A greenhouse trial has twelve trays. A model memorizes which tray IDs had a cracked moisture probe and scores 100 percent on those twelve. The next season, new trays with intact probes are scored poorly. What happened?
Select an answer to reveal the explanation.
Short Explanation
Think of a student who memorizes the twelve tray ID numbers instead of learning what a dry plant looks like. Perfect on the old trays, lost on the new season. That is overfitting, the opposite of underfitting, and extra GPUs will not un-memorize a cracked probe.
Full Explanation
Overfitting is when a model memorizes quirks in the training examples and then fails on new data. Perfect accuracy on twelve tray IDs and a collapse on the next season is that pattern. Extra GPUs do not fix a model that learned the wrong signal. Underfitting is a poor fit on the training data, which is the opposite of 100 percent training accuracy. Batch inferencing is a way to score many examples at once, not a diagnosis of this failure.