Remove tweaked benchmark for no_dist_strat in function mode.
No need to test explicit_gpu_placement + no_dist_strat + function case, since the performance issue is already fixed by unifying the Keras codepaths. However, there is still performance gap using the explicit_gpu_placement flag when run_eagerly. PiperOrigin-RevId: 283890008
Showing
想要评论请 注册 或 登录