Merge pull request #1660 from microsoft/staging

Staging to main (SARplus, SASrec, NCF, RBM etc.)

.github/workflows/actions/merge-cov/action.yml

@@ -33,9 +33,18 @@ runs:
   # different parallelized runs
   - name: Merge coverage reports
     shell: bash
+    # NOTE: Merging code reports generated from self-hosted (aml) runners can be problematic
+    #       as reports may reference the source code at an unresolvable absolute path.
+    #       For example, you may encounter errors like:
+    #            "CoverageWarning: Couldn't parse 
+    #            '/home/azureuser/runner/work/recommenders/recommenders/recommenders/evaluation/__init__.py':
+    #            No source for code"
+    # Work-around: Creating a symlink at the root that points to the default local runner folder: '/home/runner/work/recommenders'
     run: |
+      sudo mkdir -p /home/azureuser
+      sudo ln -s /home/runner /home/azureuser/runner
       python -m coverage combine .coverage*
-      python -m coverage report
+      python -m coverage report -i
       python -m coverage xml -i
       
   - name: Show merged report

.github/workflows/nightly.yml

@@ -184,6 +184,7 @@ jobs:
   build-gpu:
     runs-on: [self-hosted, Linux, gpu, nightly] # this is a union of labels to select specific self-hosted machine
     needs: static-analysis
+    timeout-minutes: 420
     strategy:
       matrix:
         python: [3.7]

.github/workflows/pr-gate.yml

@@ -129,7 +129,7 @@ jobs:
     strategy:
       matrix:
         os: [ubuntu-latest]
-        java: [11]
+        java: [8]
         python: [3.7]
         # different kinds of tests are located in tests/<unit|integration|smoke> folders
         test-kind: ['unit'] 
@@ -176,7 +176,10 @@ jobs:
         # different kinds of tests are located in tests/<unit|integration|smoke> folders
         test-kind: ['unit'] 
         # pytest markers configured in tox.ini. See https://docs.pytest.org/en/6.2.x/example/markers.html
-        test-marker: ['gpu and notebooks and not spark and not experimental', 'gpu and not notebooks and not spark and not experimental']
+        test-marker: [
+          'gpu and not notebooks and not spark and not experimental', 
+          'gpu and notebooks and not spark and not experimental'
+        ]
 
     steps:
     - uses: actions/checkout@v2

.github/workflows/sarplus.yml

@@ -19,6 +19,7 @@ on:
       - .github/workflows/sarplus.yml
 
 env:
+  SARPLUS_ROOT: ${{ github.workspace }}/contrib/sarplus
   PYTHON_ROOT: ${{ github.workspace }}/contrib/sarplus/python
   SCALA_ROOT: ${{ github.workspace }}/contrib/sarplus/scala
 
@@ -52,15 +53,20 @@ jobs:
 
       - name: Package and check
         run: |
+          # build
           cd "${PYTHON_ROOT}"
-          cp ../VERSION ./pysarplus/
+          cp "${SARPLUS_ROOT}/VERSION" ./pysarplus/VERSION
           python -m build --sdist
           python -m twine check dist/*
 
+          # set sarplus_version
+          SARPLUS_VERSION=$(cat "${SARPLUS_ROOT}/VERSION")
+          echo "sarplus_version=${SARPLUS_VERSION}" >> $GITHUB_ENV
+
       - name: Test
         run: |
           cd "${PYTHON_ROOT}"
-          python -m pip install dist/*.gz
+          python -m pip install dist/*.tar.gz
 
           cd "${SCALA_ROOT}"
           export SPARK_VERSION=$(python -m pip show pyspark | grep -i version | cut -d ' ' -f 2)
@@ -75,14 +81,13 @@ jobs:
 
           cd "${PYTHON_ROOT}"
           pytest ./tests
-          echo "sarplus_version=$(cat ../VERSION)" >> $GITHUB_ENV
 
       - name: Upload Python package as GitHub artifact
         if: github.ref == 'refs/heads/main' && matrix.python-version == '3.10'
         uses: actions/upload-artifact@v2
         with:
           name: pysarplus-${{ env.sarplus_version }}
-          path: ${{ env.PYTHON_ROOT }}/dist/*.gz
+          path: ${{ env.PYTHON_ROOT }}/dist/*.tar.gz
 
   scala-test:
     # Test sarplus with different versions of Databricks runtime, 2 LTSs and 1
@@ -129,6 +134,8 @@ jobs:
         env:
           GPG_KEY: ${{ secrets.SARPLUS_GPG_PRI_KEY_ASC }}
         run: |
+          SARPLUS_VERSION=$(cat "${SARPLUS_ROOT}/VERSION")
+
           # generate artifacts
           cd "${SCALA_ROOT}"
           export SPARK_VERSION="3.1.2"
@@ -142,18 +149,24 @@ jobs:
           export HADOOP_VERSION="3.3.1"
           export SCALA_VERSION="2.12.14"
           sbt ++${SCALA_VERSION}! package
+          sbt ++${SCALA_VERSION}! packageDoc
+          sbt ++${SCALA_VERSION}! packageSrc
+          sbt ++${SCALA_VERSION}! makePom
 
           # sign with GPG
-          cd target/scala-2.12
+          cd "${SCALA_ROOT}/target/scala-2.12"
           gpg --import <(cat <<< "${GPG_KEY}")
           for file in {*.jar,*.pom}; do gpg -ab "${file}"; done
 
           # bundle
-          jar cvf sarplus-bundle_2.12-$(cat ../VERSION).jar *.jar *.pom *.asc
-          echo "sarplus_version=$(cat ../VERSION)" >> $GITHUB_ENV
+          jar cvf sarplus-bundle_2.12-${SARPLUS_VERSION}.jar sarplus_*.jar sarplus_*.pom sarplus_*.asc
+          jar cvf sarplus-spark-3.2-plus-bundle_2.12-${SARPLUS_VERSION}.jar sarplus-spark*.jar sarplus-spark*.pom sarplus-spark*.asc
+
+          # set sarplus_version
+          echo "sarplus_version=${SARPLUS_VERSION}" >> $GITHUB_ENV
           
-      - name: Upload Scala bundle  as GitHub artifact
+      - name: Upload Scala bundle as GitHub artifact
         uses: actions/upload-artifact@v2
         with:
           name: sarplus-bundle_2.12-${{ env.sarplus_version }}
-          path: ${{ env.SCALA_ROOT }}/target/scala-2.12/sarplus-bundle_2.12-${{ env.sarplus_version }}.jar
+          path: ${{ env.SCALA_ROOT }}/target/scala-2.12/*bundle*.jar

.gitignore

@@ -175,3 +175,6 @@ examples/07_tutorials/KDD2020-tutorial/data_folder/
 
 tests/**/resources/
 reports/
+
+### pip folders
+pip-wheel*

AUTHORS.md

@@ -41,6 +41,9 @@ To contributors: please add your name to the list when you submit a patch to the
 * **[Aaron He](https://github.com/AaronHeee)**
    * Reco utils of NCF
    * Deep dive notebook demonstrating the use of NCF
+* **[Abir Chakraborty](https://github.com/aeroabir)**
+   * Self-Attentive Sequential Recommendation (SASRec)
+   * Sequential Recommendation Via Personalized Transformer (SSEPT)
 * **[Alexandros Ioannou](https://github.com/aioannou96)**
    * Standard VAE algorithm 
    * Multinomial VAE algorithm  
@@ -90,6 +93,8 @@ To contributors: please add your name to the list when you submit a patch to the
    * Improving documentation
    * Quick start notebook
    * Operationalization notebook
+* **[Nile Wilson](https://github.com/niwilso)**
+   * Term Frequency - Inverse Document Frequency (TF-IDF) quickstart, utils
 * **[Pratik Jawanpuria](https://github.com/pratikjawanpuria)**
    * RLRMC algorithm
    * GeoIMC algorithm