Method __init__

diffusers/src/diffusers/models/autoencoders/vae.py:919–958

(
        self,
        in_channels: int,
        out_channels: int,
        num_blocks: Tuple[int, ...],
        block_out_channels: Tuple[int, ...],
        upsampling_scaling_factor: int,
        act_fn: str,
        upsample_fn: str,
    )

    """

    def __init__(
        self,
        in_channels: int,
        out_channels: int,
        num_blocks: Tuple[int, ...],
        block_out_channels: Tuple[int, ...],
        upsampling_scaling_factor: int,
        act_fn: str,
        upsample_fn: str,
    ):
        super().__init__()

        layers = [
            nn.Conv2d(in_channels, block_out_channels[0], kernel_size=3, padding=1),
            get_activation(act_fn),
        ]

        for i, num_block in enumerate(num_blocks):
            is_final_block = i == (len(num_blocks) - 1)
            num_channels = block_out_channels[i]

            for _ in range(num_block):
                layers.append(AutoencoderTinyBlock(num_channels, num_channels, act_fn))

            if not is_final_block:
                layers.append(nn.Upsample(scale_factor=upsampling_scaling_factor, mode=upsample_fn))

            conv_out_channel = num_channels if not is_final_block else out_channels
            layers.append(
                nn.Conv2d(
                    num_channels,
                    conv_out_channel,
                    kernel_size=3,
                    padding=1,
                    bias=is_final_block,
                )
            )

        self.layers = nn.Sequential(*layers)
        self.gradient_checkpointing = False

    def forward(self, x: torch.Tensor) -> torch.Tensor:
        r"""The forward method of the `DecoderTiny` class."""