Jesper Wohlert wohlert
wohlert
/ baselayer.py
Last active
May 24, 2019 15:45
A metaprogramming approach with minimal overhead to perform Jacobian computation for arbitrary networks
View baselayer.py
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| """ | |
| Registers jacobian for basic PyTorch layers | |
| """ | |
| import torch | |
| from torch import nn | |
| from .metric import register_jacobian | |
| @register_jacobian(nn.Linear) |
View Sinkhorn-solver.ipynb
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
View hmc.py
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| import numpy as np | |
| class HamiltonianMC(object): | |
| """ | |
| Hamiltonian/hybrid Monte Carlo | |
| Uses Hamiltonian dynamics to make proposals using | |
| kinetic energy of the proposal. Employs leapfrog | |
| integration for time-reversibility and to ensure |
wohlert
/ Backpropagation.ipynb
Last active
October 30, 2018 15:58
Tutorial describing backpropagation in a MLP
View Backpropagation.ipynb
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.