import torch.nn as nn
from torch import Tensor


class Transpose(nn.Module):
    """Wrapper class of torch.transpose() for Sequential module."""

    def __init__(self, shape: tuple):
        super(Transpose, self).__init__()
        self.shape = shape

    def forward(self, x: Tensor) -> Tensor:
        return x.transpose(*self.shape)


class DepthwiseConv1d(nn.Module):
    """
    When groups == in_channels and out_channels == K * in_channels, where K is a positive integer,
    this operation is termed in literature as depthwise convolution.
    Args:
        in_channels (int): Number of channels in the input
        out_channels (int): Number of channels produced by the convolution
        kernel_size (int or tuple): Size of the convolving kernel
        stride (int, optional): Stride of the convolution. Default: 1
        padding (int or tuple, optional): Zero-padding added to both sides of the input. Default: 0
        bias (bool, optional): If True, adds a learnable bias to the output. Default: True
    Inputs: inputs
        - **inputs** (batch, in_channels, time): Tensor containing input vector
    Returns: outputs
        - **outputs** (batch, out_channels, time): Tensor produces by depthwise 1-D convolution.
    """

    def __init__(
        self,
        in_channels: int,
        out_channels: int,
        kernel_size: int,
        stride: int = 1,
        padding: int = 0,
        bias: bool = False,
    ) -> None:
        super(DepthwiseConv1d, self).__init__()
        assert (
            out_channels % in_channels == 0
        ), "out_channels should be constant multiple of in_channels"
        self.conv = nn.Conv1d(
            in_channels=in_channels,
            out_channels=out_channels,
            kernel_size=kernel_size,
            groups=in_channels,
            stride=stride,
            padding=padding,
            bias=bias,
        )

    def forward(self, inputs: Tensor) -> Tensor:
        return self.conv(inputs)


class ConvModule(nn.Module):
    """
    Conformer convolution module starts with a pointwise convolution and a gated linear unit (GLU).
    This is followed by a single 1-D depthwise convolution layer. Batchnorm is  deployed just after the convolution
    to aid training deep models.
    Args:
        in_channels (int): Number of channels in the input
        kernel_size (int or tuple, optional): Size of the convolving kernel Default: 31
        dropout_p (float, optional): probability of dropout
    Inputs: inputs
        inputs (batch, time, dim): Tensor contains input sequences
    Outputs: outputs
        outputs (batch, time, dim): Tensor produces by conformer convolution module.
    """

    def __init__(
        self,
        in_channels: int,
        kernel_size: int = 17,
        expansion_factor: int = 2,
        dropout_p: float = 0.1,
    ) -> None:
        super(ConvModule, self).__init__()
        assert (
            kernel_size - 1
        ) % 2 == 0, "kernel_size should be a odd number for 'SAME' padding"
        assert expansion_factor == 2, "Currently, Only Supports expansion_factor 2"

        self.sequential = nn.Sequential(
            Transpose(shape=(1, 2)),
            DepthwiseConv1d(
                in_channels,
                in_channels,
                kernel_size,
                stride=1,
                padding=(kernel_size - 1) // 2,
            ),
        )

    def forward(self, inputs: Tensor) -> Tensor:
        return inputs + self.sequential(inputs).transpose(1, 2)