Wraps `score_fn` so that the model output corresponds to a real time-dependent score function. Args: sde: An `sde_lib.SDE` object that represen

A piecewise linear function y = f(x), using xp and yp as keypoints. We implement f(x) in a differentiable way (i.e. applicable for autograd).

A piecewise linear function y = f(x), using xp and yp as keypoints. We implement f(x) in a differentiable way (i.e. applicable for autograd).

A piecewise linear function y = f(x), using xp and yp as keypoints. We implement f(x) in a differentiable way (i.e. applicable for autograd).

A piecewise linear function y = f(x), using xp and yp as keypoints. We implement f(x) in a differentiable way (i.e. applicable for autograd).

A piecewise linear function y = f(x), using xp and yp as keypoints. We implement f(x) in a differentiable way (i.e. applicable for autograd).

A piecewise linear function y = f(x), using xp and yp as keypoints. We implement f(x) in a differentiable way (i.e. applicable for autograd).

Compute the loss function. Args: model: A score model. batch: A mini-batch of training data. Returns: loss: A scalar that

Copy model parameters into a (differently-shaped) list of full-precision parameters.

Python2 support for os.makedirs(.., exist_ok=True)

The noise predicition model function that is used for DPM-Solver.

Create a wrapper function for the noise prediction model. DPM-Solver needs to solve the continuous-time diffusion ODEs. For DPMs trained on discr

Multistep DPM-Solver with the order `order` from time `t_prev_list[-1]` to time `t`. Args: x: A jnp.DeviceArray. The ini

Multistep DPM-Solver with the order `order` from time `t_prev_list[-1]` to time `t`. Args: x: A pytorch tensor. The init

Multistep DPM-Solver with the order `order` from time `t_prev_list[-1]` to time `t`. Args: x: A pytorch tensor. The init

Multistep DPM-Solver with the order `order` from time `t_prev_list[-1]` to time `t`. Args: x: A pytorch tensor. The init

Multistep DPM-Solver with the order `order` from time `t_prev_list[-1]` to time `t`. Args: x: A pytorch tensor. The init

Multistep DPM-Solver with the order `order` from time `t_prev_list[-1]` to time `t`. Args: x: A jnp.DeviceArray. The ini

1x1 convolution with PyTorch initialization. Same as NCSNv1/v2.

Return the noise prediction model.

Return the noise prediction model.

Return the noise prediction model.

Return the noise prediction model.

Return the noise prediction model.

Return the noise prediction model.

Convert [0, 1] to [0.001, 1].

Shrink an image to the given resolution.

Shrink an image to the given resolution.

Restore the parameters stored with the `store` method. Useful to validate the model with EMA parameters without affecting the original op

Restore the parameters stored with the `store` method. Useful to validate the model with EMA parameters without affecting the

Compute the sample at time `t_end` by DPM-Solver, given the initial `x` at time `t_start`. =========================================

Compute the sample at time `t_end` by DPM-Solver, given the initial `x` at time `t_start`. =========================================

Compute the sample at time `t_end` by DPM-Solver, given the initial `x` at time `t_start`. =========================================

Make a grid of images and save it into an image file. Pixel values are assumed to be within [0, 1]. Args: ndarray (array_like): 4D mini-batc

shift pixel for super-resolution with different scale factors Args: x: WxHxC or WxH sf: scale factor upper_left: shift dir

shift pixel for super-resolution with different scale factors Args: x: WxHxC or WxH sf: scale factor upper_left: shift dir

Singlestep solver DPM-Solver-3 from time `s` to time `t`. Args: x: A jnp.DeviceArray. The initial value at time `s`.

Singlestep solver DPM-Solver-3 from time `s` to time `t`. Args: x: A pytorch tensor. The initial value at time `s`.

Singlestep solver DPM-Solver-3 from time `s` to time `t`. Args: x: A pytorch tensor. The initial value at time `s`.

Singlestep solver DPM-Solver-3 from time `s` to time `t`. Args: x: A pytorch tensor. The initial value at time `s`.

Singlestep solver DPM-Solver-3 from time `s` to time `t`. Args: x: A pytorch tensor. The initial value at time `s`.

Singlestep solver DPM-Solver-3 from time `s` to time `t`. Args: x: A jnp.DeviceArray. The initial value at time `s`.

Save the current parameters for restoring later. Args: parameters: Iterable of `torch.nn.Parameter`; the parameters to be temp

Save the current parameters for restoring later. Args: parameters: Iterable of `torch.nn.Parameter`; the parameters to be

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

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

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

Flatten a torch tensor `x` and convert it to numpy.

Flatten a JAX array `x` and convert it to numpy.

Add two tensors, padding them with zeros or pooling them if necessary. Args: block_x: Output of a resnet block. orig_x: Residual branch to