Tutorials

Modified on Thu, 17 Aug 2023 at 08:43 AM

Welcome to Monolith AI! You now have access to the Monolith AI platform. If you have not already registered, please refer to the article platform registration to learn how you can create your account.

Getting Started

  • Log-in to the Monolith AI platform
  • Click on the Getting Started tab in the top left corner
  • All of the tutorials are available on this page.

Building a Notebook

We recommend the "Building a Notebook" section as a quick start. These tutorials guide you step by step through importing data into a notebook, exploring your dataset, training a model and applying that model to predict the performance of a design, or optimise your design.

By the end of these tutorials you will be familiar with the interface and manipulators used in the platform, be able to find key tools, and will have built your first workflow on tabular data within a notebook in Monolith.

Which tutorials you move onto next, depends on the type of data with which you want to work. If you are using tables or spreadsheets of data, then continue onto the “Using AI models” section of tutorials. If you are using 3D data and have CAD geometry or surface field data, then move to the “AI for 3D Data” section.

Using AI models

In this section, we delve a little deeper into rigorously comparing the performance of different AI regression models for a given problem using tabular data. We also highlight how to automatically tune the parameters of different models to output the best model for your use case. In this section, we also share how to set up a problem for multi-physics optimisation.

AI for 3D Data

These tutorials outline how to set up and train an autoencoder for parameterising geometry in Monolith. These tutorials also cover how to train a surface field model to predict performance metrics about new geometries and finally how to link these two types of model together to be able to optimise a 3D design for a desired performance.

All about the data

To really get the best out of your use case, we recommend following the “All about the data” tutorials. We recognise that customers very rarely have access to ‘perfect’ data for machine learning, and these tutorials cover different issues you may not yet realise you have with your data and how to spot those issues by exploring your data in Monolith.

Challenges

At the bottom of the tutorials page, you can find two freeform challenges. This section allows you to test your knowledge; each challenge gives you a premise but no further instructions on what steps to apply to the dataset.

We recommend the “Composite Materials challenge” as one to try if you are working with categorical variables. If you have time series data, take a look at the “Track Dynamics challenge”.

What next?

To learn more on how to move forward with your own data, click here.

Was this article helpful?

That’s Great!

Thank you for your feedback

Sorry! We couldn't be helpful

Thank you for your feedback

Let us know how can we improve this article!

Select atleast one of the reasons
CAPTCHA verification is required.

Feedback sent

We appreciate your effort and will try to fix the article