Diff between Batch Size and Steps

[SnehitVaddi]

I am currently going through the “Finetuning Large Language Models” lecture by Sharon Zhou. In lesson three, she discusses steps and batch size, but I’m a bit confused. From my understanding in deep learning and image classification, if there are 100 samples and a batch size of 10, then for each epoch, there would be 10 batches running through the model, which equates to 10 steps per epoch.

Am I understanding this correctly, or am I missing something? Can someone please educate me on this?

[Deepti_Prasad]

hi @SnehitVaddi

please refer this

Are metrics for each batch or whole dataset?

WHAT IS STEPS AND EPOCH? Steps refer to the number of batches processed by the model during training. A batch is a subset of the training data used to update the model’s weights. The number of steps defines how many times the model goes through the training data. On the other hand, an epoch is defined as one complete pass through the entire training dataset. In other words, one epoch means the model has seen the entire training data once. During an epoch, the model goes through multiple batche…