Closed t4r4d4ctyl closed 4 years ago
You are using TF 2.x, since those samples are meant to run for Cloud AI platform which supports now 1.15 and only 2.X with custom containers, you need to modify the sample. PTAL at: https://github.com/tensorflow/tensorflow/issues/26662
Description Seeing an issue when running a command in relation to this file: cloudml-samples/census/estimator/trainer/task.py
I think this must have to do with the update on this file six months ago, as I can see that a reference to "logging" was removed, but I don't know for sure.
The reason I'm attempting this is because of a Linux Academy course for the GCP Data Engineer certification. The course attempts to use this repo for interactive demos.
Environment GCP Cloud Shell Environment tensorflow==2.0.0 tensorflow-estimator==2.0.1 Python 2.7.13 (default, Sep 26 2018, 18:42:22)
Command
gcloud ai-platform local train --job-dir $MODEL_DIR --module-name trainer.task --package-path train -- --train-files $TRAIN_DATA --eval-files $EVAL_DATA --train-steps 1000 --eval-steps 100
Logs
Traceback (most recent call last):
File "/usr/lib/python2.7/runpy.py", line 174, in _run_module_as_main\
"__main__", fname, loader, pkg_name)
File "/usr/lib/python2.7/runpy.py", line 72, in _run_code
exec code in run_globals
File "/home/<user>/cloudml-samples/census/estimator/trainer/task.py", line 189, in <module>
tf.logging.set_verbosity(ARGUMENTS.verbosity)
AttributeError: 'module' object has no attribute 'logging'