fixed-point-finder
fixed-point-finder copied to clipboard
mattgolub
FixedPointFinder: A Tensorflow toolbox for identifying and characterizing fixed points in recurrent neural networks
Hello, I recently came across your work with computing fixed points from RNNs. I am interested running this type of code for my PyTorch model, so I wanted to run...
Bumps [tensorflow](https://github.com/tensorflow/tensorflow) from 2.8.0 to 2.12.1. Release notes Sourced from tensorflow's releases. TensorFlow 2.12.1 Release 2.12.1 Bug Fixes and Other Changes The use of the ambe config to build and...
![dependabot[bot] avatar](https://avatars.githubusercontent.com/in/29110?v=4 "dependabot[bot] avatar"){kind=avatar}
Bumps [scikit-learn](https://github.com/scikit-learn/scikit-learn) from 1.2.2 to 1.5.0. Release notes Sourced from scikit-learn's releases. Scikit-learn 1.5.0 We're happy to announce the 1.5.0 release. You can read the release highlights under https://scikit-learn.org/stable/auto_examples/release_highlights/plot_release_highlights_1_5_0.html and...
Technically I ran into two issues: 1) installing via torch To address this, I modified `requirements-torch` so that the last line, `pytorch=1.12.1` is instead `torch==1.12.1` (note that this installs the...
pytorch isn't an installable package. Also, probably best to force users to decide if they want a CPU or GPU install. [Relevant github issue:How to add PyTorch to requirements.txt ](https://github.com/pytorch/pytorch/issues/29745)
While reviewing the code in FixedPointFinderTorch.py, I noticed that the line: `J_list = sequential_jacobian(x_bxd, inputs_bxd)` appears to be redundant for the following reasons: 1. Unused Result: The computed value of...
Hi Matt, Thank you for developing FixedPointFinder; it’s been instrumental in analyzing fixed points in my RNN. While PyTorch inherently supports GPU computation, I noticed that the optimization process within...