• Stars
    star
    20
  • Rank 1,121,974 (Top 23 %)
  • Language
    TeX
  • Created over 1 year ago
  • Updated over 1 year ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

Source code for my PhD thesis: Backpropagation Beyond the Gradient

More Repositories

1

backpack

BackPACK - a backpropagation package built on top of PyTorch which efficiently computes quantities other than the gradient.
Python
555
star
2

cockpit

Cockpit: A Practical Debugging Tool for Training Deep Neural Networks
Python
474
star
3

unfoldNd

(N=1,2,3)-dimensional unfold (im2col) and fold (col2im) in PyTorch
Python
83
star
4

hbp

Hessian backpropagation (HBP): PyTorch extension of backpropagation for block-diagonal curvature matrix approximations
Python
20
star
5

singd

[ICML 2024] SINGD: KFAC-like Structured Inverse-Free Natural Gradient Descent (http://arxiv.org/abs/2312.05705)
Python
19
star
6

curvlinops

scipy linear operators for the Hessian, Fisher/GGN, and more in PyTorch
Python
17
star
7

vivit

[TMLR 2022] Curvature access through the generalized Gauss-Newton's low-rank structure: Eigenvalues, eigenvectors, directional derivatives & Newton steps
Python
17
star
8

einconv

Convolutions and more as einsum for PyTorch
Python
12
star
9

sirfshampoo

[ICML 2024] SIRFShampoo: Structured inverse- and root-free Shampoo in PyTorch (https://arxiv.org/abs/2402.03496)
Python
10
star
10

phd-thesis-template

LaTeX template for my PhD thesis at the University of Tuebingen
TeX
9
star
11

backobs

Use DeepOBS with BackPACK
Python
2
star
12

org-export-setup

My org-export settings
TeX
2
star
13

python-utilities

Python utility functions I often use
Python
2
star
14

wandb_preempt

Code and tutorial on integrating wandb sweeps with Slurm pre-emption
Python
1
star
15

vivit-experiments

Experiments for the TMLR 2023 paper "ViViT: Curvature Access Through the Generalized Gauss-Newton’s Low-rank Structure"
Python
1
star