diffusers-version/tora/traj_module.py [25:171]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
def avg_pool_nd(dims, *args, **kwargs):
    """
    Create a 1D, 2D, or 3D average pooling module.
    """
    if dims == 1:
        return nn.AvgPool1d(*args, **kwargs)
    elif dims == 2:
        return nn.AvgPool2d(*args, **kwargs)
    elif dims == 3:
        return nn.AvgPool3d(*args, **kwargs)
    raise ValueError(f"unsupported dimensions: {dims}")


def conv_nd(dims, *args, **kwargs):
    """
    Create a 1D, 2D, or 3D convolution module.
    """
    if dims == 1:
        return nn.Conv1d(*args, **kwargs)
    elif dims == 2:
        return nn.Conv2d(*args, **kwargs)
    elif dims == 3:
        return nn.Conv3d(*args, **kwargs)
    raise ValueError(f"unsupported dimensions: {dims}")


class Downsample(nn.Module):
    """
    A downsampling layer with an optional convolution.
    :param channels: channels in the inputs and outputs.
    :param use_conv: a bool determining if a convolution is applied.
    :param dims: determines if the signal is 1D, 2D, or 3D. If 3D, then
                 downsampling occurs in the inner-two dimensions.
    """

    def __init__(self, channels, use_conv, dims=2, out_channels=None, padding=1):
        super().__init__()
        self.channels = channels
        self.out_channels = out_channels or channels
        self.use_conv = use_conv
        self.dims = dims
        stride = 2 if dims != 3 else (1, 2, 2)
        if use_conv:
            self.op = conv_nd(
                dims,
                self.channels,
                self.out_channels,
                3,
                stride=stride,
                padding=padding,
            )
        else:
            assert self.channels == self.out_channels
            self.op = avg_pool_nd(dims, kernel_size=stride, stride=stride)

    def forward(self, x):
        assert x.shape[1] == self.channels
        return self.op(x)


class ResnetBlock(nn.Module):
    def __init__(self, in_c, out_c, down, ksize=3, sk=False, use_conv=True):
        super().__init__()
        ps = ksize // 2
        if in_c != out_c or sk == False:
            self.in_conv = nn.Conv2d(in_c, out_c, ksize, 1, ps)
        else:
            # print('n_in')
            self.in_conv = None
        self.block1 = nn.Conv2d(out_c, out_c, 3, 1, 1)
        self.act = nn.ReLU()
        self.block2 = nn.Conv2d(out_c, out_c, ksize, 1, ps)
        self.bn1 = nn.BatchNorm2d(out_c)
        self.bn2 = nn.BatchNorm2d(out_c)
        if sk == False:
            # self.skep = nn.Conv2d(in_c, out_c, ksize, 1, ps) # edit by zhouxiawang
            self.skep = nn.Conv2d(out_c, out_c, ksize, 1, ps)
        else:
            self.skep = None

        self.down = down
        if self.down == True:
            self.down_opt = Downsample(in_c, use_conv=use_conv)

    def forward(self, x):
        if self.down == True:
            x = self.down_opt(x)
        if self.in_conv is not None:  # edit
            x = self.in_conv(x)

        h = self.bn1(x)
        h = self.act(h)
        h = self.block1(h)
        h = self.bn2(h)
        h = self.act(h)
        h = self.block2(h)
        if self.skep is not None:
            return h + self.skep(x)
        else:
            return h + x


class VAESpatialEmulator(nn.Module):
    def __init__(self, kernel_size=(8, 8)):
        super().__init__()
        self.kernel_size = kernel_size

    def forward(self, x):
        """
        x: torch.Tensor: shape [B C T H W]
        """
        Hp, Wp = self.kernel_size
        H, W = x.shape[-2], x.shape[-1]
        valid_h = H - H % Hp
        valid_w = W - W % Wp
        x = x[..., :valid_h, :valid_w]
        x = rearrange(
            x,
            "B C T (Nh Hp) (Nw Wp)  -> B (Hp Wp C) T Nh Nw",
            Hp=Hp,
            Wp=Wp,
        )
        return x


class VAETemporalEmulator(nn.Module):
    def __init__(self, micro_frame_size, kernel_size=4):
        super().__init__()
        self.micro_frame_size = micro_frame_size
        self.kernel_size = kernel_size

    def forward(self, x_z):
        """
        x_z: torch.Tensor: shape [B C T H W]
        """

        z_list = []
        for i in range(0, x_z.shape[2], self.micro_frame_size):
            x_z_bs = x_z[:, :, i : i + self.micro_frame_size]
            z_list.append(x_z_bs[:, :, 0:1])
            x_z_bs = x_z_bs[:, :, 1:]
            t_valid = x_z_bs.shape[2] - x_z_bs.shape[2] % self.kernel_size
            x_z_bs = x_z_bs[:, :, :t_valid]
            x_z_bs = reduce(x_z_bs, "B C (T n) H W -> B C T H W", n=self.kernel_size, reduction="mean")
            z_list.append(x_z_bs)
        z = torch.cat(z_list, dim=2)
        return z
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -



sat/sgm/modules/traj_module.py [7:153]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
def avg_pool_nd(dims, *args, **kwargs):
    """
    Create a 1D, 2D, or 3D average pooling module.
    """
    if dims == 1:
        return nn.AvgPool1d(*args, **kwargs)
    elif dims == 2:
        return nn.AvgPool2d(*args, **kwargs)
    elif dims == 3:
        return nn.AvgPool3d(*args, **kwargs)
    raise ValueError(f"unsupported dimensions: {dims}")


def conv_nd(dims, *args, **kwargs):
    """
    Create a 1D, 2D, or 3D convolution module.
    """
    if dims == 1:
        return nn.Conv1d(*args, **kwargs)
    elif dims == 2:
        return nn.Conv2d(*args, **kwargs)
    elif dims == 3:
        return nn.Conv3d(*args, **kwargs)
    raise ValueError(f"unsupported dimensions: {dims}")


class Downsample(nn.Module):
    """
    A downsampling layer with an optional convolution.
    :param channels: channels in the inputs and outputs.
    :param use_conv: a bool determining if a convolution is applied.
    :param dims: determines if the signal is 1D, 2D, or 3D. If 3D, then
                 downsampling occurs in the inner-two dimensions.
    """

    def __init__(self, channels, use_conv, dims=2, out_channels=None, padding=1):
        super().__init__()
        self.channels = channels
        self.out_channels = out_channels or channels
        self.use_conv = use_conv
        self.dims = dims
        stride = 2 if dims != 3 else (1, 2, 2)
        if use_conv:
            self.op = conv_nd(
                dims,
                self.channels,
                self.out_channels,
                3,
                stride=stride,
                padding=padding,
            )
        else:
            assert self.channels == self.out_channels
            self.op = avg_pool_nd(dims, kernel_size=stride, stride=stride)

    def forward(self, x):
        assert x.shape[1] == self.channels
        return self.op(x)


class ResnetBlock(nn.Module):
    def __init__(self, in_c, out_c, down, ksize=3, sk=False, use_conv=True):
        super().__init__()
        ps = ksize // 2
        if in_c != out_c or sk == False:
            self.in_conv = nn.Conv2d(in_c, out_c, ksize, 1, ps)
        else:
            # print('n_in')
            self.in_conv = None
        self.block1 = nn.Conv2d(out_c, out_c, 3, 1, 1)
        self.act = nn.ReLU()
        self.block2 = nn.Conv2d(out_c, out_c, ksize, 1, ps)
        self.bn1 = nn.BatchNorm2d(out_c)
        self.bn2 = nn.BatchNorm2d(out_c)
        if sk == False:
            # self.skep = nn.Conv2d(in_c, out_c, ksize, 1, ps) # edit by zhouxiawang
            self.skep = nn.Conv2d(out_c, out_c, ksize, 1, ps)
        else:
            self.skep = None

        self.down = down
        if self.down == True:
            self.down_opt = Downsample(in_c, use_conv=use_conv)

    def forward(self, x):
        if self.down == True:
            x = self.down_opt(x)
        if self.in_conv is not None:  # edit
            x = self.in_conv(x)

        h = self.bn1(x)
        h = self.act(h)
        h = self.block1(h)
        h = self.bn2(h)
        h = self.act(h)
        h = self.block2(h)
        if self.skep is not None:
            return h + self.skep(x)
        else:
            return h + x


class VAESpatialEmulator(nn.Module):
    def __init__(self, kernel_size=(8, 8)):
        super().__init__()
        self.kernel_size = kernel_size

    def forward(self, x):
        """
        x: torch.Tensor: shape [B C T H W]
        """
        Hp, Wp = self.kernel_size
        H, W = x.shape[-2], x.shape[-1]
        valid_h = H - H % Hp
        valid_w = W - W % Wp
        x = x[..., :valid_h, :valid_w]
        x = rearrange(
            x,
            "B C T (Nh Hp) (Nw Wp)  -> B (Hp Wp C) T Nh Nw",
            Hp=Hp,
            Wp=Wp,
        )
        return x


class VAETemporalEmulator(nn.Module):
    def __init__(self, micro_frame_size, kernel_size=4):
        super().__init__()
        self.micro_frame_size = micro_frame_size
        self.kernel_size = kernel_size

    def forward(self, x_z):
        """
        x_z: torch.Tensor: shape [B C T H W]
        """

        z_list = []
        for i in range(0, x_z.shape[2], self.micro_frame_size):
            x_z_bs = x_z[:, :, i : i + self.micro_frame_size]
            z_list.append(x_z_bs[:, :, 0:1])
            x_z_bs = x_z_bs[:, :, 1:]
            t_valid = x_z_bs.shape[2] - x_z_bs.shape[2] % self.kernel_size
            x_z_bs = x_z_bs[:, :, :t_valid]
            x_z_bs = reduce(x_z_bs, "B C (T n) H W -> B C T H W", n=self.kernel_size, reduction="mean")
            z_list.append(x_z_bs)
        z = torch.cat(z_list, dim=2)
        return z
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -