what are these params -
steps_per_epochs and validation_steps
and how should i choose optimal values for these params. can i put any value, or is it according to dataset. i hope taking bad values does not give extra overload on my gpu.
what are these params -
steps_per_epochs and validation_steps
and how should i choose optimal values for these params. can i put any value, or is it according to dataset. i hope taking bad values does not give extra overload on my gpu.
Hey Mohit,
Both of these parameters represent the batch size. You can understand it by relating it to the normal ‘fit’ function we use while working with neural networks in keras. Just how we specify the batch_size seperately in that case, here we mention the number of steps we want to take per epoch to cover the entire data. If you specify the steps as ‘n’ then your batch size would be (no. of examples)/n. Also, the weights are updated after every step.
Your choose the optimal values of these parameters by analysing your data properly, these parameters totally depend on the size of your data. If the number of examples are huge then the batch size should be large as well.
I hope this clarifies your doubt
I hope I’ve cleared your doubt. I ask you to please rate your experience here
Your feedback is very important. It helps us improve our platform and hence provide you
the learning experience you deserve.
On the off chance, you still have some questions or not find the answers satisfactory, you may reopen
the doubt.