批大小与轮次Batch Size and Epochs
Batch Size and Epochs
解决什么问题What problem it solves
批大小是一次更新看多少样本,轮次是整套数据过几遍。Batch size is how many examples one update sees; an epoch is one full pass over the data.
这是什么What it is
批大小影响梯度噪声、显存占用和吞吐;轮次影响模型见过数据多少次。
两者与学习率耦合:批大小变了,合适的学习率通常也要变。
Batch size affects gradient noise, memory use and throughput; epochs affect how often the model sees the data.
Both couple with the learning rate: change the batch size and the suitable learning rate usually changes too.
为什么需要知道Why it matters
「训练 3 轮」不是通用配方;数据量、任务难度和过拟合信号决定该看多少遍。“Train for three epochs” is not a universal recipe. Data volume, task difficulty and overfitting signals decide how many passes are right.
别名Chinese aliases
batch size · epoch · 批次大小 · 训练轮次
同域词条Terms in the same area
现在能做什么What you can do now
回到名词地图看第4站的其他 46 个词 →Back to the map: the other 46 terms in station 4 →