Hi, community!
I am training a Topaz model for a Topaz Extract task. The training and test curves are shown below: the test curve looks promising, but the training curve oscillates.
My first question is, which curve should I focus on—the training curve, the test curve, or both? Under what circumstances would this be considered a good result?
My second question is, according to Topaz Train: low value of test average-precision over epochs - #2 by alexjamesnoble alex said, “Average-precision generally cannot be compared between different training parameters, and not compared to any absolute value”. How can I optimize the training parameters if I have no way to compare Average Precision？
Thanks in advance!