https://github.com/andreaschandra/ax-playground
Adaptive Experimentation Platform - Playground
https://github.com/andreaschandra/ax-playground
Last synced: 8 months ago
JSON representation
Adaptive Experimentation Platform - Playground
- Host: GitHub
- URL: https://github.com/andreaschandra/ax-playground
- Owner: andreaschandra
- License: mit
- Created: 2021-01-25T07:16:48.000Z (over 5 years ago)
- Default Branch: main
- Last Pushed: 2021-02-01T08:09:44.000Z (over 5 years ago)
- Last Synced: 2025-02-12T09:49:51.101Z (over 1 year ago)
- Language: Jupyter Notebook
- Size: 3.82 MB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# ax-playground
Adaptive Experimentation Platform - Playground
Ax is a platform for optimizing any kind of experiment, including machine learning experiments, A/B tests, and simulations. Ax can optimize discrete configurations (e.g., variants of an A/B test) using multi-armed bandit optimization, and continuous (e.g., integer or floating point)-valued configurations using Bayesian optimization. This makes it suitable for a wide range of applications.
Ax has been successfully applied to a variety of product, infrastructure, ML, and research applications at Facebook.
_https://ax.dev/docs/why-ax.html_
## Installation
The library strongly encourage you to install via pip and conda for OSX
```
conda install pytorch torchvision -c pytorch # OSX only (details below)
pip3 install ax-platform
```
## Highlights
Tutorial References:
- https://ax.dev/tutorials/tune_cnn.html
- https://towardsdatascience.com/quick-tutorial-using-bayesian-optimization-to-tune-your-hyperparameters-in-pytorch-e9f74fc133c2
- https://www.justintodata.com/hyperparameter-tuning-with-python-keras-guide/
Key points:
- Well documented (better) than BoTorch
- There are 3 types of usage in order to tune your parameters: Loop API, Service API, Developer API.
- There are 2 Algorithms: Bayesian and Bandit Optimization
- built-in feature that enables saving results to a JSON file or a MySQL database.
- able to create such as complext dependent parameter constraints.
- Integrate visualization using plotly and stunning visualization.