Main new developments
- New tutorials. We have added a new group of tutorials within the Getting Started page, focused on the role of data quantity, quality and sparsity when training machine learning models. This will help you to understand how these factors impact the accuracy and usefulness of the models you build within Monolith. Data preparation (capturing, structuring and pre-processing) is an integral part of building a machine learning solution.
- Manipulator Panel redesign. After receiving feedback on the ease of use our previous manipulator panel design, we’ve simplified the interface to reduce the total number of clicks needs to access manipulators.
- Improved responsiveness for narrow browser windows. The structures of navigation bar and of the manipulator panel now respond dynamically to varying window widths. This will make it easier for you to interact with our platform from smaller screens or within a tutorial notebook.
Other improvements
- Quick Columns: New operation to multiply a column by a constant.
- Surface Field model: Significant speedup when training on GPU, and fixed an issue in which Surface Field models had diverging loss curves.
- Labels on Parallel Coordinate plots are now more readable.
- Other bug fixes and performance improvements.
Was this article helpful?
That’s Great!
Thank you for your feedback
Sorry! We couldn't be helpful
Thank you for your feedback
Feedback sent
We appreciate your effort and will try to fix the article