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 into

Take the mean over all non-batch dimensions.

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 takes

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

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

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 con

decode index array, x, to byte string of this alphabet

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

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

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

Get the prior KL term for the variational lower-bound, measured in bits-per-dim. This term can't be optimized, as it only dep

Create a beta schedule that discretizes the given alpha_t_bar function, which defines the cumulative product of (1-beta) over time from t = [

Compute the mean for the previous step, given a function cond_fn that computes the gradient of a conditional log probability with res

Sample x_{t-1} from the model using DDIM. Same usage as p_sample().

Use DDIM to sample from the model and yield intermediate samples from each timestep of DDIM. Same usage as p_sample_loop_prog

Compute the log-likelihood of a Gaussian distribution discretizing to a given image. :param x: the target images. It is assumed that this

Forward pass of DiT. x: (N, C, H, W) tensor of spatial inputs (images or latent representations of images) t: (N,) tensor of

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/

This is the deprecated API for creating beta schedules. See get_named_beta_schedule() for the new library of schedules.

Sample x_{t-1} from the model at the given timestep. :param model: the model to sample from. :param x: the current tensor at

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

Generate samples from the model and yield intermediate samples from each timestep of diffusion. Arguments are the same as p_s

Get the distribution q(x_t | x_0). :param x_start: the [N x C x ...] tensor of noiseless inputs. :param t: the number of diff

Create a list of timesteps to use from an original diffusion process, given the number of timesteps we want to take from equally-sized portio

Create sinusoidal timestep embeddings. :param t: a 1-D Tensor of N indices, one per batch element. These ma

Drops labels to enable classifier-free guidance.

Update the reweighting using losses from a model. Sub-classes should override this method to update the reweighting using los

Get a numpy array of weights, one per diffusion step. The weights needn't be normalized, but must be positive.

Compute the entire variational lower-bound, measured in bits-per-dim, as well as other related quantities. :param model: the

Compute the log-likelihood of a continuous Gaussian distribution. :param x: the targets :param means: the Gaussian mean Tensor. :para

Create a ScheduleSampler from a library of pre-defined samplers. :param name: the name of the sampler. :param diffusion: the diffusion ob

Sample x_{t+1} from the model using DDIM reverse ODE.

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

encode a byte string into alphabet indices

Forward pass of DiT, but also batches the unconditional forward pass for classifier-free guidance.

Get a pre-defined beta schedule for the given name. The beta schedule library consists of beta schedules which remain similar in the limi

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

Compute training losses for a single timestep. :param model: the model to evaluate loss on. :param x_start: the [N x C x ...]

Update the reweighting using losses from a model. Call this method from each rank with a batch of timesteps and the correspon