from typing import List
from typing import Optional
from typing import Sequence
from typing import Tuple
from typing import Union
import logging
import torch
import torch.nn as nn
from torch.utils.checkpoint import checkpoint


def make_pad_mask(lengths, xs=None, length_dim=-1, maxlen=None):
    """Make mask tensor containing indices of padded part.

    Args:
        lengths (LongTensor or List): Batch of lengths (B,).
        xs (Tensor, optional): The reference tensor.
            If set, masks will be the same shape as this tensor.
        length_dim (int, optional): Dimension indicator of the above tensor.
            See the example.

    Returns:
        Tensor: Mask tensor containing indices of padded part.
                dtype=torch.uint8 in PyTorch 1.2-
                dtype=torch.bool in PyTorch 1.2+ (including 1.2)

    Examples:
        With only lengths.

        >>> lengths = [5, 3, 2]
        >>> make_pad_mask(lengths)
        masks = [[0, 0, 0, 0 ,0],
                 [0, 0, 0, 1, 1],
                 [0, 0, 1, 1, 1]]

        With the reference tensor.

        >>> xs = torch.zeros((3, 2, 4))
        >>> make_pad_mask(lengths, xs)
        tensor([[[0, 0, 0, 0],
                 [0, 0, 0, 0]],
                [[0, 0, 0, 1],
                 [0, 0, 0, 1]],
                [[0, 0, 1, 1],
                 [0, 0, 1, 1]]], dtype=torch.uint8)
        >>> xs = torch.zeros((3, 2, 6))
        >>> make_pad_mask(lengths, xs)
        tensor([[[0, 0, 0, 0, 0, 1],
                 [0, 0, 0, 0, 0, 1]],
                [[0, 0, 0, 1, 1, 1],
                 [0, 0, 0, 1, 1, 1]],
                [[0, 0, 1, 1, 1, 1],
                 [0, 0, 1, 1, 1, 1]]], dtype=torch.uint8)

        With the reference tensor and dimension indicator.

        >>> xs = torch.zeros((3, 6, 6))
        >>> make_pad_mask(lengths, xs, 1)
        tensor([[[0, 0, 0, 0, 0, 0],
                 [0, 0, 0, 0, 0, 0],
                 [0, 0, 0, 0, 0, 0],
                 [0, 0, 0, 0, 0, 0],
                 [0, 0, 0, 0, 0, 0],
                 [1, 1, 1, 1, 1, 1]],
                [[0, 0, 0, 0, 0, 0],
                 [0, 0, 0, 0, 0, 0],
                 [0, 0, 0, 0, 0, 0],
                 [1, 1, 1, 1, 1, 1],
                 [1, 1, 1, 1, 1, 1],
                 [1, 1, 1, 1, 1, 1]],
                [[0, 0, 0, 0, 0, 0],
                 [0, 0, 0, 0, 0, 0],
                 [1, 1, 1, 1, 1, 1],
                 [1, 1, 1, 1, 1, 1],
                 [1, 1, 1, 1, 1, 1],
                 [1, 1, 1, 1, 1, 1]]], dtype=torch.uint8)
        >>> make_pad_mask(lengths, xs, 2)
        tensor([[[0, 0, 0, 0, 0, 1],
                 [0, 0, 0, 0, 0, 1],
                 [0, 0, 0, 0, 0, 1],
                 [0, 0, 0, 0, 0, 1],
                 [0, 0, 0, 0, 0, 1],
                 [0, 0, 0, 0, 0, 1]],
                [[0, 0, 0, 1, 1, 1],
                 [0, 0, 0, 1, 1, 1],
                 [0, 0, 0, 1, 1, 1],
                 [0, 0, 0, 1, 1, 1],
                 [0, 0, 0, 1, 1, 1],
                 [0, 0, 0, 1, 1, 1]],
                [[0, 0, 1, 1, 1, 1],
                 [0, 0, 1, 1, 1, 1],
                 [0, 0, 1, 1, 1, 1],
                 [0, 0, 1, 1, 1, 1],
                 [0, 0, 1, 1, 1, 1],
                 [0, 0, 1, 1, 1, 1]]], dtype=torch.uint8)

    """
    if length_dim == 0:
        raise ValueError("length_dim cannot be 0: {}".format(length_dim))

    if not isinstance(lengths, list):
        lengths = lengths.tolist()
    bs = int(len(lengths))
    if maxlen is None:
        if xs is None:
            maxlen = int(max(lengths))
        else:
            maxlen = xs.size(length_dim)
    else:
        assert xs is None
        assert maxlen >= int(max(lengths))

    seq_range = torch.arange(0, maxlen, dtype=torch.int64)
    seq_range_expand = seq_range.unsqueeze(0).expand(bs, maxlen)
    seq_length_expand = seq_range_expand.new(lengths).unsqueeze(-1)
    mask = seq_range_expand >= seq_length_expand

    if xs is not None:
        assert xs.size(0) == bs, (xs.size(0), bs)

        if length_dim < 0:
            length_dim = xs.dim() + length_dim
        # ind = (:, None, ..., None, :, , None, ..., None)
        ind = tuple(slice(None) if i in (0, length_dim) else None for i in range(xs.dim()))
        mask = mask[ind].expand_as(xs).to(xs.device)
    return mask


class SinusoidalPositionEncoder(torch.nn.Module):
    """ """

    def __int__(self, d_model=80, dropout_rate=0.1):
        pass

    def encode(
        self, positions: torch.Tensor = None, depth: int = None, dtype: torch.dtype = torch.float32
    ):
        batch_size = positions.size(0)
        positions = positions.type(dtype)
        device = positions.device
        log_timescale_increment = torch.log(torch.tensor([10000], dtype=dtype, device=device)) / (
            depth / 2 - 1
        )
        inv_timescales = torch.exp(
            torch.arange(depth / 2, device=device).type(dtype) * (-log_timescale_increment)
        )
        inv_timescales = torch.reshape(inv_timescales, [batch_size, -1])
        scaled_time = torch.reshape(positions, [1, -1, 1]) * torch.reshape(
            inv_timescales, [1, 1, -1]
        )
        encoding = torch.cat([torch.sin(scaled_time), torch.cos(scaled_time)], dim=2)
        return encoding.type(dtype)

    def forward(self, x):
        batch_size, timesteps, input_dim = x.size()
        positions = torch.arange(1, timesteps + 1, device=x.device)[None, :]
        position_encoding = self.encode(positions, input_dim, x.dtype).to(x.device)

        return x + position_encoding
    

class PositionwiseFeedForward(torch.nn.Module):
    """Positionwise feed forward layer.

    Args:
        idim (int): Input dimenstion.
        hidden_units (int): The number of hidden units.
        dropout_rate (float): Dropout rate.

    """

    def __init__(self, idim, hidden_units, dropout_rate, activation=torch.nn.ReLU()):
        """Construct an PositionwiseFeedForward object."""
        super(PositionwiseFeedForward, self).__init__()
        self.w_1 = torch.nn.Linear(idim, hidden_units)
        self.w_2 = torch.nn.Linear(hidden_units, idim)
        self.dropout = torch.nn.Dropout(dropout_rate)
        self.activation = activation

    def forward(self, x):
        """Forward function."""
        return self.w_2(self.dropout(self.activation(self.w_1(x))))


class MultiHeadedAttentionSANM(nn.Module):
    """Multi-Head Attention layer.

    Args:
        n_head (int): The number of heads.
        n_feat (int): The number of features.
        dropout_rate (float): Dropout rate.

    """

    def __init__(
        self,
        n_head,
        in_feat,
        n_feat,
        dropout_rate,
        kernel_size,
        sanm_shfit=0,
        lora_list=None,
        lora_rank=8,
        lora_alpha=16,
        lora_dropout=0.1,
    ):
        """Construct an MultiHeadedAttention object."""
        super().__init__()
        assert n_feat % n_head == 0
        # We assume d_v always equals d_k
        self.d_k = n_feat // n_head
        self.h = n_head

        self.linear_out = nn.Linear(n_feat, n_feat)
        self.linear_q_k_v = nn.Linear(in_feat, n_feat * 3)
        self.attn = None
        self.dropout = nn.Dropout(p=dropout_rate)

        self.fsmn_block = nn.Conv1d(
            n_feat, n_feat, kernel_size, stride=1, padding=0, groups=n_feat, bias=False
        )
        # padding
        left_padding = (kernel_size - 1) // 2
        if sanm_shfit > 0:
            left_padding = left_padding + sanm_shfit
        right_padding = kernel_size - 1 - left_padding
        self.pad_fn = nn.ConstantPad1d((left_padding, right_padding), 0.0)

    def forward_fsmn(self, inputs, mask, mask_shfit_chunk=None):
        b, t, d = inputs.size()
        if mask is not None:
            mask = torch.reshape(mask, (b, -1, 1))
            if mask_shfit_chunk is not None:
                mask = mask * mask_shfit_chunk
            inputs = inputs * mask

        x = inputs.transpose(1, 2)
        x = self.pad_fn(x)
        x = self.fsmn_block(x)
        x = x.transpose(1, 2)
        x += inputs
        x = self.dropout(x)
        if mask is not None:
            x = x * mask
        return x

    def forward_qkv(self, x):
        """Transform query, key and value.

        Args:
            query (torch.Tensor): Query tensor (#batch, time1, size).
            key (torch.Tensor): Key tensor (#batch, time2, size).
            value (torch.Tensor): Value tensor (#batch, time2, size).

        Returns:
            torch.Tensor: Transformed query tensor (#batch, n_head, time1, d_k).
            torch.Tensor: Transformed key tensor (#batch, n_head, time2, d_k).
            torch.Tensor: Transformed value tensor (#batch, n_head, time2, d_k).

        """
        b, t, d = x.size()
        q_k_v = self.linear_q_k_v(x)
        q, k, v = torch.split(q_k_v, int(self.h * self.d_k), dim=-1)
        q_h = torch.reshape(q, (b, t, self.h, self.d_k)).transpose(
            1, 2
        )  # (batch, head, time1, d_k)
        k_h = torch.reshape(k, (b, t, self.h, self.d_k)).transpose(
            1, 2
        )  # (batch, head, time2, d_k)
        v_h = torch.reshape(v, (b, t, self.h, self.d_k)).transpose(
            1, 2
        )  # (batch, head, time2, d_k)

        return q_h, k_h, v_h, v

    def forward_attention(self, value, scores, mask, mask_att_chunk_encoder=None):
        """Compute attention context vector.

        Args:
            value (torch.Tensor): Transformed value (#batch, n_head, time2, d_k).
            scores (torch.Tensor): Attention score (#batch, n_head, time1, time2).
            mask (torch.Tensor): Mask (#batch, 1, time2) or (#batch, time1, time2).

        Returns:
            torch.Tensor: Transformed value (#batch, time1, d_model)
                weighted by the attention score (#batch, time1, time2).

        """
        n_batch = value.size(0)
        if mask is not None:
            if mask_att_chunk_encoder is not None:
                mask = mask * mask_att_chunk_encoder

            mask = mask.unsqueeze(1).eq(0)  # (batch, 1, *, time2)

            min_value = -float(
                "inf"
            )  # float(numpy.finfo(torch.tensor(0, dtype=scores.dtype).numpy().dtype).min)
            scores = scores.masked_fill(mask, min_value)
            self.attn = torch.softmax(scores, dim=-1).masked_fill(
                mask, 0.0
            )  # (batch, head, time1, time2)
        else:
            self.attn = torch.softmax(scores, dim=-1)  # (batch, head, time1, time2)

        p_attn = self.dropout(self.attn)
        x = torch.matmul(p_attn, value)  # (batch, head, time1, d_k)
        x = (
            x.transpose(1, 2).contiguous().view(n_batch, -1, self.h * self.d_k)
        )  # (batch, time1, d_model)

        return self.linear_out(x)  # (batch, time1, d_model)

    def forward(self, x, mask, mask_shfit_chunk=None, mask_att_chunk_encoder=None):
        """Compute scaled dot product attention.

        Args:
            query (torch.Tensor): Query tensor (#batch, time1, size).
            key (torch.Tensor): Key tensor (#batch, time2, size).
            value (torch.Tensor): Value tensor (#batch, time2, size).
            mask (torch.Tensor): Mask tensor (#batch, 1, time2) or
                (#batch, time1, time2).

        Returns:
            torch.Tensor: Output tensor (#batch, time1, d_model).

        """
        q_h, k_h, v_h, v = self.forward_qkv(x)
        fsmn_memory = self.forward_fsmn(v, mask, mask_shfit_chunk)
        q_h = q_h * self.d_k ** (-0.5)
        scores = torch.matmul(q_h, k_h.transpose(-2, -1))
        att_outs = self.forward_attention(v_h, scores, mask, mask_att_chunk_encoder)
        return att_outs + fsmn_memory

    def forward_chunk(self, x, cache=None, chunk_size=None, look_back=0):
        """Compute scaled dot product attention.

        Args:
            query (torch.Tensor): Query tensor (#batch, time1, size).
            key (torch.Tensor): Key tensor (#batch, time2, size).
            value (torch.Tensor): Value tensor (#batch, time2, size).
            mask (torch.Tensor): Mask tensor (#batch, 1, time2) or
                (#batch, time1, time2).

        Returns:
            torch.Tensor: Output tensor (#batch, time1, d_model).

        """
        q_h, k_h, v_h, v = self.forward_qkv(x)
        if chunk_size is not None and look_back > 0 or look_back == -1:
            if cache is not None:
                k_h_stride = k_h[:, :, : -(chunk_size[2]), :]
                v_h_stride = v_h[:, :, : -(chunk_size[2]), :]
                k_h = torch.cat((cache["k"], k_h), dim=2)
                v_h = torch.cat((cache["v"], v_h), dim=2)

                cache["k"] = torch.cat((cache["k"], k_h_stride), dim=2)
                cache["v"] = torch.cat((cache["v"], v_h_stride), dim=2)
                if look_back != -1:
                    cache["k"] = cache["k"][:, :, -(look_back * chunk_size[1]) :, :]
                    cache["v"] = cache["v"][:, :, -(look_back * chunk_size[1]) :, :]
            else:
                cache_tmp = {
                    "k": k_h[:, :, : -(chunk_size[2]), :],
                    "v": v_h[:, :, : -(chunk_size[2]), :],
                }
                cache = cache_tmp
        fsmn_memory = self.forward_fsmn(v, None)
        q_h = q_h * self.d_k ** (-0.5)
        scores = torch.matmul(q_h, k_h.transpose(-2, -1))
        att_outs = self.forward_attention(v_h, scores, None)
        return att_outs + fsmn_memory, cache
    
    
class MultiSequential(torch.nn.Sequential):
    """Multi-input multi-output torch.nn.Sequential."""

    def __init__(self, *args, layer_drop_rate=0.0):
        """Initialize MultiSequential with layer_drop.

        Args:
            layer_drop_rate (float): Probability of dropping out each fn (layer).

        """
        super(MultiSequential, self).__init__(*args)
        self.layer_drop_rate = layer_drop_rate

    def forward(self, *args):
        """Repeat."""
        _probs = torch.empty(len(self)).uniform_()
        for idx, m in enumerate(self):
            if not self.training or (_probs[idx] >= self.layer_drop_rate):
                args = m(*args)
        return args
    
    
class CheckpointFunction(torch.nn.Module):
    """Wrap a module function for gradient checkpointing."""
    def __init__(self, module_fn):
        super().__init__()
        self.module_fn = module_fn

    def forward(self, *args, **kwargs):
        # Use checkpointing on the forward pass of the module function
        return checkpoint(self.module_fn, *args, **kwargs)


def repeat(N, fn, layer_drop_rate=0.0, use_checkpoint=False):
    """Repeat module N times.

    Args:
        N (int): Number of repeat time.
        fn (Callable): Function to generate module.
        layer_drop_rate (float): Probability of dropping out each fn (layer).

    Returns:
        MultiSequential: Repeated model instance.

    """
    modules = []
    for n in range(N):
        module_fn = fn(n)
        if use_checkpoint:
            # Wrap the module function with checkpointing
            module_fn = CheckpointFunction(module_fn)
        modules.append(module_fn)
    return MultiSequential(*modules, layer_drop_rate=layer_drop_rate)


class LayerNorm(torch.nn.LayerNorm):
    """Layer normalization module.

    Args:
        nout (int): Output dim size.
        dim (int): Dimension to be normalized.

    """

    def __init__(self, nout, dim=-1):
        """Construct an LayerNorm object."""
        super(LayerNorm, self).__init__(nout, eps=1e-12)
        self.dim = dim

    def forward(self, x):
        """Apply layer normalization.

        Args:
            x (torch.Tensor): Input tensor.

        Returns:
            torch.Tensor: Normalized tensor.

        """
        if self.dim == -1:
            return super(LayerNorm, self).forward(x)
        return super(LayerNorm, self).forward(x.transpose(self.dim, -1)).transpose(self.dim, -1)
    
    
class EncoderLayerSANM(nn.Module):
    def __init__(
        self,
        in_size,
        size,
        self_attn,
        feed_forward,
        dropout_rate,
        normalize_before=True,
        concat_after=False,
        stochastic_depth_rate=0.0,
    ):
        """Construct an EncoderLayer object."""
        super(EncoderLayerSANM, self).__init__()
        self.self_attn = self_attn
        self.feed_forward = feed_forward
        self.norm1 = LayerNorm(in_size)
        self.norm2 = LayerNorm(size)
        self.dropout = nn.Dropout(dropout_rate)
        self.in_size = in_size
        self.size = size
        self.normalize_before = normalize_before
        self.concat_after = concat_after
        if self.concat_after:
            self.concat_linear = nn.Linear(size + size, size)
        self.stochastic_depth_rate = stochastic_depth_rate
        self.dropout_rate = dropout_rate

    def forward(self, x, mask, cache=None, mask_shfit_chunk=None, mask_att_chunk_encoder=None):
        """Compute encoded features.

        Args:
            x_input (torch.Tensor): Input tensor (#batch, time, size).
            mask (torch.Tensor): Mask tensor for the input (#batch, time).
            cache (torch.Tensor): Cache tensor of the input (#batch, time - 1, size).

        Returns:
            torch.Tensor: Output tensor (#batch, time, size).
            torch.Tensor: Mask tensor (#batch, time).

        """
        skip_layer = False
        # with stochastic depth, residual connection `x + f(x)` becomes
        # `x <- x + 1 / (1 - p) * f(x)` at training time.
        stoch_layer_coeff = 1.0
        if self.training and self.stochastic_depth_rate > 0:
            skip_layer = torch.rand(1).item() < self.stochastic_depth_rate
            stoch_layer_coeff = 1.0 / (1 - self.stochastic_depth_rate)

        if skip_layer:
            if cache is not None:
                x = torch.cat([cache, x], dim=1)
            return x, mask

        residual = x
        if self.normalize_before:
            x = self.norm1(x)

        if self.concat_after:
            x_concat = torch.cat(
                (
                    x,
                    self.self_attn(
                        x,
                        mask,
                        mask_shfit_chunk=mask_shfit_chunk,
                        mask_att_chunk_encoder=mask_att_chunk_encoder,
                    ),
                ),
                dim=-1,
            )
            if self.in_size == self.size:
                x = residual + stoch_layer_coeff * self.concat_linear(x_concat)
            else:
                x = stoch_layer_coeff * self.concat_linear(x_concat)
        else:
            if self.in_size == self.size:
                x = residual + stoch_layer_coeff * self.dropout(
                    self.self_attn(
                        x,
                        mask,
                        mask_shfit_chunk=mask_shfit_chunk,
                        mask_att_chunk_encoder=mask_att_chunk_encoder,
                    )
                )
            else:
                x = stoch_layer_coeff * self.dropout(
                    self.self_attn(
                        x,
                        mask,
                        mask_shfit_chunk=mask_shfit_chunk,
                        mask_att_chunk_encoder=mask_att_chunk_encoder,
                    )
                )
        if not self.normalize_before:
            x = self.norm1(x)

        residual = x
        if self.normalize_before:
            x = self.norm2(x)
        x = residual + stoch_layer_coeff * self.dropout(self.feed_forward(x))
        if not self.normalize_before:
            x = self.norm2(x)

        return x, mask, cache, mask_shfit_chunk, mask_att_chunk_encoder


class SANMEncoder(nn.Module):
    def __init__(
        self,
        input_size: int = 560,
        output_size: int = 512,
        attention_heads: int = 4,
        linear_units: int = 2048,
        num_blocks: int = 50,
        dropout_rate: float = 0.1,
        positional_dropout_rate: float = 0.1,
        attention_dropout_rate: float = 0.1,
        input_layer: Optional[str] = "pe",
        pos_enc_class=SinusoidalPositionEncoder,
        normalize_before: bool = True,
        concat_after: bool = False,
        positionwise_layer_type: str = "linear",
        positionwise_conv_kernel_size: int = 1,
        padding_idx: int = -1,
        interctc_layer_idx: List[int] = [],
        interctc_use_conditioning: bool = False,
        kernel_size: int = 11,
        sanm_shfit: int = 0,
        lora_list: List[str] = None,
        lora_rank: int = 8,
        lora_alpha: int = 16,
        lora_dropout: float = 0.1,
        selfattention_layer_type: str = "sanm",
        tf2torch_tensor_name_prefix_torch: str = "encoder",
        tf2torch_tensor_name_prefix_tf: str = "seq2seq/encoder",
        gradient_checkpoint=False
    ):
        super().__init__()
        self._output_size = output_size

        if input_layer == "linear":
            self.embed = torch.nn.Sequential(
                torch.nn.Linear(input_size, output_size),
                torch.nn.LayerNorm(output_size),
                torch.nn.Dropout(dropout_rate),
                torch.nn.ReLU(),
                pos_enc_class(output_size, positional_dropout_rate),
            )
        elif input_layer == "pe":
            self.embed = SinusoidalPositionEncoder()
        else:
            raise ValueError("unknown input_layer: " + input_layer)
        self.normalize_before = normalize_before
        if positionwise_layer_type == "linear":
            positionwise_layer = PositionwiseFeedForward
            positionwise_layer_args = (
                output_size,
                linear_units,
                dropout_rate,
            )
        else:
            raise NotImplementedError("Support only linear or conv1d.")

        if selfattention_layer_type == "sanm":
            encoder_selfattn_layer = MultiHeadedAttentionSANM
            encoder_selfattn_layer_args0 = (
                attention_heads,
                input_size,
                output_size,
                attention_dropout_rate,
                kernel_size,
                sanm_shfit,
                lora_list,
                lora_rank,
                lora_alpha,
                lora_dropout,
            )

            encoder_selfattn_layer_args = (
                attention_heads,
                output_size,
                output_size,
                attention_dropout_rate,
                kernel_size,
                sanm_shfit,
                lora_list,
                lora_rank,
                lora_alpha,
                lora_dropout,
            )
        self.encoders0 = repeat(
            1,
            lambda lnum: EncoderLayerSANM(
                input_size,
                output_size,
                encoder_selfattn_layer(*encoder_selfattn_layer_args0),
                positionwise_layer(*positionwise_layer_args),
                dropout_rate,
                normalize_before,
                concat_after,
            ),
            use_checkpoint=gradient_checkpoint
        )

        self.encoders = repeat(
            num_blocks - 1,
            lambda lnum: EncoderLayerSANM(
                output_size,
                output_size,
                encoder_selfattn_layer(*encoder_selfattn_layer_args),
                positionwise_layer(*positionwise_layer_args),
                dropout_rate,
                normalize_before,
                concat_after,
            ),
            use_checkpoint=gradient_checkpoint
        )
        if self.normalize_before:
            self.after_norm = LayerNorm(output_size)

        self.interctc_layer_idx = interctc_layer_idx
        if len(interctc_layer_idx) > 0:
            assert 0 < min(interctc_layer_idx) and max(interctc_layer_idx) < num_blocks
        self.interctc_use_conditioning = interctc_use_conditioning
        self.conditioning_layer = None
        self.dropout = nn.Dropout(dropout_rate)
        self.tf2torch_tensor_name_prefix_torch = tf2torch_tensor_name_prefix_torch
        self.tf2torch_tensor_name_prefix_tf = tf2torch_tensor_name_prefix_tf

    def output_size(self) -> int:
        return self._output_size

    def forward(
        self,
        xs_pad: torch.Tensor,
        ilens: torch.Tensor,
        prev_states: torch.Tensor = None,
        ctc = None,
    ) -> Tuple[torch.Tensor, torch.Tensor, Optional[torch.Tensor]]:
        """Embed positions in tensor.

        Args:
            xs_pad: input tensor (B, L, D)
            ilens: input length (B)
            prev_states: Not to be used now.
        Returns:
            position embedded tensor and mask
        """
        masks = (~make_pad_mask(ilens)[:, None, :]).to(xs_pad.device)
        xs_pad = xs_pad * self.output_size() ** 0.5
        xs_pad = self.embed(xs_pad)

        # xs_pad = self.dropout(xs_pad)
        encoder_outs = self.encoders0(xs_pad, masks)
        xs_pad, masks = encoder_outs[0], encoder_outs[1]
        intermediate_outs = []
        if len(self.interctc_layer_idx) == 0:
            encoder_outs = self.encoders(xs_pad, masks)
            xs_pad, masks = encoder_outs[0], encoder_outs[1]
        else:
            for layer_idx, encoder_layer in enumerate(self.encoders):
                encoder_outs = encoder_layer(xs_pad, masks)
                xs_pad, masks = encoder_outs[0], encoder_outs[1]

                if layer_idx + 1 in self.interctc_layer_idx:
                    encoder_out = xs_pad

                    # intermediate outputs are also normalized
                    if self.normalize_before:
                        encoder_out = self.after_norm(encoder_out)

                    intermediate_outs.append((layer_idx + 1, encoder_out))

                    if self.interctc_use_conditioning:
                        ctc_out = ctc.softmax(encoder_out)
                        xs_pad = xs_pad + self.conditioning_layer(ctc_out)

        if self.normalize_before:
            xs_pad = self.after_norm(xs_pad)

        olens = masks.squeeze(1).sum(1)
        if len(intermediate_outs) > 0:
            return (xs_pad, intermediate_outs), olens, None
        return xs_pad, olens, None