Training Requirements
A recognizer can take a significant amount of time to train.
The time required to train a recognizer is proportional to the number of training iterations. Increasing the number of iterations can result in better accuracy, but each additional iteration that you add has a smaller effect. Running too many iterations may result in overfitting, meaning the recognizer becomes so well adapted to the training data that it performs less well when recognizing unknown objects.
As a general rule, you can calculate a reasonable number of iterations by multiplying the number of object classes in the recognizer by 2000.
You must choose the number of iterations to run before you begin training your recognizer. Changing the number of iterations invalidates the training and Media Server must begin training again from the beginning.
Media Server can help you find the optimum number of iterations, by setting aside some of your training images for evaluation purposes. For example, you could use 80% of your training images for training a recognizer and 20% for evaluating its performance. Media Server only sets aside training images when you enable snapshots. A snapshot captures the state of a recognizer after a certain number of iterations. For example, you can choose to run 2000 iterations and take snapshots of your recognizer after every 500 iterations. You can then test the performance of the recognizer (using the reserved images) at each snapshot. The snapshot that represents the greatest number of iterations usually performs best; if you see a reduction in performance this indicates overfitting.