Practice / Trust and use

Get started

From open frameworks and datasets to a surrogate trained on your own simulations.

• Pick data close to your physics: Start from an open dataset in the same regime (external aero, internal flow, thermal), or your own simulation archive.
• Start from a reference model: Frameworks ship tested implementations and training recipes; reproduce a published baseline before changing anything.
• Validate on held-out designs: Measure error on designs never seen in training, and report the training range alongside every prediction.
• Fine-tune on your own cases: Adapt a pretrained model to your product line with a smaller set of your own simulations.

Related

• Frameworks: Where to start in code: open neural frameworks, plus classical surrogate and UQ toolkits.
• Datasets: Public simulation datasets to benchmark on before you have your own.
• Design use: Millisecond answers turn design studies, optimisation and real-time tools into routine work.