Jupyter notebooks for testing need to be configured with model checkpoint directory

aws-samples / amazon-eks-machine-learning-with-terraform-and-kubeflow

Distributed training using Kubeflow on Amazon EKS

Apache License 2.0

82 stars 42 forks source link

Jupyter notebooks for testing need to be configured with model checkpoint directory #29

Closed ajayvohra2005 closed 2 years ago

ajayvohra2005 commented 2 years ago

Jupyter notebooks used for testing are currently not configured to automatically point to the model checkpoint directory for trained models. The user has to manually select the model checkpoint directory.

ajayvohra2005 commented 2 years ago

Hash commit 3935e4fb22bd48ef3f1d8aafbe99cf5f811a27d9 resolves this issue. Notebooks use LOGDIR environment variable to locate the model checkpoint directory.