Fix issues with the xgboost_synthetic example and deploying the model. (#682)

* Fix issues with the xgboost_synthetic example and deploying the model.

* install newer version of fairing
* modify preprocessor to use custom dockerfile
* use newer 0.7 base image.
* Fix endpoint.

Related to:

kubeflow/examples#673 model doesn't deploy its crash looping
Related to kubeflow/examples#655 update example to work with 0.7

* Add some comments to the notebook.
This commit is contained in:
Jeremy Lewi 2019-11-25 14:55:10 -08:00 committed by Kubernetes Prow Robot
parent 91374e6d27
commit 712c29a18e
2 changed files with 303 additions and 578 deletions

File diff suppressed because it is too large Load Diff

View File

@ -11,7 +11,7 @@ import subprocess
from pathlib import Path
KFP_PACKAGE = 'https://storage.googleapis.com/ml-pipeline/release/0.1.32/kfp.tar.gz'
FAIRING_PACKAGE = 'git+git://github.com/kubeflow/fairing.git@7c93e888c3fc98bdf5fb0140e90f6407ce7a807b' # pylint: disable=line-too-long
FAIRING_PACKAGE = 'git+git://github.com/kubeflow/fairing.git@9b0d4ed4796ba349ac6067bbd802ff1d6454d015' # pylint: disable=line-too-long
def notebook_setup():
# Install the SDK