Batch size (ML)
Batch size is the size of each data subset that the model goes through while training. They tend to be in the range 32-512. There isn’t an ideal size for all iterations. That depends on underlying math.
Search
Jan 15, 2024
Batch size is the size of each data subset that the model goes through while training. They tend to be in the range 32-512. There isn’t an ideal size for all iterations. That depends on underlying math.