KrakathorOfTheLavaPeople
Reputation: 91
BASIC_TPU is not a valid choice in ML Engine
I've been following this guide to using TPUs in ML Engine but get an error that BASIC_TPU is an invalid choice:
ERROR: (gcloud.ml-engine.jobs.submit.training) argument --scale-tier: Invalid choice: 'BASIC_TPU'.
Valid choices are [BASIC, BASIC_GPU, CUSTOM, PREMIUM_1, STANDARD_1].
According to the doc, this should be a valid choice. Has anyone else run into this? Is this just the result of TPUs being in beta?
Upvotes: 1
Views: 87
Answers (1)
Related Questions
- Why is the TPU not recognized on my Google Cloud TPU VM instance?
- Google Cloud TPU -- no TPU being used
- How to select a specific TPU in Google Cloud?
- TPU core error on Google Cloud Platform (Cannot find any TPU cores in the system. Please double check Tensorflow master address and TPU worker)
- Simple model can't run on tpu (on colab)
- training MNIST with TPU generates errors
- Starting Google Cloud TPU is missing VM image specification
- Getting error in TPU with python mnist_tpu.py --use_tpu=false --tpu=''
- how to use "Cloud TPU" in google compute engine?
- TPUEstimator does not work with use_tpu=False