wassname/viz_torch_optim
Videos of deep learning optimizers moving on 3D problem-landscapes
This tool helps deep learning researchers and practitioners understand how different optimization algorithms navigate complex problem landscapes. It takes a chosen PyTorch optimizer and a toy problem function, then generates a video animation showing the optimizer's path across a 3D surface. This is designed for those who want to visualize and compare the behavior of optimizers like Adam and SGD.
107 stars. No commits in the last 6 months.
Use this if you are a deep learning researcher or student who wants clear visual explanations of how various optimization algorithms find solutions, especially on challenging, noisy surfaces.
Not ideal if you are looking to visualize optimizer performance on your own custom, high-dimensional neural network models, as this is designed for specific toy problems.
Stars
107
Forks
13
Language
Jupyter Notebook
License
—
Category
Last pushed
Jul 25, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/wassname/viz_torch_optim"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
nschaetti/EchoTorch
A Python toolkit for Reservoir Computing and Echo State Network experimentation based on...
metaopt/torchopt
TorchOpt is an efficient library for differentiable optimization built upon PyTorch.
opthub-org/pytorch-bsf
PyTorch implementation of Bezier simplex fitting
gpauloski/kfac-pytorch
Distributed K-FAC preconditioner for PyTorch
pytorch/xla
Enabling PyTorch on XLA Devices (e.g. Google TPU)