import math
import torch
from torch import nn
from torch.nn import functional as F

from torch.nn import Conv1d
from torch.nn.utils import weight_norm, remove_weight_norm
from utils.util import init_weights, get_padding, convert_pad_shape, convert_pad_shape, subsequent_mask, fused_add_tanh_sigmoid_multiply
from .common.transforms import piecewise_rational_quadratic_transform

LRELU_SLOPE = 0.1

class LayerNorm(nn.Module):
  def __init__(self, channels, eps=1e-5):
    super().__init__()
    self.channels = channels
    self.eps = eps

    self.gamma = nn.Parameter(torch.ones(channels))
    self.beta = nn.Parameter(torch.zeros(channels))

  def forward(self, x):
    x = x.transpose(1, -1)
    x = F.layer_norm(x, (self.channels,), self.gamma, self.beta, self.eps)
    return x.transpose(1, -1)

 
class ConvReluNorm(nn.Module):
  def __init__(self, in_channels, hidden_channels, out_channels, kernel_size, n_layers, p_dropout):
    super().__init__()
    self.in_channels = in_channels
    self.hidden_channels = hidden_channels
    self.out_channels = out_channels
    self.kernel_size = kernel_size
    self.n_layers = n_layers
    self.p_dropout = p_dropout
    assert n_layers > 1, "Number of layers should be larger than 0."

    self.conv_layers = nn.ModuleList()
    self.norm_layers = nn.ModuleList()
    self.conv_layers.append(nn.Conv1d(in_channels, hidden_channels, kernel_size, padding=kernel_size//2))
    self.norm_layers.append(LayerNorm(hidden_channels))
    self.relu_drop = nn.Sequential(
        nn.ReLU(),
        nn.Dropout(p_dropout))
    for _ in range(n_layers-1):
      self.conv_layers.append(nn.Conv1d(hidden_channels, hidden_channels, kernel_size, padding=kernel_size//2))
      self.norm_layers.append(LayerNorm(hidden_channels))
    self.proj = nn.Conv1d(hidden_channels, out_channels, 1)
    self.proj.weight.data.zero_()
    self.proj.bias.data.zero_()

  def forward(self, x, x_mask):
    x_org = x
    for i in range(self.n_layers):
      x = self.conv_layers[i](x * x_mask)
      x = self.norm_layers[i](x)
      x = self.relu_drop(x)
    x = x_org + self.proj(x)
    return x * x_mask


class DDSConv(nn.Module):
  """
  Dilated and Depth-Separable Convolution
  """
  def __init__(self, channels, kernel_size, n_layers, p_dropout=0.):
    super().__init__()
    self.channels = channels
    self.kernel_size = kernel_size
    self.n_layers = n_layers
    self.p_dropout = p_dropout

    self.drop = nn.Dropout(p_dropout)
    self.convs_sep = nn.ModuleList()
    self.convs_1x1 = nn.ModuleList()
    self.norms_1 = nn.ModuleList()
    self.norms_2 = nn.ModuleList()
    for i in range(n_layers):
      dilation = kernel_size ** i
      padding = (kernel_size * dilation - dilation) // 2
      self.convs_sep.append(nn.Conv1d(channels, channels, kernel_size, 
          groups=channels, dilation=dilation, padding=padding
      ))
      self.convs_1x1.append(nn.Conv1d(channels, channels, 1))
      self.norms_1.append(LayerNorm(channels))
      self.norms_2.append(LayerNorm(channels))

  def forward(self, x, x_mask, g=None):
    if g is not None:
      x = x + g
    for i in range(self.n_layers):
      y = self.convs_sep[i](x * x_mask)
      y = self.norms_1[i](y)
      y = F.gelu(y)
      y = self.convs_1x1[i](y)
      y = self.norms_2[i](y)
      y = F.gelu(y)
      y = self.drop(y)
      x = x + y
    return x * x_mask


class WN(torch.nn.Module):
  def __init__(self, hidden_channels, kernel_size, dilation_rate, n_layers, gin_channels=0, p_dropout=0):
    super(WN, self).__init__()
    assert(kernel_size % 2 == 1)
    self.hidden_channels =hidden_channels
    self.kernel_size = kernel_size,
    self.dilation_rate = dilation_rate
    self.n_layers = n_layers
    self.gin_channels = gin_channels
    self.p_dropout = p_dropout

    self.in_layers = torch.nn.ModuleList()
    self.res_skip_layers = torch.nn.ModuleList()
    self.drop = nn.Dropout(p_dropout)

    if gin_channels != 0:
      cond_layer = torch.nn.Conv1d(gin_channels, 2*hidden_channels*n_layers, 1)
      self.cond_layer = torch.nn.utils.weight_norm(cond_layer, name='weight')

    for i in range(n_layers):
      dilation = dilation_rate ** i
      padding = int((kernel_size * dilation - dilation) / 2)
      in_layer = torch.nn.Conv1d(hidden_channels, 2*hidden_channels, kernel_size,
                                 dilation=dilation, padding=padding)
      in_layer = torch.nn.utils.weight_norm(in_layer, name='weight')
      self.in_layers.append(in_layer)

      # last one is not necessary
      if i < n_layers - 1:
        res_skip_channels = 2 * hidden_channels
      else:
        res_skip_channels = hidden_channels

      res_skip_layer = torch.nn.Conv1d(hidden_channels, res_skip_channels, 1)
      res_skip_layer = torch.nn.utils.weight_norm(res_skip_layer, name='weight')
      self.res_skip_layers.append(res_skip_layer)

  def forward(self, x, x_mask, g=None, **kwargs):
    output = torch.zeros_like(x)
    n_channels_tensor = torch.IntTensor([self.hidden_channels])

    if g is not None:
      g = self.cond_layer(g)

    for i in range(self.n_layers):
      x_in = self.in_layers[i](x)
      if g is not None:
        cond_offset = i * 2 * self.hidden_channels
        g_l = g[:,cond_offset:cond_offset+2*self.hidden_channels,:]
      else:
        g_l = torch.zeros_like(x_in)

      acts = fused_add_tanh_sigmoid_multiply(
          x_in,
          g_l,
          n_channels_tensor)
      acts = self.drop(acts)

      res_skip_acts = self.res_skip_layers[i](acts)
      if i < self.n_layers - 1:
        res_acts = res_skip_acts[:,:self.hidden_channels,:]
        x = (x + res_acts) * x_mask
        output = output + res_skip_acts[:,self.hidden_channels:,:]
      else:
        output = output + res_skip_acts
    return output * x_mask

  def remove_weight_norm(self):
    if self.gin_channels != 0:
      torch.nn.utils.remove_weight_norm(self.cond_layer)
    for l in self.in_layers:
      torch.nn.utils.remove_weight_norm(l)
    for l in self.res_skip_layers:
     torch.nn.utils.remove_weight_norm(l)


class ResBlock1(torch.nn.Module):
    def __init__(self, channels, kernel_size=3, dilation=(1, 3, 5)):
        super(ResBlock1, self).__init__()
        self.convs1 = nn.ModuleList([
            weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=dilation[0],
                               padding=get_padding(kernel_size, dilation[0]))),
            weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=dilation[1],
                               padding=get_padding(kernel_size, dilation[1]))),
            weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=dilation[2],
                               padding=get_padding(kernel_size, dilation[2])))
        ])
        self.convs1.apply(init_weights)

        self.convs2 = nn.ModuleList([
            weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=1,
                               padding=get_padding(kernel_size, 1))),
            weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=1,
                               padding=get_padding(kernel_size, 1))),
            weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=1,
                               padding=get_padding(kernel_size, 1)))
        ])
        self.convs2.apply(init_weights)

    def forward(self, x, x_mask=None):
        for c1, c2 in zip(self.convs1, self.convs2):
            xt = F.leaky_relu(x, LRELU_SLOPE)
            if x_mask is not None:
                xt = xt * x_mask
            xt = c1(xt)
            xt = F.leaky_relu(xt, LRELU_SLOPE)
            if x_mask is not None:
                xt = xt * x_mask
            xt = c2(xt)
            x = xt + x
        if x_mask is not None:
            x = x * x_mask
        return x

    def remove_weight_norm(self):
        for l in self.convs1:
            remove_weight_norm(l)
        for l in self.convs2:
            remove_weight_norm(l)


class ResBlock2(torch.nn.Module):
    def __init__(self, channels, kernel_size=3, dilation=(1, 3)):
        super(ResBlock2, self).__init__()
        self.convs = nn.ModuleList([
            weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=dilation[0],
                               padding=get_padding(kernel_size, dilation[0]))),
            weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=dilation[1],
                               padding=get_padding(kernel_size, dilation[1])))
        ])
        self.convs.apply(init_weights)

    def forward(self, x, x_mask=None):
        for c in self.convs:
            xt = F.leaky_relu(x, LRELU_SLOPE)
            if x_mask is not None:
                xt = xt * x_mask
            xt = c(xt)
            x = xt + x
        if x_mask is not None:
            x = x * x_mask
        return x

    def remove_weight_norm(self):
        for l in self.convs:
            remove_weight_norm(l)


class Log(nn.Module):
  def forward(self, x, x_mask, reverse=False, **kwargs):
    if not reverse:
      y = torch.log(torch.clamp_min(x, 1e-5)) * x_mask
      logdet = torch.sum(-y, [1, 2])
      return y, logdet
    else:
      x = torch.exp(x) * x_mask
      return x
    

class Flip(nn.Module):
  def forward(self, x, *args, reverse=False, **kwargs):
    x = torch.flip(x, [1])
    if not reverse:
      logdet = torch.zeros(x.size(0)).to(dtype=x.dtype, device=x.device)
      return x, logdet
    else:
      return x


class ElementwiseAffine(nn.Module):
  def __init__(self, channels):
    super().__init__()
    self.channels = channels
    self.m = nn.Parameter(torch.zeros(channels,1))
    self.logs = nn.Parameter(torch.zeros(channels,1))

  def forward(self, x, x_mask, reverse=False, **kwargs):
    if not reverse:
      y = self.m + torch.exp(self.logs) * x
      y = y * x_mask
      logdet = torch.sum(self.logs * x_mask, [1,2])
      return y, logdet
    else:
      x = (x - self.m) * torch.exp(-self.logs) * x_mask
      return x


class ResidualCouplingLayer(nn.Module):
  def __init__(self,
      channels,
      hidden_channels,
      kernel_size,
      dilation_rate,
      n_layers,
      p_dropout=0,
      gin_channels=0,
      mean_only=False):
    assert channels % 2 == 0, "channels should be divisible by 2"
    super().__init__()
    self.channels = channels
    self.hidden_channels = hidden_channels
    self.kernel_size = kernel_size
    self.dilation_rate = dilation_rate
    self.n_layers = n_layers
    self.half_channels = channels // 2
    self.mean_only = mean_only

    self.pre = nn.Conv1d(self.half_channels, hidden_channels, 1)
    self.enc = WN(hidden_channels, kernel_size, dilation_rate, n_layers, p_dropout=p_dropout, gin_channels=gin_channels)
    self.post = nn.Conv1d(hidden_channels, self.half_channels * (2 - mean_only), 1)
    self.post.weight.data.zero_()
    self.post.bias.data.zero_()

  def forward(self, x, x_mask, g=None, reverse=False):
    x0, x1 = torch.split(x, [self.half_channels]*2, 1)
    h = self.pre(x0) * x_mask
    h = self.enc(h, x_mask, g=g)
    stats = self.post(h) * x_mask
    if not self.mean_only:
      m, logs = torch.split(stats, [self.half_channels]*2, 1)
    else:
      m = stats
      logs = torch.zeros_like(m)

    if not reverse:
      x1 = m + x1 * torch.exp(logs) * x_mask
      x = torch.cat([x0, x1], 1)
      logdet = torch.sum(logs, [1,2])
      return x, logdet
    else:
      x1 = (x1 - m) * torch.exp(-logs) * x_mask
      x = torch.cat([x0, x1], 1)
      return x


class ConvFlow(nn.Module):
  def __init__(self, in_channels, filter_channels, kernel_size, n_layers, num_bins=10, tail_bound=5.0):
    super().__init__()
    self.in_channels = in_channels
    self.filter_channels = filter_channels
    self.kernel_size = kernel_size
    self.n_layers = n_layers
    self.num_bins = num_bins
    self.tail_bound = tail_bound
    self.half_channels = in_channels // 2

    self.pre = nn.Conv1d(self.half_channels, filter_channels, 1)
    self.convs = DDSConv(filter_channels, kernel_size, n_layers, p_dropout=0.)
    self.proj = nn.Conv1d(filter_channels, self.half_channels * (num_bins * 3 - 1), 1)
    self.proj.weight.data.zero_()
    self.proj.bias.data.zero_()

  def forward(self, x, x_mask, g=None, reverse=False):
    x0, x1 = torch.split(x, [self.half_channels]*2, 1)
    h = self.pre(x0)
    h = self.convs(h, x_mask, g=g)
    h = self.proj(h) * x_mask

    b, c, t = x0.shape
    h = h.reshape(b, c, -1, t).permute(0, 1, 3, 2) # [b, cx?, t] -> [b, c, t, ?]

    unnormalized_widths = h[..., :self.num_bins] / math.sqrt(self.filter_channels)
    unnormalized_heights = h[..., self.num_bins:2*self.num_bins] / math.sqrt(self.filter_channels)
    unnormalized_derivatives = h[..., 2 * self.num_bins:]

    x1, logabsdet = piecewise_rational_quadratic_transform(x1,
        unnormalized_widths,
        unnormalized_heights,
        unnormalized_derivatives,
        inverse=reverse,
        tails='linear',
        tail_bound=self.tail_bound
    )

    x = torch.cat([x0, x1], 1) * x_mask
    logdet = torch.sum(logabsdet * x_mask, [1,2])
    if not reverse:
        return x, logdet
    else:
        return x

class Encoder(nn.Module):
  def __init__(self, hidden_channels, filter_channels, n_heads, n_layers, kernel_size=1, p_dropout=0., window_size=4, **kwargs):
    super().__init__()
    self.hidden_channels = hidden_channels
    self.filter_channels = filter_channels
    self.n_heads = n_heads
    self.n_layers = n_layers
    self.kernel_size = kernel_size
    self.p_dropout = p_dropout
    self.window_size = window_size

    self.drop = nn.Dropout(p_dropout)
    self.attn_layers = nn.ModuleList()
    self.norm_layers_1 = nn.ModuleList()
    self.ffn_layers = nn.ModuleList()
    self.norm_layers_2 = nn.ModuleList()
    for i in range(self.n_layers):
      self.attn_layers.append(MultiHeadAttention(hidden_channels, hidden_channels, n_heads, p_dropout=p_dropout, window_size=window_size))
      self.norm_layers_1.append(LayerNorm(hidden_channels))
      self.ffn_layers.append(FFN(hidden_channels, hidden_channels, filter_channels, kernel_size, p_dropout=p_dropout))
      self.norm_layers_2.append(LayerNorm(hidden_channels))

  def forward(self, x, x_mask):
    attn_mask = x_mask.unsqueeze(2) * x_mask.unsqueeze(-1)
    x = x * x_mask
    for i in range(self.n_layers):
      y = self.attn_layers[i](x, x, attn_mask)
      y = self.drop(y)
      x = self.norm_layers_1[i](x + y)

      y = self.ffn_layers[i](x, x_mask)
      y = self.drop(y)
      x = self.norm_layers_2[i](x + y)
    x = x * x_mask
    return x


class Decoder(nn.Module):
  def __init__(self, hidden_channels, filter_channels, n_heads, n_layers, kernel_size=1, p_dropout=0., proximal_bias=False, proximal_init=True, **kwargs):
    super().__init__()
    self.hidden_channels = hidden_channels
    self.filter_channels = filter_channels
    self.n_heads = n_heads
    self.n_layers = n_layers
    self.kernel_size = kernel_size
    self.p_dropout = p_dropout
    self.proximal_bias = proximal_bias
    self.proximal_init = proximal_init

    self.drop = nn.Dropout(p_dropout)
    self.self_attn_layers = nn.ModuleList()
    self.norm_layers_0 = nn.ModuleList()
    self.encdec_attn_layers = nn.ModuleList()
    self.norm_layers_1 = nn.ModuleList()
    self.ffn_layers = nn.ModuleList()
    self.norm_layers_2 = nn.ModuleList()
    for i in range(self.n_layers):
      self.self_attn_layers.append(MultiHeadAttention(hidden_channels, hidden_channels, n_heads, p_dropout=p_dropout, proximal_bias=proximal_bias, proximal_init=proximal_init))
      self.norm_layers_0.append(LayerNorm(hidden_channels))
      self.encdec_attn_layers.append(MultiHeadAttention(hidden_channels, hidden_channels, n_heads, p_dropout=p_dropout))
      self.norm_layers_1.append(LayerNorm(hidden_channels))
      self.ffn_layers.append(FFN(hidden_channels, hidden_channels, filter_channels, kernel_size, p_dropout=p_dropout, causal=True))
      self.norm_layers_2.append(LayerNorm(hidden_channels))

  def forward(self, x, x_mask, h, h_mask):
    """
    x: decoder input
    h: encoder output
    """
    self_attn_mask = subsequent_mask(x_mask.size(2)).to(device=x.device, dtype=x.dtype)
    encdec_attn_mask = h_mask.unsqueeze(2) * x_mask.unsqueeze(-1)
    x = x * x_mask
    for i in range(self.n_layers):
      y = self.self_attn_layers[i](x, x, self_attn_mask)
      y = self.drop(y)
      x = self.norm_layers_0[i](x + y)

      y = self.encdec_attn_layers[i](x, h, encdec_attn_mask)
      y = self.drop(y)
      x = self.norm_layers_1[i](x + y)
      
      y = self.ffn_layers[i](x, x_mask)
      y = self.drop(y)
      x = self.norm_layers_2[i](x + y)
    x = x * x_mask
    return x


class MultiHeadAttention(nn.Module):
  def __init__(self, channels, out_channels, n_heads, p_dropout=0., window_size=None, heads_share=True, block_length=None, proximal_bias=False, proximal_init=False):
    super().__init__()
    assert channels % n_heads == 0

    self.channels = channels
    self.out_channels = out_channels
    self.n_heads = n_heads
    self.p_dropout = p_dropout
    self.window_size = window_size
    self.heads_share = heads_share
    self.block_length = block_length
    self.proximal_bias = proximal_bias
    self.proximal_init = proximal_init
    self.attn = None

    self.k_channels = channels // n_heads
    self.conv_q = nn.Conv1d(channels, channels, 1)
    self.conv_k = nn.Conv1d(channels, channels, 1)
    self.conv_v = nn.Conv1d(channels, channels, 1)
    self.conv_o = nn.Conv1d(channels, out_channels, 1)
    self.drop = nn.Dropout(p_dropout)

    if window_size is not None:
      n_heads_rel = 1 if heads_share else n_heads
      rel_stddev = self.k_channels**-0.5
      self.emb_rel_k = nn.Parameter(torch.randn(n_heads_rel, window_size * 2 + 1, self.k_channels) * rel_stddev)
      self.emb_rel_v = nn.Parameter(torch.randn(n_heads_rel, window_size * 2 + 1, self.k_channels) * rel_stddev)

    nn.init.xavier_uniform_(self.conv_q.weight)
    nn.init.xavier_uniform_(self.conv_k.weight)
    nn.init.xavier_uniform_(self.conv_v.weight)
    if proximal_init:
      with torch.no_grad():
        self.conv_k.weight.copy_(self.conv_q.weight)
        self.conv_k.bias.copy_(self.conv_q.bias)
      
  def forward(self, x, c, attn_mask=None):
    q = self.conv_q(x)
    k = self.conv_k(c)
    v = self.conv_v(c)
    
    x, self.attn = self.attention(q, k, v, mask=attn_mask)

    x = self.conv_o(x)
    return x

  def attention(self, query, key, value, mask=None):
    # reshape [b, d, t] -> [b, n_h, t, d_k]
    b, d, t_s, t_t = (*key.size(), query.size(2))
    query = query.view(b, self.n_heads, self.k_channels, t_t).transpose(2, 3)
    key = key.view(b, self.n_heads, self.k_channels, t_s).transpose(2, 3)
    value = value.view(b, self.n_heads, self.k_channels, t_s).transpose(2, 3)

    scores = torch.matmul(query / math.sqrt(self.k_channels), key.transpose(-2, -1))
    if self.window_size is not None:
      assert t_s == t_t, "Relative attention is only available for self-attention."
      key_relative_embeddings = self._get_relative_embeddings(self.emb_rel_k, t_s)
      rel_logits = self._matmul_with_relative_keys(query /math.sqrt(self.k_channels), key_relative_embeddings)
      scores_local = self._relative_position_to_absolute_position(rel_logits)
      scores = scores + scores_local
    if self.proximal_bias:
      assert t_s == t_t, "Proximal bias is only available for self-attention."
      scores = scores + self._attention_bias_proximal(t_s).to(device=scores.device, dtype=scores.dtype)
    if mask is not None:
      scores = scores.masked_fill(mask == 0, -1e4)
      if self.block_length is not None:
        assert t_s == t_t, "Local attention is only available for self-attention."
        block_mask = torch.ones_like(scores).triu(-self.block_length).tril(self.block_length)
        scores = scores.masked_fill(block_mask == 0, -1e4)
    p_attn = F.softmax(scores, dim=-1) # [b, n_h, t_t, t_s]
    p_attn = self.drop(p_attn)
    output = torch.matmul(p_attn, value)
    if self.window_size is not None:
      relative_weights = self._absolute_position_to_relative_position(p_attn)
      value_relative_embeddings = self._get_relative_embeddings(self.emb_rel_v, t_s)
      output = output + self._matmul_with_relative_values(relative_weights, value_relative_embeddings)
    output = output.transpose(2, 3).contiguous().view(b, d, t_t) # [b, n_h, t_t, d_k] -> [b, d, t_t]
    return output, p_attn

  def _matmul_with_relative_values(self, x, y):
    """
    x: [b, h, l, m]
    y: [h or 1, m, d]
    ret: [b, h, l, d]
    """
    ret = torch.matmul(x, y.unsqueeze(0))
    return ret

  def _matmul_with_relative_keys(self, x, y):
    """
    x: [b, h, l, d]
    y: [h or 1, m, d]
    ret: [b, h, l, m]
    """
    ret = torch.matmul(x, y.unsqueeze(0).transpose(-2, -1))
    return ret

  def _get_relative_embeddings(self, relative_embeddings, length):
    max_relative_position = 2 * self.window_size + 1
    # Pad first before slice to avoid using cond ops.
    pad_length = max(length - (self.window_size + 1), 0)
    slice_start_position = max((self.window_size + 1) - length, 0)
    slice_end_position = slice_start_position + 2 * length - 1
    if pad_length > 0:
      padded_relative_embeddings = F.pad(
          relative_embeddings,
          convert_pad_shape([[0, 0], [pad_length, pad_length], [0, 0]]))
    else:
      padded_relative_embeddings = relative_embeddings
    used_relative_embeddings = padded_relative_embeddings[:,slice_start_position:slice_end_position]
    return used_relative_embeddings

  def _relative_position_to_absolute_position(self, x):
    """
    x: [b, h, l, 2*l-1]
    ret: [b, h, l, l]
    """
    batch, heads, length, _ = x.size()
    # Concat columns of pad to shift from relative to absolute indexing.
    x = F.pad(x, convert_pad_shape([[0,0],[0,0],[0,0],[0,1]]))

    # Concat extra elements so to add up to shape (len+1, 2*len-1).
    x_flat = x.view([batch, heads, length * 2 * length])
    x_flat = F.pad(x_flat, convert_pad_shape([[0,0],[0,0],[0,length-1]]))

    # Reshape and slice out the padded elements.
    x_final = x_flat.view([batch, heads, length+1, 2*length-1])[:, :, :length, length-1:]
    return x_final

  def _absolute_position_to_relative_position(self, x):
    """
    x: [b, h, l, l]
    ret: [b, h, l, 2*l-1]
    """
    batch, heads, length, _ = x.size()
    # padd along column
    x = F.pad(x, convert_pad_shape([[0, 0], [0, 0], [0, 0], [0, length-1]]))
    x_flat = x.view([batch, heads, length**2 + length*(length -1)])
    # add 0's in the beginning that will skew the elements after reshape
    x_flat = F.pad(x_flat, convert_pad_shape([[0, 0], [0, 0], [length, 0]]))
    x_final = x_flat.view([batch, heads, length, 2*length])[:,:,:,1:]
    return x_final

  def _attention_bias_proximal(self, length):
    """Bias for self-attention to encourage attention to close positions.
    Args:
      length: an integer scalar.
    Returns:
      a Tensor with shape [1, 1, length, length]
    """
    r = torch.arange(length, dtype=torch.float32)
    diff = torch.unsqueeze(r, 0) - torch.unsqueeze(r, 1)
    return torch.unsqueeze(torch.unsqueeze(-torch.log1p(torch.abs(diff)), 0), 0)


class FFN(nn.Module):
  def __init__(self, in_channels, out_channels, filter_channels, kernel_size, p_dropout=0., activation=None, causal=False):
    super().__init__()
    self.in_channels = in_channels
    self.out_channels = out_channels
    self.filter_channels = filter_channels
    self.kernel_size = kernel_size
    self.p_dropout = p_dropout
    self.activation = activation
    self.causal = causal

    if causal:
      self.padding = self._causal_padding
    else:
      self.padding = self._same_padding

    self.conv_1 = nn.Conv1d(in_channels, filter_channels, kernel_size)
    self.conv_2 = nn.Conv1d(filter_channels, out_channels, kernel_size)
    self.drop = nn.Dropout(p_dropout)

  def forward(self, x, x_mask):
    x = self.conv_1(self.padding(x * x_mask))
    if self.activation == "gelu":
      x = x * torch.sigmoid(1.702 * x)
    else:
      x = torch.relu(x)
    x = self.drop(x)
    x = self.conv_2(self.padding(x * x_mask))
    return x * x_mask
  
  def _causal_padding(self, x):
    if self.kernel_size == 1:
      return x
    pad_l = self.kernel_size - 1
    pad_r = 0
    padding = [[0, 0], [0, 0], [pad_l, pad_r]]
    x = F.pad(x, convert_pad_shape(padding))
    return x

  def _same_padding(self, x):
    if self.kernel_size == 1:
      return x
    pad_l = (self.kernel_size - 1) // 2
    pad_r = self.kernel_size // 2
    padding = [[0, 0], [0, 0], [pad_l, pad_r]]
    x = F.pad(x, convert_pad_shape(padding))
    return x