Add operations to minimize `loss` by updating `var_list` with SGD and compute the batch size for the next step according to the CABS rule. Args: :losses: A rank 1 `Tensor` containing the individual loss values for each example in the batch. You can *not* insert a scalar me
(self, losses, var_list=None, global_bs=None)
source not stored for this graph (policy: none)
no outgoing calls
no test coverage detected