ltx23_video_encoder3d

Packages

LTX-2.3 video encoder

Description

Pixel video [B, 3, F, H, W] -> latent statistics [B, 2 * latent_channels, F/8, H/32, W/32] (mean and a uniform log-var channel broadcast across the latent channels).

Usage

ltx23_video_encoder3d(in_channels = 3L, out_channels = 128L,
                      block_out_channels = c(256L, 512L, 1024L, 1024L),
                      spatio_temporal_scaling = c(TRUE, TRUE, TRUE, TRUE),
                      layers_per_block = c(4L, 6L, 4L, 2L, 2L),
                      downsample_type = c("spatial", "temporal", "spatiotemporal", "spatiotemporal"),
                      patch_size = 4L, patch_size_t = 1L,
                      resnet_norm_eps = 1e-06, is_causal = TRUE,
                      spatial_padding_mode = "zeros")

Arguments

block_out_channels: Integer vector. Per-block output channels.
spatio_temporal_scaling: Logical vector per block.
layers_per_block: Integer vector (blocks then mid).
downsample_type: Character vector per block.
resnet_norm_eps: Numeric.
is_causal: Logical.
spatial_padding_mode: Character.
in_channels,out_channels: Integers. Pixel and latent channels.
patch_size,patch_size_t: Integers. Pixel patchification.