automlbenchmark VS mercury

Here is benchmark done by independent team of researchers https://openml.github.io/automlbenchmark/

I think most of overfitting is avoided with early stoppoing technique.

The underfitting can be avoidwd with using large training time.

Your list excludes most of well-known open-source AutoML tools such as auto-sklearn, AutoGluon, LightAutoML, MLJarSupervised, etc. These tools have been very extensively benchmarked by the OpenML AutoML Benchmark (https://github.com/openml/automlbenchmark) and have papers published, so they are pretty well-known to the AutoML community.

Regarding H2O.ai: Frankly, you don't seem to understand H2O.ai's AutoML offerings.

I'm the creator of H2O AutoML, which is open source, and there's no "enterprise version" of H2O AutoML. The interface is simple -- all you need to specify is the training data and target. We have included DNNs in our set of models since the first release of the tool in 2017. Read more here: https://docs.h2o.ai/h2o/latest-stable/h2o-docs/automl.html We also offer full explainability for our models: https://docs.h2o.ai/h2o/latest-stable/h2o-docs/explain.html

H2O.ai develops another AutoML tool called Driverless AI, which is proprietary. You might be conflating the two. Neither of these tools need to be used on the H2O AI Cloud. Both tools pre-date our cloud by many years and can be used on a user's own laptop/server very easily.

Your Features & Roadmap list in the README indicates that your tool does not yet offer DNNs, so either you should update your post here or update your README if it's incorrect: https://github.com/blobcity/autoai/blob/main/README.md#featu...

Lastly, I thought I would mention that there's already an AutoML tool called "AutoAI" by IBM. Generally, it's not a good idea to have name collisions in a small space like the AutoML community. https://www.ibm.com/support/producthub/icpdata/docs/content/...

I'm also curious how does it compare! The package will be included in the newest comparison done by OpenML people https://github.com/openml/automlbenchmark

I have some old comparison of closed-source old system

I'm build dashboards in Jupyter Lab. My plotting libraries are Altair, matplotlib, seaborn, Plotly - all work well in notebook.

My favorite is Altair. It provides interactivity for charts, so you can move/zoom your plots and have tooltips. It is much lighter than Plotly after saving the notebook to ipynb file. Altair charts looks much better than in matplotlib. One drawback, that exporting to PDF doesn't work. To serve notebook as dashboard with code hidden, I use Mercury framework, you can check example https://runmercury.com/tutorials/vega-altair-dashboard/

disclaimer: I'm author of Mercury framework https://github.com/mljar/mercury

Web App is using two open-source packages that I've created:

- MLJAR AutoML - Python package for AutoML on tabular data https://github.com/mljar/mljar-supervised

- Mercury - framework for converting Jupyter Notebooks into Web App https://github.com/mljar/mercury

You can run Web App locally. What is more, you can adjust notebook's code for your needs. For example, you can set different validation strategies or evalutaion metrics or longer training times. The notebooks in the repo are good starting point for you to develop more advanced apps.

The GitHub repository https://github.com/mljar/mercury

We are working on open-source web framework Mercury that converts Python notebooks to Web Apps.

It is very opinionated:

- it has no callbacks - we automatically re-execute cells below updated widget

- it has no layout widgets, all input widgets are always in the left sidebar

Thanks to above decisions you don't need to change notebook's code to have web app and fit to the framework.

The simplicity of the framework is very important to us. We also care about deployment simplicity. That's why we created a shared hosting service called Mercury Cloud. You can deploy notebook by uploading a file.

The GitHub repository https://github.com/mljar/mercury

Documentation https://RunMercury.com/docs/

Mercury Cloud https://cloud.runmercury.com

Analysis was done in Jupyter Notebook with Python 3.10, Pandas, Matplotlib, wordcloud and Mercury framework.

Analysis was done in Jupyter Notebook with Python 3.10 kernel, Pandas, Matplotlib, wordcloud and Mercury framework to share notebook as a web application with widgets and code hidden. Gif created in Canva.