Extract values from a 1-D numpy array for a batch of indices. :param arr: the 1-D numpy array. :param timesteps: a tensor of indices i

r"""Returns the mean of the scalars that were accumulated for the given statistic between the last two calls to `update()`, or NaN if

r""" Copy from torch.nn.utils.clip_grad_norm_ Clips gradient norm of an iterable of parameters. The norm is computed over all gradients

r"""Copies current values of the internal counters to the user-visible state and resets them for the next round. If `keep_previous=Tr

Step the EMA model towards the current model.

Wrapper for the underlying `conv2d()` and `conv_transpose2d()` implementations.

Take the mean over all non-batch dimensions.

Importance-sample timesteps for a batch. :param batch_size: the number of timesteps. :param device: the torch device to sa

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

r"""Returns the standard deviation of the scalars that were accumulated for the given statistic between the last two calls to `update(

End DDP training.

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

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

Apply the model to get p(x_{t-1} | x_t), as well as a prediction of the initial x, x_0. :param model: the model, which tak

Compute the mean and variance of the diffusion posterior: q(x_{t-1} | x_t, x_0)

Set requires_grad flag for all parameters in a model.

write the loss information to a tensorboard file. Only for pytorch DDP mode.

r"""Returns the raw moments that were accumulated for the given statistic between the last two calls to `update()`, or zero if no scal

Get a term for the variational lower-bound. The resulting units are bits (rather than nats, as one might expect). This all

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

Traverses the object name and returns the last (rightmost) python object.

r"""Pad, upsample, filter, and downsample a batch of 2D images. Performs the following sequence of operations for each channel: 1. Upsample

Fast CUDA implementation of `upfirdn2d()` using custom ops.

A fast approximation of the cumulative distribution function of the standard normal.

Compute what the p_mean_variance output would have been, should the model's score function be conditioned by cond_fn. See

Create a tensorboard that saves losses.

Generate samples from the model using DDIM. Same usage as p_sample_loop().

Finds a pre-trained Latte model, downloading it if necessary. Alternatively, loads a model from a local path.

Flush written text to both stdout and a file, if open.

Searches for the underlying module behind the name to some python object. Returns the module and the object name (original name with module part r

r"""Test whether the given object or class is persistent, i.e., whether it will save its source code when pickled.

r"""Returns the names of all statistics broadcasted so far that match the regular expression specified at construction time.

Compute the KL divergence between two gaussians. Shapes are automatically broadcasted, so batches can be compared to scalars, among ot

Generate samples from the model. :param model: the model module. :param shape: the shape of the samples, (N, C, H, W).

Diffuse the data for a given number of diffusion steps. In other words, sample from q(x_t | x_0). :param x_start: the init

Initialize distributed training environment. support both slurm and torch.distributed.launch see torch.distributed.init_process_group() for mo

Write text to stdout (and a file) and optionally flush.