Closed Once88 closed 5 years ago
I find estimator.evaluate is much slower than estimator.train in train.py , and when estimator.evaluate is running, the gpu usage is always 0.
my tensorflow version is 1.12
The updated version does not depend on tf.estimator anymore, it may solve your problem.
I find estimator.evaluate is much slower than estimator.train in train.py , and when estimator.evaluate is running, the gpu usage is always 0.