realnvp_imagenet32_gen(pretrained: bool = False, overrides: object = {})RealNVP generator for Imagenet 32 x 32.
Same bijection as realnvp_imagenet32, wrapped with the
bits/dim objective and the ancestral sampler.
Model Size
Parameters
pretrainedbool= FalseReserved for future weight registration; currently a no-op.
**overridesobject= {}Optional
RealNVPConfig field overrides forwarded into
the underlying config.Returns
RealNVPForImageGenerationGenerator configured with the Imagenet 32 x 32 setup.
Notes
Reference: Dinh, Sohl-Dickstein, and Bengio, "Density Estimation Using Real NVP", ICLR, 2017 (arXiv:1605.08803). Reported: 4.28 bits/dim.
Examples
>>> from lucid.models.generative.realnvp import realnvp_imagenet32_gen
>>> model = realnvp_imagenet32_gen().eval()
>>> model.generate(n_samples=2).samples.shape
(2, 3, 32, 32)