A maple-sugar cooperative’s training log shows how many times the full note set was seen, how many notes were used per update, and how large each weight nudge was. Which three knobs are those?
Select an answer to reveal the explanation.
Short Explanation
How many times the full set was seen, how many notes per update, and how large each nudge was. Those knobs are epoch, batch size, and learning rate. Hardware, RAG, and NCCL are not those three.
Full Explanation
An epoch is one pass over the training set, batch size is how many examples drive each update, and learning rate is how large that update is. Those are the common training knobs. Hardware, RAG, and Professional parallelism settings are not those three.