Merge pull request #542 from yahoo/leewyang_sd

migrate build from travis to screwdriver

README.md

@@ -7,7 +7,10 @@ Please see LICENSE file in the project root for terms.
 > _TensorFlowOnSpark brings scalable deep learning to Apache Hadoop and Apache Spark
 clusters._
 
-[![Build Status](https://travis-ci.org/yahoo/TensorFlowOnSpark.svg?branch=master)](https://travis-ci.org/yahoo/TensorFlowOnSpark) [![PyPI version](https://badge.fury.io/py/tensorflowonspark.svg)](https://badge.fury.io/py/tensorflowonspark)
+[![Build Status](https://cd.screwdriver.cd/pipelines/6384/badge?nocache=true)](https://cd.screwdriver.cd/pipelines/6384)
+[![Package](https://img.shields.io/badge/package-pypi-blue.svg)](https://pypi.org/project/tensorflowonspark/)
+[![Downloads](https://img.shields.io/pypi/dm/tensorflowonspark.svg)](https://img.shields.io/pypi/dm/tensorflowonspark.svg)
+[![Documentation](https://img.shields.io/badge/Documentation-latest-blue.svg)](https://yahoo.github.io/TensorFlowOnSpark/)
 
 By combining salient features from the [TensorFlow](https://www.tensorflow.org) deep learning framework with [Apache Spark](http://spark.apache.org) and [Apache Hadoop](http://hadoop.apache.org), TensorFlowOnSpark enables distributed
 deep learning on a cluster of GPU and CPU servers.

docs/source/conf.py

@@ -28,9 +28,9 @@
 author = 'Yahoo Inc'
 
 # The short X.Y version
-version = '2.2.1'
+version = '2.2.2'
 # The full version, including alpha/beta/rc tags
-release = '2.2.1'
+release = '2.2.2'
 
 
 # -- General configuration ---------------------------------------------------

pom.xml

@@ -5,7 +5,7 @@
     <modelVersion>4.0.0</modelVersion>
     <groupId>com.yahoo.ml</groupId>
     <artifactId>tensorflowonspark</artifactId>
-    <version>2.2.0-SNAPSHOT</version>
+    <version>2.2.2-SNAPSHOT</version>
     <packaging>jar</packaging>
     <name>tensorflowonspark</name>
     <description>Spark Scala inferencing for TensorFlowOnSpark</description>

requirements.txt

@@ -2,7 +2,7 @@ h5py>=2.9.0
 numpy>=1.14.0
 packaging
 py4j==0.10.7
-pyspark==2.4.5
+pyspark==2.4.7
 scipy
 setuptools>=41.0.0
 sphinx

screwdriver.yaml

@@ -0,0 +1,56 @@
+# Copyright 2017, Verizon Inc.
+# Licensed under the terms of the apache license.  See the LICENSE file in the project root for terms
+
+version: 4
+shared:
+  environment:
+    PACKAGE_DIRECTORY: tensorflowonspark
+    SPARK_HOME: ${SD_ROOT_DIR}/spark
+    TOX_ARGS: '--verbose'
+    TOX_ENVLIST: py37
+  annotations:
+    screwdriver.cd/cpu: HIGH
+    screwdriver.cd/ram: HIGH
+
+jobs:
+  validate_test:
+    template: python/validate_unittest
+    requires: [~commit, ~pr]
+    steps:
+      - prevalidate_code: |
+          source scripts/install_spark.sh
+
+  validate_lint:
+    template: python/validate_lint
+    requires: [~commit, ~pr]
+
+  validate_codestyle:
+    template: python/validate_codestyle
+    requires: [~commit, ~pr]
+
+  validate_safetydb:
+    template: python/validate_safety
+    requires: [~commit, ~pr]
+
+#  validate_security:
+#    template: python/validate_security
+#    requires: [~commit, ~pr]
+
+  publish_test_pypi:
+    template: python/package_python
+    environment:
+      PUBLISH: True
+      TWINE_REPOSITORY_URL: https://test.pypi.org/legacy/
+    requires: [validate_test, validate_lint, validate_codestyle, validate_safetydb]
+    steps:
+      - update_version: |
+          echo 'using version from setup.cfg'
+
+  publish_pypi:
+    template: python/package_python
+    environment:
+      PUBLISH: True
+    requires: [publish_test_pypi]
+    steps:
+      - update_version: |
+          echo 'using version from setup.cfg'

scripts/install_spark.sh

@@ -0,0 +1,11 @@
+#!/bin/bash -x
+
+# Install JDK8
+yum install -y java-1.8.0-openjdk
+export JAVA_HOME=/usr/lib/jvm/jre-1.8.0
+
+# Install Spark
+export SPARK_VERSION=2.4.7
+curl -LO http://www-us.apache.org/dist/spark/spark-${SPARK_VERSION}/spark-${SPARK_VERSION}-bin-hadoop2.7.tgz
+mkdir $SPARK_HOME
+tar -xf spark-${SPARK_VERSION}-bin-hadoop2.7.tgz -C $SPARK_HOME --strip-components=1

scripts/start_spark.sh

@@ -0,0 +1,12 @@
+#!/bin/bash -x
+#export SPARK_HOME=/opt/spark
+#export SPARK_LOCAL_IP=127.0.0.1
+#export PATH=$SPARK_HOME/bin:$PATH
+#
+## Start Spark Standalone Cluster
+#export SPARK_CLASSPATH=./lib/tensorflow-hadoop-1.0-SNAPSHOT.jar
+#export MASTER=spark://$(hostname):7077
+#export SPARK_WORKER_INSTANCES=2; export CORES_PER_WORKER=1
+#export TOTAL_CORES=$((${CORES_PER_WORKER}*${SPARK_WORKER_INSTANCES}))
+
+${SPARK_HOME}/sbin/start-master.sh; ${SPARK_HOME}/sbin/start-slave.sh -c ${CORES_PER_WORKER} -m 1G ${MASTER}

scripts/stop_spark.sh

@@ -0,0 +1,3 @@
+#!/bin/bash -x
+
+${SPARK_HOME}/sbin/stop-slave.sh; ${SPARK_HOME}/sbin/stop-master.sh

sd.allow

@@ -0,0 +1,3 @@
+version: 1
+push:
+  - screwdriver:6384

setup.cfg

@@ -1,5 +1,76 @@
+# Copyright 2017, Verizon Inc.
+# Licensed under the terms of the apache license.  See the LICENSE file in the project root for terms
 [metadata]
-description-file = README.md
+author = Lee Yang
+author_email = [email protected]
+classifiers =
+    Intended Audience :: Developers
+    Intended Audience :: Science/Research
+    License :: OSI Approved :: Apache Software License
+    Topic :: Software Development :: Libraries
+    Programming Language :: Python :: 3 :: Only
+    Programming Language :: Python :: 3.6
+    Programming Language :: Python :: 3.7
+    Programming Language :: Python :: 3.8
+description = Deep learning with TensorFlow on Apache Spark clusters
+license = Apache 2.0
+long_description = file:README.md
+long_description_content_type = text/markdown
+name = tensorflowonspark
+url = https://github.com/yahoo/TensorFlowOnSpark
+version = 2.2.2
+
+[options]
+packages =
+    tensorflowonspark
+
+# The install_requires should include abstract package dependencies
+# here (do not specify specific versions)
+
+install_requires =
+   setuptools>38.0
+
+# By default new packages require at minimum the current supported Python release.
+python_requires = >="3.6"
+zip_safe = True
+
+[options.extras_require]
+# This config section allows you to define optional dependencies.  For the general case, the defaults will
+# work fine.  So these settings aren't required.  However, many of the screwdriver CI Pipeline steps
+# will install the appropriate extras for that step.  This makes it possible to install packages that install
+# or enhance the functionality of the CI Pipeline step.
+# Such as packages that implement plugins or themes for the step in question.
+
+# Additional packages for testing (test step)
+# test =
+
+# Additonal packages needed for documentation generation (doc_build/doc_publish steps)
+# If you want to use a sphinx theme from a package, list it here.
+# doc_build =
+
+# Additional packages needed for mypy type checking
+# mypy =
+
+# Additional packages needed for pep8/pycodestyle style checking
+# pycodestyle =
+
+# Additional packages needed for pylint code analysis
+# pylint =
+
+[options.entry_points]
+# Console script entry points are used to create wrapper scripts that run a specific function, the resulting wrapper
+# is installed in the bin directory.
+
+# They are defined using the following format:
+#    scriptname = modulename:function
+# console_scripts =
+#     TFoS=ouroath.TFoS.cli:main
+
+[screwdrivercd.version]
+# Base the autoversion build number on the screwdriver build number
+# This requires the CI Pipeline to have a build step that runs before
+# any packaging steps.
+version_type = sdv4_SD_BUILD
 
 [bdist_wheel]
 universal = 1

setup.py

@@ -1,29 +1,26 @@
-from setuptools import setup
+#!/usr/bin/env python
+# Copyright 2017, Verizon Inc.
+# Licensed under the terms of the apache license.  See the LICENSE file in the project root for terms
+"""
+Package setup file for python module 'tensorflowonspark'
+"""
+import setuptools
+import sys
 
-with open('README.md') as f:
-  long_description = f.read()
 
-setup(
-  name='tensorflowonspark',
-  packages=['tensorflowonspark'],
-  version='2.2.1',
-  description='Deep learning with TensorFlow on Apache Spark clusters',
-  long_description=long_description,
-  long_description_content_type='text/markdown',
-  author='Yahoo, Inc.',
-  url='https://github.com/yahoo/TensorFlowOnSpark',
-  keywords=['tensorflowonspark', 'tensorflow', 'spark', 'machine learning', 'yahoo'],
-  install_requires=['packaging'],
-  license='Apache 2.0',
-  classifiers=[
-    'Intended Audience :: Developers',
-    'Intended Audience :: Science/Research',
-    'License :: OSI Approved :: Apache Software License',
-    'Topic :: Software Development :: Libraries',
-    'Programming Language :: Python :: 2',
-    'Programming Language :: Python :: 2.7',
-    'Programming Language :: Python :: 3',
-    'Programming Language :: Python :: 3.5',
-    'Programming Language :: Python :: 3.6'
-  ]
-)
+def setuptools_version_supported():
+    major, minor, patch = setuptools.__version__.split('.')
+    if int(major) > 38:
+        return True
+    return False
+
+
+if __name__ == '__main__':
+    # Check for a working version of setuptools here because earlier versions did not
+    # support python_requires.
+    if not setuptools_version_supported():
+        print('Setuptools version 38.0.0 or higher is needed to install this package')
+        sys.exit(1)
+
+    # We're being run from the command line so call setup with our arguments
+    setuptools.setup()

tensorflowonspark/__init__.py

@@ -2,4 +2,4 @@
 
 logging.basicConfig(level=logging.INFO, format="%(asctime)s %(levelname)s (%(threadName)s-%(process)d) %(message)s")
 
-__version__ = "2.2.1"
+__version__ = "2.2.2"