algorithmic-efficiency
algorithmic-efficiency copied to clipboard
Added test for Evaluating Timing Consistency in MNIST Workload Training for PyTorch and JAX and changed the int32 to uint32
A test is added to evaluate timing consistency in the MNIST training workload using PyTorch and JAX. It ensures that the total reported training time matches the sum of submission, evaluation, and logging times. The test verifies timing accuracy and checks that the expected number of evaluations are completed during training.