Training Le continued to decline. Before the 6500th iteration, when the validate function is called, the LE of the test is falling, while the recall is rising, and everything is fine. But after that, the LE of the test started to rise, but the recall is still getting better, why?
Training Le continued to decline. Before the 6500th iteration, when the validate function is called, the LE of the test is falling, while the recall is rising, and everything is fine. But after that, the LE of the test started to rise, but the recall is still getting better, why?