https://github.com/greentfrapp/lucent

Lucid library adapted for PyTorch
https://github.com/greentfrapp/lucent

Last synced: 3 months ago
JSON representation

Lucid library adapted for PyTorch

• Host: GitHub
• URL: https://github.com/greentfrapp/lucent
• Owner: greentfrapp
• License: apache-2.0
• Created: 2020-05-09T18:07:01.000Z (almost 5 years ago)
• Default Branch: dev
• Last Pushed: 2024-06-08T15:39:07.000Z (9 months ago)
• Last Synced: 2024-11-05T14:24:09.977Z (4 months ago)
• Language: Python
• Size: 38.6 MB
• Stars: 605
• Watchers: 16
• Forks: 88
• Open Issues: 19
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• Awesome-pytorch-list-CNVersion - Lucent
• Awesome-pytorch-list - Lucent

README

        
![](https://github.com/greentfrapp/lucent/raw/master/images/lucent_header.jpg)

# Lucent

[![Travis build status](https://img.shields.io/travis/greentfrapp/lucent.svg)](https://travis-ci.org/greentfrapp/lucent)

[![Code coverage](https://img.shields.io/coveralls/github/greentfrapp/lucent.svg)](https://coveralls.io/github/greentfrapp/lucent)

*PyTorch + Lucid = Lucent*

The wonderful [Lucid](https://github.com/tensorflow/lucid) library adapted for the wonderful PyTorch!

**Lucent is not affiliated with Lucid or OpenAI's Clarity team, although we would love to be!**

Credit is due to the original Lucid authors, we merely adapted the code for PyTorch and we take the blame for all issues and bugs found here.

# Usage

Lucent is still in pre-alpha phase and can be installed locally with the following command:

```

pip install torch-lucent

```

In the spirit of Lucid, get up and running with Lucent immediately, thanks to Google's [Colab](https://colab.research.google.com/notebooks/welcome.ipynb)! 

You can also clone this repository and run the notebooks locally with [Jupyter](http://jupyter.org/install.html).

## Quickstart

```python

import torch

from lucent.optvis import render

from lucent.modelzoo import inceptionv1

device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")

model = inceptionv1(pretrained=True)

model.to(device).eval()

render.render_vis(model, "mixed4a:476")

```

## Tutorials







## Other Notebooks

Here, we have tried to recreate some of the Lucid notebooks! You can also check out the [lucent-notebooks](https://github.com/greentfrapp/lucent-notebooks) repo to clone all the notebooks.























# Recommended Readings

* [Feature Visualization](https://distill.pub/2017/feature-visualization/)

* [The Building Blocks of Interpretability](https://distill.pub/2018/building-blocks/)

* [Using Artificial Intelligence to Augment Human Intelligence](https://distill.pub/2017/aia/)

* [Visualizing Representations: Deep Learning and Human Beings](http://colah.github.io/posts/2015-01-Visualizing-Representations/)

* [Differentiable Image Parameterizations](https://distill.pub/2018/differentiable-parameterizations/)

* [Activation Atlas](https://distill.pub/2019/activation-atlas/)

## Related Talks

* [Lessons from a year of Distill ML Research](https://www.youtube.com/watch?v=jlZsgUZaIyY) (Shan Carter, OpenVisConf)

* [Machine Learning for Visualization](https://www.youtube.com/watch?v=6n-kCYn0zxU) (Ian Johnson, OpenVisConf)

# Slack

Check out `#proj-lucid` and `#circuits` on the [Distill slack](http://slack.distill.pub)!

# Additional Information

## License and Disclaimer

You may use this software under the Apache 2.0 License. See [LICENSE](https://github.com/greentfrapp/lucent/blob/master/LICENSE).