Getting started
Using Siml.ai's Model Engineer, you can visually create and optimize "neural physics simulators" for various applications using deep learning techniques. The visual editor is node-based, with each node representing a specific operation. Nodes include Geometry Node, Neural Network Node, Equation Node, Constraint Node, and Geometry Node.
Figure 1.: Drag and drop nodes that can be used to build a simulator model.
With Siml.ai (opens in a new tab)'s user-friendly interface, creating and training a simulator from scratch becomes incredibly easy. Here's how you can create a simulator in a few simple steps:
- Click on “Create Simulator” when building a new simulator or click one from the existing list to resume development.
- Start with the visual editor based on a node-based graph. Drag and drop these nodes into the visual editor to build your simulator model.
- Upload the geometry. Use the Geometry Viewer for an interactive rendering of your model's geometry.
- Select the equation and preferred neural network architecture from the node dropdown.
- When adding the constraint node, you can adjust parameters and define boundaries by declaring variables to achieve your desired outcome effortlessly!
- Customize the specifications for training the simulator within the editor using the Configuration tab.
For more detailed information on using specific features, refer to the following documentation (next parts under current section from left sidebar). This will provide you with a deeper understanding of Siml.ai (opens in a new tab)'s capabilities.