward,
    is_flash_attn_2_available,
    is_flash_attn_greater_or_equal_2_10,
    logging,
    replace_return_docstrings,
)

from .configuration_meralion import MERaLiONConfig, MERaLiONSpeechConfig
from .modeling_text_decoder import MERaLiONTextForCausalLM


if is_flash_attn_2_available():
    from transformers.modeling_flash_attention_utils import _flash_attention_forward


logger = logging.get_logger(__name__)

_CONFIG_FOR_DOC = "MERaLiONConfig"


def sinusoids(length: int, channels: int, max_timescale: float = 10000) -> torch.Tensor:
    """Returns sinusoids for positional embedding"""
    if channels % 2 != 0:
        raise ValueError(
            f"Number of channels has to be divisible by 2 for sinusoidal positional embeddings, got {channels} channels."
        )
    log_timescale_increment = math.log(max_timescale) / (channels // 2 - 1)
    inv_timescales = torch.exp(-log_timescale_increment * torch.arange(channels // 2))
    scaled_time = torch.arange(length).view(-1, 1) * inv_timescales.view(1, -1)
    return torch.cat([scaled_time.sin(), scaled_time.cos()], dim=1)


# Copied from transformers.models.bart.modeling_bart.shift_tokens_right
def shift_tokens_right(input_ids: torch.Tensor, pad_token_id: int, decoder_start_token_id: int):
    """
    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

    if pad_token_id is None:
        raise ValueError("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


# Copied from transformers.models.llama.modeling_llama._prepare_4d_causal_attention_mask_with_cache_position
def _prepare_4d_causal_attention_mask_with_cache_position(
    attention_mask: torch.Tensor,
    sequence_length: int,
    target_length: int,
    dtype: torch.dtype,
    device: torch.device,
    min_dtype: float,
    cache_position: torch.Tensor,
    batch_size: int,
):
    """
    Creates a causal 4D mask of shape `(batch_size, 1, query_length, key_value_length)` from a 2D mask of shape
    `(batch_size, key_value_length)`, or if the input `attention_mask` is already 4D, do nothing.

    Args:
        attention_mask (`torch.Tensor`):
            A 2D attention mask of shape `(batch_size, key_value_length)` or a 4D attention mask of shape `(batch_size, 1, query_length, key_value_length)`.
        sequence_length (`int`):
            The sequence length being processed.
        target_length (`int`):
            The target length: when generating with static cache, the mask should be as long as the static cache, to account for the 0 padding, the part of the cache that is not filled yet.
        dtype (`torch.dtype`):
            The dtype to use for the 4D attention mask.
        device (`torch.device`):
            The device to plcae the 4D attention mask on.
        min_dtype (`float`):
            The minimum value representable with the dtype `dtype`.
        cache_position (`torch.Tensor`):
            Indices depicting the position of the input sequence tokens in the sequence.
        batch_size (`torch.Tensor`):
            Batch size.
    """
    if attention_mask is not None and attention_mask.dim() == 4:
        # In this case we assume that the mask comes already in inverted form and requires no inversion or slicing.
        causal_mask = attention_mask
    else:
        causal_mask = torch.full((sequence_length, target_length), fill_value=min_dtype, dtype=dtype, device=device)
        if sequence_length != 1:
            causal_mask = torch.triu(causal_mask, diagonal=1)
        causal_mask *= torch.arange(target_length, device=device) > cache_position.reshape(-1, 1)
        causal_mask = causal_mask[None, None, :, :].expand(batch_size, 1, -1, -1)
        if attention_mask is not None:
            causal_mask = causal_mask.clone()  # copy to contiguous memory for in-place edit
            mask_length = attention_mask.shape[-1]
            padding_mask = causal_mask[:, :, :, :mask_length] + attention_mask[:, None, None, :]
            padding_mask = padding_mask == 0
            causal_mask[:, :, :, :mask_length] = causal_mask[:, :, :, :mask_length].masked_fill(
                padding_mask, min_dtype
            )
    return causal_mask


class MERaLiONSpeechAttention(nn.Module):
    """Multi-headed attention from 'Attention Is All You Need' paper"""

    def __init__(
        self,
        embed_dim: int,
        num_heads: int,
        dropout: float = 0.0,
        is_decoder: bool = False,
        bias: bool = True,
        is_causal: bool = False,
        layer_idx: Optional[int] = None,
        config: Optional[MERaLiONSpeechConfig] = None,
    ):
        super().__init__()
        self.embed_dim = embed_dim
        self.num_heads = num_heads
        self.dropout = dropout
        self.head_dim = embed_dim // num_heads
        self.config = config

        if (self.head_dim * num_heads) != self.embed_dim:
            raise ValueError(
                f"embed_dim must be divisible by num_heads (got `embed_dim`: {self.embed_dim}"
                f" and `num_heads`: {num_heads})."
            )
        self.scaling = self.head_dim**-0.5
        self.is_decoder = is_decoder
        self.is_causal = is_causal

        if layer_idx is None and is_decoder:
            logger.warning_once(
                f"Instantiating a decoder {self.__class__.__name__} without passing `layer_idx` is not recommended and "
                "will to errors during the forward call, if caching is used. Please make sure to provide a `layer_idx` "
                "when creating this class."
            )
        self.layer_idx = layer_idx

        self.k_proj = nn.Linear(embed_dim, embed_dim, bias=False)
        self.v_proj = nn.Linear(embed_dim, embed_dim, bias=bias)
        self.q_proj = nn.Linear(embed_dim, embed_dim, bias=bias)
        self.out_proj = nn.Linear(embed_dim, embed_dim, bias=bias)

    # Copied from transformers.models.bart.modeling_bart.BartAttention._shape with BART->speech
    def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
        return tensor.view(bsz, seq_len, self.num_heads, self.head_dim).transpose(1, 2).contiguous()

    def forward(
        self,
        hidden_states: torch.Tensor,
        key_value_states: Optional[torch.Tensor] = None,
        past_key_value: Optional[EncoderDecoderCache] = None,
        attention_mask: Optional[torch.Tensor] = None,
        layer_head_mask: Optional[torch.Tensor] = None,
        output_attentions: bool = False,
        cache_position: Optional[torch.LongTensor] = None,
    ) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[Tuple[torch.Tensor]]]:
        """Input shape: Batch x Time x Channel"""

        # if key_value_states are provided this layer is used as a cross-attention layer
        # for the decoder
        is_cross_attention = key_value_states is not None
        bsz, tgt_len, _ = hidden_states.size()

        # get query proj
        query_states = self._shape(self.q_proj(hidden_states) * self.scaling, tgt_len, bsz)

        if past_key_value is not None:
            is_updated = past_key_value.is_updated.get(self.layer_idx)
            if is_cross_attention:
                # after the first generated id, we can subsequently re-use all key/value_states from cache
                past_key_value.is_updated[self.layer_idx] = True
                past_key_value = past_key_value.cross_attention_cache
            else:
                past_key_value = past_key_value.self_attention_cache

        # use key_value_states if cross attention
        current_states = key_value_states if key_value_states is not None else hidden_states
        if is_cross_attention and past_key_value and is_updated:
            # reuse k,v, cross_attentions
            key_states = past_key_value.key_cache[self.layer_idx]
            value_states = past_key_value.value_cache[self.layer_idx]
        else:
            key_states = self._shape(self.k_proj(current_states), -1, bsz)
            value_states = self._shape(self.v_proj(current_states), -1, bsz)
            if past_key_value is not None:
                # save all key/value_states to cache to be re-used for fast auto-regressive generation
                cache_position = cache_position if not is_cross_attention else None
                key_states, value_states = past_key_value.update(
                    key_states, value_states, self.layer_idx, {"cache_position": cache_position}
                )

        attn_weights = torch.matmul(query_states, key_states.transpose(2, 3))

        if attention_mask is not None:  # no matter the length, we just slice it
            causal_mask = attention_mask[:, :, :, : key_states.shape[-2]]
            attn_weights = attn_weights + causal_mask

        attn_weights = nn.functional.softmax(attn_weights, dim=-1)

        if layer_head_mask is not None:
            if layer_head_mask.size() != (self.num_heads,):
                raise ValueError(
                    f"Head mask for a single layer should be of size {(self.num_heads,)}, but is"
                    f" {layer_head_mask.size()}"
                )
            attn_weights = layer_head_mask.view(1, -1, 1, 1) * attn_weights

        attn_probs = nn.functional.dropout(attn_weights, p=self.dropout, training=self.training)
        attn_output = torch.matmul(attn_probs, value_states)

        if attn_output.size() != (bsz, self.num_heads, tgt_len, self.head_dim):
            raise ValueError(
                f"`attn_output` should be of size {(bsz, self.num_heads, tgt_len, self.head_dim)}, but is"
                f" {attn_output.size()}"
            )

        attn_output = attn_output.transpose(1, 2)
        # Use the `embed_dim` from the config (stored in the class) rather than `hidden_state` because `attn_output` can be
        # partitioned across GPUs when using tensor-parallelism.
        attn_output = attn_output.reshape(bsz, tgt_len, self.embed_dim)

        attn_output = self.out_proj(attn_output)

        return attn_output, attn_weights, past_key_value


class MERaLiONSpeechFlashAttention2(MERaLiONSpeechAttention):
    """
    MERaLiONSpeech flash attention module. This module inherits from `MERaLiONSpeechAttention` as the weights of the module stays
    untouched. The only required change would be on the forward pass where it needs to correctly call the public API of
    flash attention and deal with padding tokens in case the input contains any of them.
    """

    # Copied from transformers.models.llama.modeling_llama.LlamaFlashAttention2.__init__
    def __init__(self, *args, **kwargs):
        super().__init__(*args, **kwargs)

        # TODO: Should be removed once Flash Attention for RoCm is bumped to 2.1.
        # flash_attn<2.1 generates top-left aligned causal mask, while what is needed here is bottom-right alignement, that was made default for flash_attn>=2.1. This attribute is used to handle this difference. Reference: https://github.com/Dao-AILab/flash-attention/releases/tag/v2.1.0.
        # Beware that with flash_attn<2.1, using q_seqlen != k_seqlen (except for the case q_seqlen == 1) produces a wrong mask (top-left).
        self._flash_attn_uses_top_left_mask = not is_flash_attn_greater_or_equal_2_10()

    def forward(
        self,
        hidden_states: torch.Tensor,
        key_value_states: Optional[torch.Tensor] = None,
        past_key_value: Optional[EncoderDecoderCache] = None,
        attention_mask: Optional[torch.Tensor] = None,
        layer_head_mask: Optional[torch.Tensor] = None,
        output_attentions: bool = False,
        cache_position: Optional[torch.LongTensor] = None,
    ) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[Tuple[torch.Tensor]]]:
        if isinstance(past_key_value, StaticCache):
            raise ValueError(
                "The `static` cache implementation is not compatible with `attn_implementation='flash_attention_2'`. "
                "Use `attn_implementation='sdpa'` in the meantime, and open an issue at https://github.com/huggingface/transformers"
            )
        # SpeechFlashAttention2 attention does not support output_attentions
        if output_attentions:
            raise ValueError("SpeechFlashAttention2 attention does not support output_attentions")

        # if key_value_states are provided this layer is used as a cross-attention layer
        # for the decoder
        is_cross_attention = key_value_states is not None
        bsz, tgt_len, _ = hidden_states.size()

        # get query proj
        query_states = torch.reshape(self.q_proj(hidden_states), (bsz, tgt_len, self.num_heads, self.head_dim))

        if past_key_value is not None:
            is_updated = past_key_value.is_updated.get(self.layer_idx)
            if is_cross_attention:
                # after the first generated id, we can subsequently re-use all key/value_states from cache
                past_key_value.is_updated[self.layer_idx] = True
                past_key_value = past_key_value.cross_attention_cache
            else:
                past_key_value = past_key_value.self_attention_cache

        # use key_value_states if cross attention
        current_states = key_value_states if key_value_states is not None else hidden_states
        if is_cross_attention and past_key_value and is_updated:
            # reuse k,v, cross_attentions
            key_states = past_key_value.key_cache[self.layer_idx]
            value_states = past_key_value.value_cache[self.layer_idx]
        else:
            key_states = self._shape(self.k_proj(current_states), -1, bsz)
            value_states = self._shape(self.v_proj(current_states), -1, bsz)
            if past_key_value is not None:
                # save all key/value_states to cache to be re-used for fast auto-regressive generation
                cache_position = cache_position if not is_cross_attention else None
                key_states, value_states = past_key_value.update(
                    key_states, value_states, self.layer_idx, {"cache_position": cache_position}
                )

        # TODO: These transpose are quite inefficient but Flash Attention requires the layout [batch_size, sequence_length, num_heads, head_dim]
        #  We would need to refactor the KV cache to be able to avoid many of these transpose/reshape/view.
        key_states = key_states.transpose(1, 2)
        value_states = value_states.transpose(1, 2)

        causal_mask = attention_mask
        if attention_mask is not None:  # no matter the length, we just slice it
            causal_mask = attention_mask[:, :, :, : key_states.shape[-2]]

        # In PEFT, usually we cast the layer norms in float32 for training stability reasons
        # therefore the input hidden states gets silently casted in float32. Hence, we need
        # cast them back in the correct dtype just to be sure everything works as expected.
        # This might slowdown training & inference so it is recommended to not cast the LayerNorms
        # in fp32. (LlamaRMSNorm handles it correctly)

        input_dtype = query_states.dtype
        if input_dtype == torch.float32:
            if torch.is_autocast_enabled():
                target_dtype = torch.get_autocast_gpu_dtype()
            # Handle the case where the model is quantized
            elif hasattr(self.config, "_pre_quantization_dtype"):
                target_dtype = self.config._pre_quantization_dtype
            else:
                target_dtype = self.q_proj.weight.dtype

            logger.warning_once(
                f"The input hidden states seems to be silently casted in float32, this might be related to"
                f" the fact you have upcasted embedding or layer norm layers in float32. We will cast back the input in"
                f" {target_dtype}."
            )

            query_states = query_states.to(target_dtype)
            key_states = key_states.to(target_dtype)
            value_states = value_states.to(target_dtype)

        attn_output = _flash_attention_forward(
            query_states,
            key_states,
            value_states,
            causal_mask,
            tgt_len,
            dropout=self.dropout if self.training else 0.0,
            is_causal=self.is_causal,
            use_top_left_mask=self._flash_attn_uses_top_left_mask,
        )

        attn_output = attn_output.reshape(bsz, tgt_len, -1)
        attn_output = self.out_proj(attn_output)

        if not output_attentions:
            attn_weights = None

        return attn_output, attn_weights, past_key_value


class MERaLiONSpeechSdpaAttention(MERaLiONSpeechAttention):
    def forward(
        self,
        hidden_states: torch.Tensor,
        key_value_states: Optional[torch.Tensor] = None,
        past_key_value: Optional[EncoderDecoderCache] = None,
        attention_mask: Optional[torch.Tensor] = None,
        layer_head_mask: Optional[torch.Tensor] = None,
        output_attentions: bool = False,
        cache_position: Optional[torch.LongTensor] = None,
    ) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[Tuple[torch.Tensor]]]:
        """Input shape: Batch x Time x Channel"""
        if output_attentions or layer_head_mask is not None:
            # TODO: Improve this warning with e.g. `model.config._attn_implementation = "manual"` once this is implemented.
            logger.warning_once(
                "MERaLiONSpeechModel is using MERaLiONSpeechSdpaAttention, but `torch.nn.functional.scaled_dot_product_attention` does not support `output_attentions=True` or `layer_head_mask` not None. Falling back to the manual attention"
                ' implementation, but specifying the manual implementation will be required from Transformers version v5.0.0 onwards. This warning can be removed using the argument `attn_implementation="eager"` when loading the model.'
            )
            return super().forward(
                hidden_states,
                key_value_states=key_value_states,
                past_key_value=past_key_value,
                attention_mask=attention_mask,
                layer_head_mask=layer_head_mask,
                output_attentions=output_attentions,
                cache_position=cache_position,
            )

        # if key_value_states are provided this layer is used as a cross-attention layer
        # for the decoder
        is_cross_attention = key_value_states is not None
        bsz, tgt_len, _ = hidden_states.size()

        # get query proj
        query_states = self._shape(self.q_proj(hidden_states), tgt_len, bsz)

        if past_key_value is not None:
            is_updated = past_key_value.is_updated.get(self.layer_idx)
            if is_cross_attention:
                # after the first generated id, we can subsequently re-use all key/value_states from cache
                past_key_value.is_updated[self.layer_idx] = True
                past_key_value = past_key_value.cross_attention_cache
            else:
                past_key_value = past_key_value.self_attention_cache

        # use key_value_states if cross attention
        current_states = key_value_states if key_value_states is not None else hidden_states
        if is_cross_attention and past_key_value and is_updated:
            # reuse k,v, cross_attentions
            key_states = past_key_value.key_cache[self.layer_idx]
            value_states = past_key_value.value_cache[self.layer_idx]
        else:
            key_states = self._shape(self.k_proj(current_states), -1, bsz)
            value_states = self._shape(self.v_proj(current_states), -1, bsz)
            if past_key_value is not None:
                # save all key/value_states to cache to be re-used for fast auto-regressive generation
                cache_position = cache_position if not is_cross_attention else None
                key_states, value_states = past_key_value.update(
                    key_states, value_states, self.layer_idx, {"cache_position": cache_position}
                )

        causal_mask = attention_mask
        if attention_mask is not None:  # no matter the length, we just slice it
            causal_mask = attention_mask[:, :, :, : key_states.shape[-2]]

        # We dispatch to SDPA's Flash Attention or Efficient kernels via this `is_causal` if statement instead of an inline conditional assignment
        # in SDPA to support both torch.compile's dynamic shapes and full graph options. An inline conditional prevents dynamic shapes from compiling.
        # The tgt_len > 1 is necessary to match with AttentionMaskConverter.to_causal_4d that does not create a causal mask in case tgt_len == 1.
        is_causal = True if self.is_causal and causal_mask is None and tgt_len > 1 else False

        # NOTE: SDPA with memory-efficient backend is currently (torch==2.1.2) bugged when using non-contiguous inputs and a custom attn_mask,
        # but we are fine here as `_shape` do call `.contiguous()`. Reference: https://github.com/pytorch/pytorch/issues/112577
        attn_output = torch.nn.functional.scaled_dot_product_attention(
            query_states,
            key_states,
            value_states,
            attn_mask=causal_mask,
            dropout_p=self.dropout if self.training else 0.0,
            is_causal=is_causal,
        )

        if attn_output.size() != (bsz, self.num_heads, tgt_len, self.head_dim):
            raise ValueError(
                f"`attn_output` should be of size {(bsz, self.num_heads, tgt_len, self.head_dim)}, but is"
                f" {attn_output.size()}"
            )

        attn_output = attn_output.transpose(1, 2)

        # Use the `embed_dim` from the config (stored in the class) rather than `hidden_state` because `attn_output` can be
        # partitioned across GPUs when using tensor-parallelism.
        attn_output = attn_output.reshape(bsz, tgt_len, self.embed_dim)

        attn_output = self.out_proj(attn_output)

        return attn_output, None, past_key_value


MERALION_SPEECH_ATTENTION_CLASSES = {
    "eager": MERaLiONSpeechAttention,
    "flash_attention_2": MERaLiONSpeechFlashAttention2,
    "sdpa": MERaLiONSpeechSdpaAttention,
}


# Copied from transformers.models.mbart.modeling_mbart.MBartEncoderLayer with MBart->Speech, MBART->WHISPER
class MERaLiONSpeechEncoderLayer(nn.Module):
    def __init__(self, config: MERaLiONSpeechConfig):
        super().__init__()
        self.embed_dim = config.d_model

        self.self_attn = MERALION_SPEECH_ATTENTION_CLASSES[config._attn_implementation](
            embed_dim=self.embed_dim,
            num_heads=config.encoder_attention_heads,
            dropout=config.attention_dropout,
            config=config,
        )
        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.Tensor,
        attention_mask: torch.Tensor,
        layer_head_mask: torch.Tensor,
        output_attentions: bool = False,
    ) -> torch.Tensor:
        """
        Args:
            hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)`
            attention_mask (`torch.FloatTensor`): attention mask of size
                `(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values.
            layer_head_mask (`torch.FloatTensor`): mask for attention heads in a given layer of size
                `(encoder_attention_heads,)`.
            output_attentions (`bool`, *optional*):
                Whether or not to return the attentions tensors of all attention layers. See `attentions` under
                returned tensors for more detail.
        """
        residual = hidden_states
        hidden_states = self.self_attn_layer_norm(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

        residual = hidden_states
        hidden_states = self.final_layer_norm(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

        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


class MERaLiONSpeechPreTrainedModel(PreTrainedModel):
    config_class = MERaLiONSpeechConfig
    base_model_prefix = "model"
    main_input_name = "input_features"
    supports_gradient_checkpointing = True
    _no_split_modules = ["MERaLiONSpeechEncoderLayer", "MERaLiONSpeechDecoderLayer"]
    _supports_flash_attn_2 = True
    _supports_sdpa = True
    _supports_cache_class = True
    _supports_static_cache = True

    def _init_weights(self, module):
        std = self.config.init_std
        if isinstance(module, (nn.Linear, nn.Conv1d)):
            module.weight.data.normal_(mean=0.0, std=std)
            if module.bias is not None:
                module.bias.data.zero_()
        elif isinstance(module, nn.Embedding):
            module.weight.data.normal_(mean=0.0, std=std)
            if module.padding_idx is not None:
                module.weight.data[module.padding_idx].zero_()
        elif isinstance(module, MERaLiONSpeechEncoder):
            with torch.no_grad():
                embed_positions = module.embed_positions.weight
                embed_positions.copy_(sinusoids(*embed_positions.shape))

    def _get_feat_extract_output_lengths(self, input_lengths: torch.LongTensor):
        """
        Computes the output length of the convolutional layers
        """
        input_lengths = (input_lengths - 1) // 2 + 1

        return input_lengths


MERALION_SPEECH_START_DOCSTRING = r"""
    This model inherits from [`PreTrainedModel`]. Check the superclass documentation for the generic methods the
    library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads
    etc.)

    This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass.
    Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage
    and behavior.

    Parameters:
        config ([`MERaLiONSpeechConfig`]):
            Model configuration class with all the parameters of the model. Initializing with a config file does not
            load the weights associated with the model, only the configuration. Check out the
            [`~PreTrainedModel.from_pretrained`] method to load the model weights.
"""

MERALION_SPEECH_INPUTS_DOCSTRING = r"""
    Args:
        input_features (`torch.FloatTensor` of shape `(batch_size, feature_size, sequence_length)`):
            Float values mel features extracted from the raw speech waveform. Raw speech waveform can be obtained by
            loading a `.flac` or `.wav` audio file into an array of type `List[float]` or a `numpy.ndarray`, *e.g.* via
            the soundfile library (`pip install soundfile`). To prepare the array into `input_features`, the
            [`AutoFeatureExtractor`] should be used for extracting the mel features, padding and conversion into a
            tensor of type `torch.FloatTensor`. See [`~SpeechFeatureExtractor.__call__`]
        attention_mask (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
            Mask to avoid performing *SpecAugment* data augmentation 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#attention-mask)
        decoder_input_ids (`torch.LongTensor` of shape `(batch_size, target_sequence_length)`, *optional*):
            Indices of decoder input sequence tokens in the vocabulary.

            Indices can be obtained using [`SpeechTokenizer`]. See [`PreTrainedTokenizer.encode`] and
            [`PreTrainedTokenizer.__call__`] for details.

            [What are decoder input IDs?](../glossary#decoder-input-ids)

            Speech uses the `decoder_start_token_id` as the starting token for `decoder_input_ids` generation. If
            `past_key_values` is used, optionally only the last `decoder_input_ids` have to be input (see
            `past_key_values`).
        decoder_attention_mask (`torch.LongTensor` of shape `(batch_size, target_sequence_length)`, *optional*):
            Default behavior: generate a tensor that ignores pad tokens in `decoder_input_ids`. Causal mask will also
            be used by default.

            If you want to change padding behavior, you should read
            [`modeling_speech._prepare_decoder_attention_mask`] and modify to your needs. See diagram 1 in [the BART
            paper](https://arxiv.org/abs/1910.13461) for more information on the default strategy.
        head_mask (`torch.Tensor` of shape `(encoder_layers, encoder_attention_heads)`, *optional*):
            Mask to nullify selected heads of the attention modules in the encoder. Mask values selected in `[0, 1]`:

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

        decoder_head_mask (`torch.Tensor` of shape `(decoder_layers, decoder_attention_heads)`, *optional*):
            Mask to nullify selected heads of the attention modules in the decoder. Mask values selected in `[0, 1]`:

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

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

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

        encoder_outputs (`tuple(tuple(torch.FloatTensor)`, *optional*):
            Tuple consists of (`last_hidden_state`, *optional*: `hidden_states`, *optional*: `attentions`)
            `last_hidden_state` of shape `(batch_size, sequence_length, hidden_size)`, *optional*) is a sequence of
            hidden-states at the output of the last layer of the encoder. Used in the cross-attention of the decoder.
        past_key_values (`EncoderDecoderCache` or `tuple(tuple(torch.FloatTensor))`, *optional*):
            Pre-computed hidden-states that can be used to speed up auto-regressive (sequential) decoding. There are
            four sets of pre-computed hidden-states: key and values states in the self-attention blocks (2) and
            in the cross-attention blocks (2). The `past_key_values` are returned when `use_cache=True` is passed or
            when `config.use_cache=True`

            Two formats are allowed:
            - An [`~cache_utils.EncoderDecoderCache`] instance;
            - Tuple of `tuple(torch.FloatTensor)` of length `config.n_layers`, with each tuple having 2 tensors of shape
            `(batch_size, num_heads, sequence_length, embed_size_per_head)`) and 2 additional tensors of shape
            `(batch_size, num_heads, encoder_sequence_length, embed_size_per_head)`.

            If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those that
            don't have their past key value states given to this model) of shape `(batch_size, 1)` instead of all
            `decoder_input_ids` of shape `(batch_size, sequence_length)`.
        decoder_inputs_embeds (`torch.FloatTensor` of shape `(batch_size, target_sequence_length, hidden_size)`, *optional*):
            Optionally, instead of passing `decoder_input_ids` you can choose to directly pass an embedded
            representation. If `past_key_values` is used, optionally only the last `decoder_inputs_embeds` have to be
            input (see `past_key_values`). This is useful if you want more control over how to convert
            `decoder_input_ids` indices into associated vectors than the model's internal embedding lookup matrix.
        use_cache (`bool`, *optional*):
            If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding (see
            `past_key_values`).
        output_attentions (`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 (`bool`, *optional*):
            Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
            more detail.
        return_dict (`bool`, *optional*):
            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
        cache_position (`torch.LongTensor` of shape `(sequence_length)`, *optional*):
            Indices depicting the position of the input sequence tokens in the sequence. It is used to update the cache
            in the correct position and to infer the complete sequence length.
"""

MERALION_SPEECH_ENCODER_INPUTS_DOCSTRING = r"""
    Args:
        input_features (`torch.FloatTensor` of shape `(batch_size, feature_size, sequence_length)`):
            Float values mel features extracted from the raw speech waveform. Raw speech waveform can be obtained by
            loading a `.flac` or `.wav` audio file into an array of type `List[float]` or a `numpy.ndarray`, *e.g.* via
            the soundfile library (`pip install soundfile`). To prepare the array into `input_features`, the
            [`AutoFeatureExtractor`] should be used for extracting the mel features, padding and conversion into a
            tensor of type `torch.FloatTensor`. See [`~SpeechFeatureExtractor.__call__`]
        head_mask (`torch.Tensor` of shape `(encoder_layers, encoder_attention_heads)`, *optional*):
            Mask to nullify selected heads of the attention modules in the encoder. Mask values selected in `[0, 1]`:

            - 1 indicates the head is **not masked**,
            - 0 indicates the head is **masked**.
        encoder_outputs (`tuple(tuple(torch.FloatTensor)`, *optional*):
            Tuple consists of (`last_hidden_state`, *optional*: `hidden_states`, *optional*: `attentions`)
            `last_hidden_state` of shape `(batch_size, sequence_length, hidden_size)`, *optional*) is a sequence of
            hidden-states at the output of the last layer of the encoder.
        output_attentions (`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 (`bool`, *optional*):
            Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
            more detail.
        return_dict (`bool`, *optional*):
            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
"""


class MERaLiONSpeechEncoder(MERaLiONSpeechPreTrainedModel):
    """
    Transformer encoder consisting of *config.encoder_layers* self attention layers. Each layer is a
    [`MERaLiONSpeechEncoderLayer`].

    Args:
        config: MERaLiONSpeechConfig
    """

    def __init__(self, config: MERaLiONSpeechConfig):
        super().__init__(config)
        self.dropout = config.dropout
        self.layerdrop = config.encoder_layerdrop

        embed_dim = config.d_model
        self.num_mel_bins = config.num_mel_bins
        self.padding_idx = config.pad_token_id
        self.max_source_positions = config.max_source_positions
        self.embed_scale = math.sqrt(embed_dim) if config.scale_embedding else 1.0

        self.conv1 = nn.Conv1d(self.num_mel_bins, embed_dim, kernel_size=3, padding=1)
        self.conv2 = nn.Conv1d(embed_dim, embed_dim, kernel_size=3, stride=2, padding=1)

        self.embed_positions = nn.Embedding(self.max_source_positions, embed_dim)
        self.embed_positions.requires_grad_(False)

        self.layers = nn.ModuleList([MERaLiONSpeechEncoderLayer(config) for _ in range(config.encoder_layers)])
        self.layer_norm = nn.LayerNorm(config.d_model)

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

    def _freeze_parameters(self):
        for param in self.parameters():
            param.requires_grad = False
        self._requires_grad = False

    def get_input_embeddings(self) -> nn.Module:
        return self.conv1

    def set_input_embeddings(self, value: nn.Module):
        self.conv1 = value

    def forward(
        self,
        input_features,
        attention_mask=None,
        head_mask=None,
        output_attentions=None,
        output_hidden_states=None,
        return_dict=None,
    ):
        r"""
        Args:
            input_features (`torch.LongTensor` of shape `(batch_size, feature_size, sequence_length)`):
                Float values of mel features extracted from the raw speech waveform. Raw speech waveform can be
                obtained by loading a `.flac` or `.wav` audio file into an array of type `List[float]` or a
                `numpy.ndarray`, *e.g.* via the soundfile library (`pip install soundfile`). To prepare the array into
                `input_features`, the [`AutoFeatureExtractor`] should be used for extracting the mel features, padding
                and conversion into a tensor of type `torch.FloatTensor`. See [`~SpeechFeatureExtractor.__call__`]
            attention_mask (`torch.Tensor`)`, *optional*):
                Speech does not support masking of the `input_features`, this argument is preserved for compatibility,
                but it is not used. By default the silence in the input log mel spectrogram are ignored.
            head_mask (`torch.Tensor` of shape `(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**.
            output_attentions (`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 (`bool`, *optional*):
                Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors
                for more detail.
            return_dict (`bool`, *optional*):
                Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
        """

        expected_seq_length = self.config.max_source_positions * self.conv1.stride[0] * self.conv2.stride[0]
        if input_features.shape[-1] != expected_seq_length:
            raise ValueError(
                f"Speech expects the mel input features to be of length {expected_seq_length}, but found {input_features.shape[-1]}. Make sure to pad the input mel features to {expected_seq_length}."
            )

        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
        inputs_embeds = nn.functional.gelu(self.conv1(input_features))
        inputs_embeds = nn.functional.gelu(self.conv2(inputs_embeds))

        inputs_embeds = inputs_embeds.permute(0, 2, 1)
        embed_pos = self.embed_positions.weight

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

        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):
            if output_hidden_states:
                encoder_states = encoder_states + (hidden_states,)
            # add LayerDrop (see https://arxiv.org/abs/1909.11556 for description)
            to_drop = False
            if self.training:
                dropout_probability = torch.rand([])
                if dropout_probability < self.layerdrop:  # skip the layer
                    to_drop = True

            if to_drop:
                layer_outputs = (None, None)
            else:
                if self.gradient_checkpointing and self.training:
                    layer_outputs = self._gradient_checkpointing_func(
                        encoder_layer.__call__,
                        hidden_states,
                        None,
                        (head_mask[idx] if head_mask is not None else None),
                        output_attentions,
                    )
                else:
                    layer_outputs = encoder_layer(
                        hidden_states,
                        None,
                        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],)

        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
        )


# copied from Qwen2AudioCausalLMOutputWithPast
@dataclass
class MERaLiONOutputWithPast(ModelOutput):
    """
    Base class for MERaLiON causal language model (or autoregressive) outputs.

    Args:
        loss (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` is provided):
            Language modeling loss (for next-token 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).
        past_key_values (`tuple(tuple(torch.FloatTensor))`, *optional*, returned when `use_cache=True` is passed or when `config.use_cache=True`):
            Tuple of `tuple(torch.FloatTensor)` of length `config.n_layers`, with each tuple having 2 tensors of shape
            `(batch_size, num_heads, sequence_length, embed_size_per_head)`)

            Contains pre-computed hidden-states (key and values in the self-attention blocks) that can be used (see
            `past_key_values` input) to speed up sequential decoding.
        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, if the model has an embedding layer, +
            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 optional 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.
        attention_mask (`torch.FloatTensor`, *optional*):
            Attentions mask, used to update attention mask and position_ids.
    """

    loss: Optional[torch.FloatTensor] = None
    logits: torch.FloatTensor = None
    past_key_values: Optional[List[torch.FloatTensor]] = None
    hidden_states: Optional[Tuple[torch.FloatTensor]] = None
    attentions: Optional[Tuple[torch.FloatTensor]] = None
    attention_mask: Optional[torch.FloatTensor] = None


MERALION_START_DOCSTRING = r"""
    This model inherits from [`PreTrainedModel`]. Check the superclass documentation for the generic methods the
    library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads
    etc.)

    This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass.
    Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage
    and behavior.

    Parameters:
        config ([`MERaLiONConfig`]):
            Model configuration class with all the parameters of the model. Initializing with a config file does not
            load the weights associated with the model, only the configuration. Check out the
            [`~PreTrainedModel.from_pretrained`] method to load the model weights.
"""


@add_start_docstrings(
    "The bare MERaLiON Model outputting raw hidden-states without any specific head on top.",
    MERALION_START_DOCSTRING,
)
class MERaLiONPreTrainedModel(PreTrainedModel):
    config_class = MERaLiONConfig
    base_model_prefix = "model"
    supports_gradient_checkpointing = True
    _no_split_modules = ["MERaLiONSpeechEncoderLayer", "MERaLiONSpeechDecoderLayer", "MERaLiONTextDecoderLayer"]
    _supports_flash_attn_2 = True
    _supports_sdpa = True
    _supports_cache_class = True
    _supports_static_cache = True

    def _init_weights(self, module):
        # important: this ported version of Qwen2Audio isn't meant for training from scratch - only
        # inference and fine-tuning - so the proper init weights code has been removed
        std = self.config.init_std if hasattr(self.config, "init_std") else self.config.speech_config.init_std

        if isinstance(module, (nn.Linear, nn.Conv1d)):
            module.weight.data.normal_(mean=0.0, std=std)
            if module.bias is not None:
                module.bias.data.zero_()
        elif isinstance(module, nn.Embedding):
            module.weight.data.normal_(mean=0.0, std=std)
            if module.padding_idx is not None:
                module.weight.data[module.padding_idx].zero_()

    @property
    def _supports_sdpa(self):
        """
        Retrieve language_model's attribute to check whether the model supports
        SDPA or not.
        """
        return self.text_decoder._supports_sdpa

class MERaLiONSpeechAudioAdaper(nn.Module):
    def __init__(
        self,
        config,
        **kwargs
    ):
        super(MERaLiONSpeechAudioAdaper, self).__init__()
        speech_audio_encoder_output_dim = config.speech_config.d_model
        llm_input_hidden_size = config.text_config.hidden_size
        speech_mlp_scale_factor = config.speech_mlp_scale_factor

        self.speech_mlp_scale_factor = speech_mlp_scale_factor
        self.mlp_adapter = nn.Sequential(
            nn.Linear(
                in_features=speech_audio_encoder_output_dim * speech_mlp_scale_factor,
                out_features=speech_audio_encoder_output_dim
            ),
            nn.SiLU(),
            nn.Dropout(0.1),
        )

        self.speech_llm_proj = nn.Sequential(
                nn.Linear(
                    speech_audio_encoder_output_dim,
                    speech_audio_encoder_output_dim * 4
                ),
                nn.SiLU(),
                nn.Dropout(0.1),

                nn.Linear(
                    speech_audio_encoder_output_dim * 4,
                    llm_input_hidden_size
                ),
            )

    def forward(self, speech_embeds, **kwargs):
        B, T, C = speech_embeds.shape
        speech_embeds = self.mlp_adapter(
            speech_embeds.reshape(
                B,
                T // self.speech_mlp_scale_factor,
                C * self.speech_mlp_scale_factor,
            )
        )
        return self.speech_llm_proj(speech_embeds)


MERALION_INPUTS_DOCSTRING = r"""
    Args:
        input_ids (`torch.LongTensor` of shape `(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 [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
            [`PreTrainedTokenizer.__call__`] for details.

            [What are input IDs?](../glossary#input-ids)
        input_features (`torch.FloatTensor` of shape `(batch_size, feature_size, feature_sequence_length)`, *optional*):
            Float values mel features extracted from the raw speech waveform. Raw speech waveform can be obtained by
            loading a `.flac` or `.wav` audio file into an array of type `List[float]` or a `numpy.ndarray`, *e.g.* via
            the soundfile library (`pip install soundfile`). To prepare the array into `input_features`, the
            [`AutoFeatureExtractor`] should be used for extracting the mel features, padding and conversion into a
            tensor of type `torch.FloatTensor`. See [`~WhisperFeatureExtractor.__call__`]
        attention_mask (`torch.Tensor` of shape `(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#attention-mask)

            Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
            [`PreTrainedTokenizer.__call__`] for details.

            If `past_key_values` is used, optionally only the last `decoder_input_ids` have to be input (see
            `past_key_values`).

            If you want to change padding behavior, you should read [`modeling_opt._prepare_decoder_attention_mask`]
            and modify to your needs. See diagram 1 in [the paper](https://arxiv.org/abs/1910.13461) for more
            information on the default strategy.

            - 1 indicates the head is **not masked**,
            - 0 indicates the head is **masked**.
        feature_attention_mask (`torch.Tensor` of shape `(batch_size, feature_sequence_length)`, *optional*):
            Mask to avoid performing attention on padding feature indices. Mask values selected in `[0, 1]`:

            - 1 for tokens that are **not masked**,
            - 0 for tokens that are **masked**.
        position_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
            Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0,
            config.n_positions - 1]`. [What are position IDs?](../glossary#position-ids)
        past_key_values (`tuple(tuple(torch.FloatTensor))`, *optional*, returned when `use_cache=True` is passed or when `config.use_cache=True`):
            Tuple of `tuple(torch.FloatTensor)` of length `config.n_layers`, with each tuple having 2 tensors of shape
            `(batch_size, num_heads, sequence_length, embed_size_per_head)`) and 2 additional tensors of shape
            `(batch_size, num_heads, encoder_sequence_length, embed_size_per_head)`.

            Contains pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention
            blocks) that can be used (see `past_key_values` input) to speed up sequential decoding.

            If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those that
            don't have their past key value states given to this model) of shape `(batch_size, 1)` instead of all
            `decoder_input_ids` of shape `(batch_size, sequence_length)`.
        inputs_embeds (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*):
            Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This
            is useful if you want more control over how to convert `input_ids` indices into associated vectors than the
            model's internal embedding lookup matrix.
        use_cache (`bool`, *optional*):
            If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding (see
            `past_key_values`).
        output_attentions (`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 (`bool`, *optional*):
            Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
            more detail.
        return_dict (`bool`, *optional*):
            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
"""

@add_start_docstrings(
    """The MERALION model which consists of a audio backbone and a language model.""",
    MERALION_START_DOCSTRING,
)
class MERaLiONForConditionalGeneration(MERaLiONPreTrainedModel, GenerationMixin):
    def __init__(self, config: MERaLiONConfig):
        config.text_config._attn_implementation = config._attn_implementation
        config.speech_config._attn_implementation = config._attn_implementation

        super().__init__(config)

        self.speech_encoder = MERaLiONSpeechEncoder(config.speech_config)
        # self.speech_encoder = AutoModel.from_config(config.audio_config, attn_implementation=config._attn_implementation)

        self.ln_speech = nn.LayerNorm(config.speech_config.d_model)
        self.speech_audio_adapter = MERaLiONSpeechAudioAdaper(config)
        self.vocab_size = config.text_config.vocab_size
        self.text_decoder = MERaLiONTextForCausalLM(config.text_config)
        self.pad_token_id = self.config.pad_token_id if self.config.pad_token_id is not None else -1
        self._padding_side = "left"  # set it to left by default, user can use setter to change padding_sides
        self.post_init()

    @property
    def padding_side(self):
        return self._padding_side

    @padding_side.setter
    def padding_side(self, padding_side: str):
        if padding_side not in ["left", "right"]:
            raise ValueError(f"{padding_side} is not `left` or `right`.")
        self._padding_side = padding_side

    # Copied from transformers.models.llava.modeling_llava.LlavaForConditionalGeneration.get_input_embeddings
    def get_input_embeddings(self):
        return self.text_decoder.get_input_embeddings()

    # Copied from transformers.models.llava.modeling_llava.LlavaForConditionalGeneration.set_input_embeddings
    def set_input_embeddings(self, value):
        self.text_decoder.set_input_embeddings(value)

    # Copied from transformers.models.llava.modeling_llava.LlavaForConditionalGeneration.get_output_embeddings
    def get_output_embeddings(self):
        return self.text_decoder.get_output_embeddings()

    # Copied from transformers.models.llava.modeling_llava.LlavaForConditionalGeneration.set_output_embeddings
    def set_output_embeddings(self, new_embeddings):
        self.text_decoder.set_output_embeddings(new_embeddings)

    # Copied from transformers.models.llava.modeling_llava.LlavaForConditionalGeneration.set_decoder
    def set_decoder(self, decoder):
        self.text_decoder.set_decoder(decoder)

    # Copied from transformers.models.llava.modeling_llava.LlavaForConditionalGeneration.get_decoder
    def get_decoder(self):
        return self.text_decoder.get_decoder()

    # Copied from transformers.models.llava.modeling_llava.LlavaForConditionalGeneration.tie_weights
    def tie_weights(self):
        return self.text_decoder.tie_weights()

    # Copied from transformers.models.llava.modeling_llava.LlavaForConditionalGeneration.resize_token_embeddings
    def resize_token_embeddings(self, new_num_tokens: Optional[int] = None, pad_to_multiple_of=None) -> nn.Embedding:
        model_embeds = self.text_decoder.resize_token_embeddings(new_num_tokens, pad_to_multiple_of)
        # update vocab size
        self.config.text_config.vocab_size = model_embeds.num_embeddings
        self.vocab_size = model_embeds.num_embeddings
        return model_embeds

    def _get_multimodal_input_embeds(
        self,
        input_ids_left,
        input_ids_right,
        attention_mask_left,
        attention_mask_right,
        speech_audio_contexts_embeds,
        speech_audio_contexts_atts,
    ):
        input_embeds_left = self.text_decoder.base_model.embed_tokens(input_ids_left)
        input_embeds_right = self.text_decoder.base_model.embed_tokens(input_ids_right)

        multimodal_embeds = torch.cat(
            [
                input_embeds_left,
                speech_audio_contexts_embeds,
                input_embeds_right,
            ],
            dim=1,
        )

        multimodal_attention_mask = torch.cat(
            [
                attention_mask_left,
                speech_audio_contexts_atts,
                attention_mask_right,
            ],
            dim=1,
        )
        return multimodal_embeds, multimodal_attention_mask

    @add_start_docstrings_to_model_forward(MERALION_INPUTS_DOCSTRING)
    @replace_return_docstrings(output_type=MERaLiONOutputWithPast, config_class=_CONFIG_FOR_DOC)
    def forward(
        self,
        input_ids: torch.LongTensor = None,
        input_features: torch.FloatTensor = None,
        attention_mask: Optional[torch.Tensor] = None,
        feature_attention_mask: Optional[torch.Tensor] = None,
        position_ids: Optional[torch.LongTensor] = None,
        past_key_values: Optional[List[torch.FloatTensor]] = None,
        inputs_embeds: Optional[torch.FloatTensor] = None,
        labels: Optional[torch.LongTensor] = None,
        use_cache: Optional[bool] = None,
        cache_position: Optional[torch.LongTensor] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        return_dict: Optional[bool] = None,
    ) -> Union[Tuple, MERaLiONOutputWithPast]:
        r"""
        Args:
            labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
                Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
                config.vocab_size]` or -100 (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]`.

        Returns:
        """

        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

        speech_encoder_device = self.speech_encoder.device

        if input_features is not None:
            input_features = input_features.to(speech_encoder_device)
            feature_attention_mask = feature_attention_mask.to(speech_encoder_device)

            if inputs_embeds is None:
                speech_contexts_embeds = self.speech_encoder(input_features, attention_mask=feature_attention_mask).last_hidden_state
                speech_contexts_embeds = self.ln_speech(speech_contexts_embeds)
                speech_audio_contexts_embeds = self.speech_audio_adapter(speech_contexts_embeds)

                inputs_embeds = self.text_decoder.base_model.embed_tokens(input_ids)

                speech_mask = (input_ids == self.config.speech_token_index).unsqueeze(-1)
                speech_mask = speech_mask.expand_as(inputs_embeds).to(inputs_embeds.device)

                inputs_embeds = inputs_embeds.masked_scatter(speech_mask, speech_audio_contexts_embeds)

                input_ids = None

        outputs = self.text_decoder(
            input_ids=input_ids,
            attention_mask=attention_mask,
            position_ids=position_ids,
            past_key_values=past_key_values,
            inputs_embeds=inputs_embeds,
            use_cache=use_cache,
            cache_position=cache_position,
            output_attentions=output_attentions,
            output_hidden_states=output_hidden_states,
            return_dict=return_dict,
            labels=labels
        )

        return outputs

    # from transformers.models.gemma2.modeling_gemma2.Gemma2ForCausalLM.prepare_inputs_for_generation
    def prepare_inputs_for_generation(
        self,
        input_ids,
        attention_mask=None,
        input_features=None,
        feature_attention_mask=None,
        past_key_values=None,
        inputs_embeds=None,
        cache_position=None,
        position_ids=None,
        use_cache=None,
        **kwargs,
    ):
        # If we have cache: let's slice `input_ids` through `cache_position`, to keep only the unprocessed tokens
        # Exception 1: when passing input_embeds, input_ids may be missing entries
        # Exception 2: some generation methods do special slicing of input_ids, so we don't need to do it here
        is_first_step = cache_position[0].item() == 0
        if past_key_values is not None:
            if inputs_embeds is not None:  # Exception 1
                input_ids = input_ids[:, -cache_position.shape[0] :]
            elif input_ids.shape[1] != cache_position.shape[0]:  # Default case (the "else", a no op, is Exception 2)
                input_ids = input_ids[:, cache_position]

        if attention_mask is not None and position_ids is None:
            # create position_ids on the fly for batch generation
            position_ids = attention_mask.long().cumsum(-1) - 1
            position_ids.masked_fill_(attention_mask == 0, 1)
            if past_key_values:
                position_ids = position_ids[:, -input_ids.shape[1] :]
                # This `clone` call is needed to avoid recapturing cuda graphs with `torch.compile`'s
                # `mode="reduce-overhead`, as otherwise the input `position_ids` would have various stride
                # during the decoding. Here, simply using `.contiguous()` is not sufficient as in the
                # batch size = 1 case, `position_ids` is already contiguous but with varying stride
                # which retriggers a capture.
                position_ids = position_ids.clone(memory_format=torch.contiguous_format)

        # if `inputs_embeds` are passed, we only want to use them in the 1st generation step
        if inputs_embeds is not None and is_first_step:
            model_inputs = {"inputs_embeds": inputs_embeds, "input_ids": None}
        else:
            # The clone here is for the same reason as for `position_ids`.
            model_inputs = {"input_ids": input_ids.clone(memory_format=torch.contiguous_format), "inputs_embeds": None}

        if (
            isinstance(past_key_values, HybridCache)
            and attention_mask.ndim == 2
            and not self.config._attn_implementation == "flash_attention_2"
        ):
            if model_inputs["inputs_embeds"] is not None:
                batch_size, sequence_length, _ = model_inputs["inputs_embeds"].shape
                device = model_inputs["inputs_embeds"].device
            else:
                batch_size, sequence_length = model_inputs["input_ids"].shape
                device = model_inputs["input_ids"].device
            dtype = self.text_decoder.lm_head.weight.dtype
            min_dtype = torch.finfo(dtype).min
            attention_mask = _prepare_4d_causal_attention_mask_with_cache_position(
                attention_mask,
                sequence_length=sequence_length,
                target_length=past_key_values.get_max_length(),
                dtype=dtype,
                device=device,
                min_dtype=min_dtype,
                cache_position=cache_position,
                batch_size=batch_size,
            )

        model_inputs.update(
            {
                "attention_mask": attention_mask,
                "position_ids": position_ids,
                "cache_position": cache_position,
                "past_key_values": past_key_values,
                "use_cache": use_cache
            }
        )

        # Input ids will only be used from the second step. 
        if is_first_step:
            model_inputs["input_features"] = input_features
            model_inputs["feature_attention_mask"] = feature_attention_mask

        return model_inputs

    def _reorder_cache(self, *args, **kwargs):
        return self.text_decoder._reorder_cache(*args, **kwargs)