sub-mod

need to install bazel
https://copr.fedorainfracloud.org/coprs/vbatts/bazel/

oc new-app --template=tensorflow-build-image  --param=APPLICATION_NAME=tf-fedora27-builder-image-36 --param=S2I_IMAGE=registry.fedoraproject.org/f27/s2i-core   --param=DOCKER_FILE_PATH=Dockerfile.fedora27 --param=NB_PYTHON_VER=3.6 --param=VERSION=4

oc new-app --template=tensorflow-build-job  --param=APPLICATION_NAME=tf-build-fc27-363 --param=BUILDER_IMAGESTREAM=docker-registry.default.svc:5000/dh-prod-analytics-factory/tf-fedora27-builder-image-36:3  --param=NB_PYTHON_VER=3.6  --param=GIT_TOKEN= --param=CUSTOM_BUILD="gcc -march=native -Q --help=target|grep march && gcc -march=native -E -v - </dev/null 2>&1 | grep cc1 && grep flags -m1 /proc/cpuinfo | cut -d ":" -f 2 | tr '[:upper:]' '[:lower:]' && lscpu && bazel build -c opt --copt='-march=core-avx2' --cxxopt='-D_GLIBCXX_USE_CXX11_ABI=0' --local_resources 2048,2.0,1.0 --verbose_failures //tensorflow/tools/pip_package:build_pip_package"



gcc -march=native -Q --help=target|grep marc

sub-mod

"""This file contains some basic model components"""

import tensorflow as tf
from tensorflow.python.ops.rnn_cell import DropoutWrapper
from tensorflow.python.ops import variable_scope as vs
from tensorflow.python.ops import rnn_cell
from operator import mul


class LSTMEncoder(object):

sub-mod

---
apiVersion: v1
data:
  jupyterhub_config.py: |
    import json
    import os
    from kubespawner.spawner import KubeSpawner
    from jhub_remote_user_authenticator.remote_user_auth import RemoteUserAuthenticator
    from oauthenticator.github import GitHubOAuthenticator

sub-mod

CRD and ClusterRole Both need cluster-admin

Create a CRD ...just the definition as cluster-admin

cat >> crd.yaml << EOF
apiVersion: apiextensions.k8s.io/v1beta1
kind: CustomResourceDefinition
metadata:
  name: crontabs.stable.example.com

sub-mod

import os
import sys
import tensorflow as tf
import numpy as np
from tensorflow.python.saved_model import builder as saved_model_builder
from tensorflow.python.saved_model import signature_constants
from tensorflow.python.saved_model import signature_def_utils
from tensorflow.python.saved_model import tag_constants
from tensorflow.python.saved_model import utils
from tensorflow.python.util import compat

sub-mod

1) Login as developer and make a Project

oc login -u developer -p developer
oc new-project test

2) Make the SA(tf-job-operator-sa) cluster-admin in the test Namespace

Since CRD & its instance has to be created by cluster-admin we have to make SA a cluster-admin

sub-mod

{
  "apiVersion": "tensorflow.org/v1alpha1",
  "kind": "TfJob",
  "metadata": {
    "name": "distributed-job"
  },
  "spec": {
    "replicaSpecs": [
      {
        "replicas": 1,

sub-mod

{
  "apiVersion":"v1",
  "kind":"Template",
  "template":"kubeflow-operator",
  "labels":{
    "template":"kubeflow-operator"
  },
  "message":"Kubeflow TfJob Operator",
  "metadata":{
    "name":"kubeflow-operator",

sub-mod

{
  "apiVersion": "tensorflow.org/v1alpha1",
  "kind": "TfJob",
  "metadata": {
    "name": "sample-tfjob"
  },
  "spec": {
    "replicaSpecs": [
      {
        "template": {

sub-mod

Notes for setting up GPU with Openshift

• Centos7 or RHEL7
• Openshift 3.7+
• CUDA9 and CuDNN7

1. Check the GPU model

lspci -nn | grep VGA  
lspci | grep -i --color 'vga\|3d\|2d'