Merge pull request #580 from EducationalTestingService/release/2.0

Preparing for SKLL v2.0 Release

Author: desilinguist

CONTRIBUTING.md

@@ -19,7 +19,7 @@ How to contribute
 
 3. Create an isolated environment for SKLL development. We recommend using the [conda](https://conda.io/en/latest/) package manager. To create a `conda` environment, run the following command in the root of the working directory:
 
-         $ conda create -n sklldev -c conda-forge --file conda_requirements.txt
+         $ conda create -n sklldev -c conda-forge nose --file conda_requirements.txt
 
 4. Activate the conda environment
 

DistributeTests.ps1

@@ -0,0 +1,59 @@
+<#  
+.SYNOPSIS  
+    Distribute the tests in VSTS pipeline across multiple agents 
+.DESCRIPTION  
+    This script divides test files across multiple agents for running on Azure DevOps.
+    It is adapted from the script in this repository: 
+    https://github.com/PBoraMSFT/ParallelTestingSample-Python/blob/master/DistributeTests.ps1
+
+    The distribution is basically identical to the way we do it in .travis.yaml
+#>
+
+$tests = Get-ChildItem .\tests\ -Filter "test*.py" # search for test files with specific pattern.
+$totalAgents = [int]$Env:SYSTEM_TOTALJOBSINPHASE # standard VSTS variables available using parallel execution; total number of parallel jobs running
+$agentNumber = [int]$Env:SYSTEM_JOBPOSITIONINPHASE  # current job position
+$testCount = $tests.Count
+
+# below conditions are used if parallel pipeline is not used. i.e. pipeline is running with single agent (no parallel configuration)
+if ($totalAgents -eq 0) {
+    $totalAgents = 1
+}
+if (!$agentNumber -or $agentNumber -eq 0) {
+    $agentNumber = 1
+}
+
+Write-Host "Total agents: $totalAgents"
+Write-Host "Agent number: $agentNumber"
+Write-Host "Total tests: $testCount"
+
+$testsToRun= @()
+
+if ($agentNumber -eq 1) {
+    $testsToRun = $testsToRun + "tests/test_featureset.py"
+    $testsToRun = $testsToRun + "tests/test_utilities.py"
+}
+elseif ($agentNumber -eq 2) {
+    $testsToRun = $testsToRun + "tests/test_output.py"
+}
+elseif ($agentNumber -eq 3) {
+    $testsToRun = $testsToRun + "tests/test_regression.py"
+}
+elseif ($agentNumber -eq 4) {
+    $testsToRun = $testsToRun + "tests/test_input.py"
+    $testsToRun = $testsToRun + "tests/test_preprocessing.py"
+    $testsToRun = $testsToRun + "tests/test_metrics.py"
+    $testsToRun = $testsToRun + "tests/test_custom_learner.py"
+    $testsToRun = $testsToRun + "tests/test_logutils.py"
+    $testsToRun = $testsToRun + "tests/test_examples.py"
+}
+elseif ($agentNumber -eq 5) {
+    $testsToRun = $testsToRun + "tests/test_classification.py"
+    $testsToRun = $testsToRun + "tests/test_cv.py"
+    $testsToRun = $testsToRun + "tests/test_ablation.py"
+}
+
+# join all test files seperated by space. pytest runs multiple test files in following format pytest test1.py test2.py test3.py
+$testFiles = $testsToRun -Join " "
+Write-Host "Test files $testFiles"
+# write these files into variable so that we can run them using pytest in subsequent task. 
+Write-Host "##vso[task.setvariable variable=pytestfiles;]$testFiles" 

MANIFEST.in

@@ -1,7 +1,6 @@
 include LICENSE.txt
 include *.rst
 include requirements.txt
-include requirements_rtd.txt
-recursive-include doc *
-recursive-include examples *
-recursive-include tests *
+recursive-exclude doc *
+recursive-exclude tests *
+recursive-exclude examples *

README.rst

@@ -15,8 +15,8 @@ SciKit-Learn Laboratory
 .. image:: https://img.shields.io/pypi/l/skll.svg
    :alt: License
 
-.. image:: https://img.shields.io/conda/v/desilinguist/skll.svg
-   :target: https://anaconda.org/desilinguist/skll
+.. image:: https://img.shields.io/conda/v/ets/skll.svg
+   :target: https://anaconda.org/ets/skll
    :alt: Conda package for SKLL
 
 .. image:: https://img.shields.io/pypi/pyversions/skll.svg
@@ -37,6 +37,25 @@ our project is to make it so that you can run scikit-learn experiments without
 actually needing to write any code other than what you used to generate/extract
 the features.
 
+Installation
+~~~~~~~~~~~~
+
+You can install using either ``pip`` or ``conda``. See details `here <https://skll.readthedocs.io/en/latest/getting_started.html>`__.
+
+Requirements
+~~~~~~~~~~~~
+
+-  Python 3.6+
+-  `beautifulsoup4 <http://www.crummy.com/software/BeautifulSoup/>`__
+-  `gridmap <https://pypi.org/project/gridmap/>`__ (only required if you plan
+   to run things in parallel on a DRMAA-compatible cluster)
+-  `joblib <https://pypi.org/project/joblib/>`__
+-  `pandas <http://pandas.pydata.org>`__
+-  `ruamel.yaml <http://yaml.readthedocs.io/en/latest/overview.html>`__
+-  `scikit-learn <http://scikit-learn.org/stable/>`__
+-  `seaborn <http://seaborn.pydata.org>`__
+-  `tabulate <https://pypi.org/project/tabulate/>`__
+
 Command-line Interface
 ~~~~~~~~~~~~~~~~~~~~~~
 
@@ -81,7 +100,6 @@ like:
   probability = true
   models = output
 
-
 For more information about getting started with ``run_experiment``, please check
 out `our tutorial <https://skll.readthedocs.org/en/latest/tutorial.html>`__, or
 `our config file specs <https://skll.readthedocs.org/en/latest/run_experiment.html>`__.
@@ -125,20 +143,6 @@ A Note on Pronunciation
 SciKit-Learn Laboratory (SKLL) is pronounced "skull": that's where the learning
 happens.
 
-Requirements
-~~~~~~~~~~~~
-
--  Python 3.6+
--  `beautifulsoup4 <http://www.crummy.com/software/BeautifulSoup/>`__
--  `gridmap <https://pypi.org/project/gridmap/>`__ (only required if you plan
-   to run things in parallel on a DRMAA-compatible cluster)
--  `joblib <https://pypi.org/project/joblib/>`__
--  `pandas <http://pandas.pydata.org>`__
--  `ruamel.yaml <http://yaml.readthedocs.io/en/latest/overview.html>`__
--  `scikit-learn <http://scikit-learn.org/stable/>`__
--  `seaborn <http://seaborn.pydata.org>`__
--  `tabulate <https://pypi.org/project/tabulate/>`__
-
 Talks
 ~~~~~
 

appveyor.yml

@@ -0,0 +1,53 @@
+# Python package
+# Create and test a Python package on multiple Python versions.
+# Add steps that analyze code, save the dist with the build record, publish to a PyPI-compatible index, and more:
+# https://docs.microsoft.com/azure/devops/pipelines/languages/python
+
+variables:
+  MPLBACKEND: Agg
+
+trigger:
+  branches:
+    include:
+    - '*'  # must quote since "*" is a YAML reserved character; we want a string
+
+jobs:
+
+- job: 'SkllTests'
+  pool:
+    vmImage: 'windows-latest'
+  strategy:
+    parallel: 5
+
+  steps:
+  - powershell: Write-Host "##vso[task.prependpath]$env:CONDA\Scripts"
+    displayName: Add conda to PATH
+
+  - script: |
+      conda update --quiet --yes conda
+      conda info -a
+    displayName: "Update conda"
+
+  - script: |
+      conda create --name sklldev --yes --quiet -c conda-forge -c defaults python=%PYTHON_VERSION% numpy=1.17 nose --file conda_requirements.txt
+      conda init cmd.exe
+      CALL activate sklldev
+      pip install -e .
+    displayName: 'Install dependencies & code'
+
+  - powershell: ./DistributeTests.ps1 
+    displayName: 'PowerShell Script to distribute tests'
+
+  - script: |
+      CALL activate sklldev
+      echo $(pytestfiles)
+      nosetests --with-xunit $(pytestfiles)
+    displayName: 'Run tests'
+    continueOnError: true
+
+  - task: PublishTestResults@2
+    displayName: 'Publish Test Results'
+    inputs:
+      testResultsFiles: 'nosetests.xml'
+      testRunTitle: 'SKLL tests'
+    condition: succeededOrFailed()

azure-pipelines.yml

@@ -1,7 +1,7 @@
 How to create and test conda package.
 
 1. To create the SKLL conda package run:
-   `conda build -c defaults -c conda-forge --python=3.6 --numpy=1.14 skll`
-2. Upload the package to anaconda.org using `anaconda upload <path>`.
+   `conda build -c defaults -c conda-forge --numpy=1.17 skll`
+2. Upload the package to anaconda.org using `anaconda upload --user ets <path>`.
 3. Test the package:
-   `conda create -n foobar -c defaults -c conda-forge -c desilinguist python=3.6 skll=1.5`
+   `conda create -n foobar -c conda-forge -c ets skll=2.0`

conda-recipe/README.md

@@ -1,15 +1,16 @@
 package:
   name: skll
-  version: 1.5.3
+  version: 2.0
 
 source:
   path: ../../../skll
 
 build:
   number: 0
+  noarch: python
   script:
     - cd $SRC_DIR
-    - $PYTHON setup.py install
+    - "{{ PYTHON }} -m pip install . --no-deps -vv"
   entry_points:
     - compute_eval_from_predictions = skll.utilities.compute_eval_from_predictions:main
     - filter_features = skll.utilities.filter_features:main
@@ -41,25 +42,29 @@ build:
 
 requirements:
   build:
-    - python
-    - scikit-learn ==0.20.1
-    - joblib >=0.8
-    - setuptools
-    - tabulate
     - beautifulsoup4
+    - joblib >=0.8
     - numpy
-    - scipy
+    - pandas
+    - python >=3.6
     - ruamel.yaml
+    - scikit-learn ==0.21.3
+    - scipy
+    - seaborn
+    - setuptools
+    - tabulate
 
   run:
-    - python
-    - scikit-learn ==0.20.1
-    - joblib >=0.8
-    - tabulate
     - beautifulsoup4
+    - joblib >=0.8
     - numpy
-    - scipy
+    - pandas
+    - python >=3.6
     - ruamel.yaml
+    - scikit-learn ==0.21.3
+    - seaborn
+    - scipy
+    - tabulate
 
 test:
   # Python imports
@@ -79,4 +84,6 @@ test:
 
 about:
   home: http://github.com/EducationalTestingService/skll
-  license: BSD 3-clause
+  license: BSD-3-Clause
+  license_file: LICENSE.txt
+  summary: SciKit-Learn Laboratory makes it easier to run machine learning experiments with scikit-learn.

conda-recipe/skll/meta.yaml

@@ -1,9 +1,9 @@
-scikit-learn==0.21.3
 beautifulsoup4
-tabulate
+joblib>=0.8
 numpy
-scipy
 pandas
-seaborn
-joblib>=0.8
 ruamel.yaml
+scikit-learn==0.21.3
+scipy
+seaborn
+tabulate

conda_requirements.txt

@@ -8,7 +8,7 @@ SKLL can be installed via ``pip``::
 
 or via ``conda``::
 
-    conda install -c conda-forge -c desilinguist python=3.6 skll
+    conda install -c conda-forge -c ets skll
 
 It can also be downloaded directly from
 `GitHub <https://github.com/EducationalTestingService/skll>`_.