I eagerly anticipate the day CUDA no longer has a stranglehold on ML libraries; it's just absurd that one must purchase Magical Hardware when damn near every la

I eagerly anticipate the day CUDA no longer has a stranglehold on ML libraries; it's just absurd that one must purchase Magical Hardware when damn near every laptop has a GPU in it just not the GPU