I am trying to address an issue caused by overfitting of a model. Unfortunately I don't know how to increase the interval of model.cpk that legacy/train.py outputs during training. Is there a way to reduce the time between each saving of model.cpk and to disable its deletion. I am training small models and can afford the increased storage requirement.
from How to configure tensorflow legacy/train.py model.cpk output interval
No comments:
Post a Comment