Model Evaluation and Validation

Model Evaluation

When model training is complete, the system will automatically select the best model for evaluation and calculate relevant metrics. These metrics give us a rough idea of the model's training performance.

AI Creator allows you to compare the evaluation results with the labeled data. By viewing this comparison, you can better assess whether the trained model meets your needs.

Model Validation

After the model evaluation, we can use unseen data samples to validate whether the trained model performs well on samples it hasn't encountered during training. The model validation feature allows us to perform this task.

On the Model Validation page, select the dataset to be used for validation. Here, you can only choose the test set from the Dataset, then click Start Validation.

Once validation is complete, the results will automatically appear. You can switch between images on the left to view the model's recognition results for each.

At this point, we have completed the entire model training process. If you're interested in learning more about model optimization, deployment, and generating applications for smart cameras, please proceed to the next step.

Next Step

After completing the model training, if we wish to deploy it to edge devices, we need to go through Model Optimization and Deployment.