Re-basin via implicit Sinkhorn differentiation
Implementation of paper Re-basin via implicit Sinkhorn differentiation (Accepted at CVPR 2023).
Installation
pip install sinkhorn-rebasin
Running the examples
| Basics |  |
| Models alignment | |
| Linear mode connectivity | |
Models alignment
cd examples
python main_alignment_{mlp|cnn|resnet}.py
| Example | Layer from $\theta_A$ | Layer from $\pi_{\mathcal{P}}(\theta_A)$ | Layer from $\theta_B$ |
|---|---|---|---|
| MLP |  |
 |
 |
| VGG |  |
 |
 |
| ResNet18 |  |
 |
 |
Linear mode connectivity
cd examples
python main_lmc_{mlp|cnn|resnet}.py
| Dataset | Model | Accuracy LMC | Cross Entropy Loss LMC |
|---|---|---|---|
| Mnist | MLP |  |
 |
| Mnist | VGG |  |
 |
| Imagenette-320 | ResNet18 |  |
 |
1 maintainer
Metadata
1 maintainer
Release files for sinkhorn-rebasin 0.0.2
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| sinkhorn_rebasin-0.0.2-py3-none-any.whl | Python 3 | none | any | Details for sinkhorn_rebasin-0.0.2-py3-none-any.whl |
Release files / sinkhorn_rebasin-0.0.2-py3-none-any.whl
| Download URL | sinkhorn_rebasin-0.0.2-py3-none-any.whl |
|---|---|
| Size | 11.5 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
bffd8d6c15487a289745bfe89727a52fbc0c6f21ab07a4aebe89074897695983
|
|
BLAKE2b-256 checksum How to use checksums |
0541fdb4cb7162cf3bed3d046697018f4526bf12fe51c25f1a6ff3091e315fcb
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.9.12
|
Security for sinkhorn-rebasin 0.0.2
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.
