Train a GAN using the techniques described in the paper "Training Generative Adversarial Networks with Limited Data". Examples: \b #

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

r"""Class decorator that extends a given class to save its source code when pickled. Example: from torch_utils import persistence

r"""Broadcasts the given set of scalars to all interested instances of `Collector`, across device and process boundaries. This function is ex

Project given image to the latent space of pretrained network pickle. Examples: \b python projector.py --outdir=out --target=~/mytargeti

r"""Convenience getter. `collector[name]` is a synonym for `collector.mean(name)`.

r"""Hook that is called internally by the `pickle` module to unpickle a persistent object.

Ask the user the question until the user inputs a valid answer.

r"""Fused bias and activation function. Adds bias `b` to activation tensor `x`, evaluates activation function `act`, and scales the result by

Finds the python class with the given name and constructs it with the given arguments.

r"""2D convolution with optional up/downsampling. Padding is performed only once at the beginning, not between the operations. Args:

Takes in a list of tuples of (src, dst) paths and copies files. Will create all necessary directories.

r"""Downsample a batch of 2D images using the given 2D FIR filter. By default, the result is padded so that its shape is a fraction of the input.