Load a PyTorch file without redundant fetches across MPI ranks.

The same as logkv(), but if called many times, values averaged.

Log a dictionary of key-value pairs

Convert rotations given as rotation matrices to axis/angle. Args: matrix: Rotation matrices as tensor of shape (..., 3, 3). Ret

Converts rotation matrices to 6D rotation representation by Zhou et al. [1] by dropping the last row. Note that 6D representation is not uniq

Make a standard normalization layer. :param channels: number of input channels. :return: an nn.Module for normalization.

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).

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 using Pseudo Linear Multistep. Same usage as p_sample_loop().

Usage: @profile("my_func") def my_func(): code

q0: starting quaternion q1: ending quaternion t: array of points along the way Returns: Tensor of Slerps: t.shape + q0.shape

Apply the rotation given by a quaternion to a 3D point. Usual torch rules for broadcasting apply. Args: quaternion: Tensor of qu

Multiply two quaternions representing rotations, returning the quaternion representing their composition, i.e. the versor with nonnegative re

Generate a single random 3x3 rotation matrix. Args: dtype: Type to return device: Device of returned tensor. Default: if Non

Scale the parameters of a module and return it.

Set logging threshold on current logger.

Setup a distributed process group.

Synchronize a sequence of Tensors across ranks from rank 0.

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

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

Update target parameters to be closer to those of source parameters using an exponential moving average. :param target_params: the targe

Zero out the parameters of a module and return it.