https://github.com/omarrayyann/KAN-Conv2D

Drop-in convolutional KAN replacement of Conv2d
https://github.com/omarrayyann/KAN-Conv2D

Last synced: about 1 month ago
JSON representation

Drop-in convolutional KAN replacement of Conv2d

• Host: GitHub
• URL: https://github.com/omarrayyann/KAN-Conv2D
• Owner: omarrayyann
• License: apache-2.0
• Created: 2024-05-20T00:50:12.000Z (8 months ago)
• Default Branch: main
• Last Pushed: 2024-05-22T07:04:36.000Z (8 months ago)
• Last Synced: 2024-05-23T02:16:43.151Z (8 months ago)
• Language: Python
• Size: 153 KB
• Stars: 3
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• awesome-kan - KAN-Conv2D - in Convolutional KAN built on multiple implementations ([Original pykan](https://github.com/KindXiaoming/pykan) / [efficient-kan](https://github.com/Blealtan/efficient-kan) / [FastKAN](https://github.com/ZiyaoLi/fast-kan)) to support the original paper hyperparameters. |  (Library / ConvKANs)
• awesome-kan - KAN-Conv2D - in Convolutional KAN built on multiple implementations ([Original pykan](https://github.com/KindXiaoming/pykan) / [efficient-kan](https://github.com/Blealtan/efficient-kan) / [FastKAN](https://github.com/ZiyaoLi/fast-kan)) to support the original paper hyperparameters. |  (Library / ConvKANs)

README

        
# Convolutional 2D KAN Implementation

This repositry contains 3 drop-in convolutional KAN replacements. Each work on top of a different KAN implementation:

1. [Efficient implementation of Kolmogorov-Arnold Network (KAN)](https://github.com/Blealtan/efficient-kan)

2. [Original KAN implementation](https://github.com/KindXiaoming/pykan)

3. [Fast KAN implementation](https://github.com/ZiyaoLi/fast-kan)

# Installation

```bash

git clone [email protected]/omarrayyann/KAN-Conv2D

cd KAN-Conv2D

pip install -r requirements.txt

```

# Usage

You should be able to just replace ```torch.nn.Conv2D()``` with ```ConvKAN()```

```python3

from ConvKAN import ConvKAN

# Implementation built on the efficient KAN Implementation (https://github.com/Blealtan/efficient-kan)

conv = ConvKAN(in_channels=3, out_channels=4, kernel_size=3, stride=1, padding=1, version="Efficient")

# Implementation built on the original KAN Implementation (https://github.com/KindXiaoming/pykan)

conv = ConvKAN(in_channels=3, out_channels=4, kernel_size=3, stride=1, padding=1, version="Original")

# Implementation built on the fast KAN Implementation (https://github.com/ZiyaoLi/fast-kan)

conv = ConvKAN(in_channels=3, out_channels=4, kernel_size=3, stride=1, padding=1, version="Fast")

```