Training a model

Trained models can be deployed for use in visual inspections.

Before you begin

Ensure that you have the following GPU storage:

  • At least one GPU available.
  • Enough GPUs available for your workload. The number of active GPU tasks that you can run at the same time depends on the number of GPUs on your server.

About this task

For more information about model types, see Model types.

Procedure

  1. Open the relevant data set and then click Train model.
  2. Select the relevant training type and then click Train model.
    Note: When a model is in training, a graph shows the relative performance of the model over time. For more information, see the Graph for training models.
  3. Optional: To stop training a model that is in training or is in the training queue, click Stop training.

What to do next

If training fails for Detectron2, Faster R-CNN, High resolution, or Anomaly optimized models, you can slow the training learning rate, and then run the training again. Training for these models might fail due to the speed of the learning rate. For example, a training configuration that contains the same model and data set might succeed and fail in two different training tasks that use the same learning rate