https://github.com/fabiangroeger96/wandb-callbacks

Additional Callbacks for Weights & Biases to monitor your models even better :mag_right:
https://github.com/fabiangroeger96/wandb-callbacks

callback experiment-track hyperparameter-search keras machine-learning tensorflow weights-and-biases

Last synced: 15 days ago
JSON representation

Additional Callbacks for Weights & Biases to monitor your models even better :mag_right:

• Host: GitHub
• URL: https://github.com/fabiangroeger96/wandb-callbacks
• Owner: FabianGroeger96
• License: mit
• Created: 2021-04-26T16:12:25.000Z (over 3 years ago)
• Default Branch: main
• Last Pushed: 2021-06-02T16:36:42.000Z (over 3 years ago)
• Last Synced: 2024-10-04T21:04:07.446Z (about 1 month ago)
• Topics: callback, experiment-track, hyperparameter-search, keras, machine-learning, tensorflow, weights-and-biases
• Language: Python
• Homepage: https://fabiangroeger96.github.io/wandb-callbacks/
• Size: 5.79 MB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • Funding: .github/FUNDING.yml
  • License: LICENSE

Awesome Lists containing this project

README

        




 

[![PyPI version](https://badge.fury.io/py/wandb-callbacks.svg)](https://badge.fury.io/py/wandb-callbacks)

[![GitHub Issues](https://img.shields.io/github/issues/FabianGroeger96/wandb-callbacks)](https://img.shields.io/github/issues/FabianGroeger96/wandb-callbacks)

[![License](https://img.shields.io/github/license/FabianGroeger96/wandb-callbacks)](https://img.shields.io/github/license/FabianGroeger96/wandb-callbacks)

![Contribotion](https://img.shields.io/badge/Contribution-Welcome-brightgreen)






# Weights & Biases Callbacks

`wandb-callbacks` provides some additional Callbacks for Weights & Biases.

Callbacks currently implemented:

* `ActivationCallback`

  * visualizes the activations of a layer

  * activations are computed for a sample of each class

* `DeadReluCallback`

  * logs the number of dead relus in each layer

  * prints warning if the percentage is higher than a threshold

* `GradCAMCallback`

  *  [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://arxiv.org/pdf/1610.02391.pdf)

  *  produces a coarse localization map highlighting the important regions in the image for predicting the class of the image

## Installation

### Last Stable Release

```python

pip install wandb-callbacks

```

### Latest Development Changes

```bash

git clone https://github.com/FabianGroeger96/wandb-callbacks

```

## Sample Implementation

Can be found in `notebooks/sample_implementation.ipynb`.

## Contributing

Open to ideas and for helpers to develop the package further.