xla_device() over directly specifying TPU cores. A future Colab tutorial will show how to easily train a network using multiple cores (or you can look at an ...
確定! 回上一頁