Update the state with extracted features. Args: imgs: tensor with images feed to the feature extractor real: bool ind

Compute the noise coefficient along the path

Simple block wrapping a mixer class with LayerNorm/RMSNorm and residual connection" This Block has a slightly different structure co

Function to reshape time t to broadcastable dimension of x Args: t: [batch_dim,], time vector x: [batch_dim,...], data point

x: (batch, dim, seqlen) weight: (dim, width) bias: (dim,) activation: either None or "silu" or "swish" out: (batch, dim, seqlen)

Update the state with extracted features.

We always output sde according to score parametrization;

Args: clip (torch.tensor): Video clip to be cropped. Size is (T, C, H, W)

if return_last_state is True, returns (out, last_state) last_state has shape (batch, dim, dstate). Note that the gradient of the last state is

Compute the data coefficient along the path

Calculate FID score based on accumulated extracted features from the two distributions.

Decoding, either greedy or with top-k or top-p sampling. If top-k = 0, don't limit the number of candidates (pure sampling). Top-k and top-p c

Take the mean over all non-batch dimensions.

Adapted from `KID Score`_

forward loop of sde

Sampling x0 & t based on shape of x1 (if needed) Args: x1 - data point; [batch, *dim]

Converts a PIL Image to an input for the model

Initializes Unit3D module.

Args: input_features: numpy.ndarray([N, feature_dim], dtype=np.float32) nearest_k: int Returns: Distances to kth near

Args: data_x: numpy.ndarray([N, feature_dim], dtype=np.float32) data_y: numpy.ndarray([N, feature_dim], dtype=np.float32) Ret

Estimate a covariance matrix given data. Covariance indicates the level to which two variables vary together. If we examine N-dimensional sam

Create a logger that writes to a log file and stdout.

function for creating Transport object **Note**: model prediction defaults to velocity Args: - path_type: type of path to use; default to

embed_dim: output dimension for each position pos: a list of positions to be encoded: size (M,) out: (M, D)

grid_size: int of the grid height and width return: pos_embed: [grid_size*grid_size, embed_dim] or [1+grid_size*grid_size, embed_dim] (w/

Set the logits for none top-p values to -inf. Done in-place.

Plot a single or multiple values from the metric. Args: val: Either a single result from calling `metric.forward` or `metric.comp

Set requires_grad flag for all parameters in a model.

returns a sampling function with given ODE settings Args: - sampling_method: type of sampler used in solving the ODE; default to be Do

Argument: state: (batch, dim, dstate) x: (batch, dim) dt: (batch, dim) A: (dim, dstate) B: (batch, dstate

Update the state with extracted features.

Update the state with extracted features.

Step the EMA model towards the current model.

TODO: generalize here by adding all private functions ending with steps to it

Get the last step function of the SDE solver