Tests included instruction versions in TensorFlow and PyTorch, working Jupyter notebooks, and dealing with medium to massive datasets more than prolonged periods. The focus was on sustained efficiency as opposed to brief benchmark outcomes.We prioritized laptops with devoted NPUs effective a