Testing a custom object detection model

You can test a custom object detection model in IBM Watson Studio.

 

Before you begin

Train your custom model

 

Testing your model

After training your model, you can test your custom model in the Test area of the Object Detection model builder:

  • Test individual image files by dragging them from your computer onto the test area.
     
  • Test all the images in a .zip file from the the data panel. (If the data panel isn’t open, you can open it by clicking the Find and add data icon (The find data icon).)

Your model will return confidence scores for objects that match the training data. For example, uploading an image of a brain scan to the tumor detection model shows that the model identifies a possible tumor and gives a confidence score of 45%. Confidence scores will rise as the performance of the model improves through more training.

Testing object recognition

Retraining a model

To retrain your model, click Edit and retrain and continue labeling images in your data set. You can continue to retrain and test until you are satisfied with the performance of the model.

Next steps