Merge pull request #54 from simai-ml/ci_github

Ci GitHub

Author: vtaquet

.appveyor.yml

@@ -1,4 +1,4 @@
-name: Publish Python Package to PyPi
+name: Publish to PyPi
 
 on:
   release:

.github/workflows/python-publish.yml renamed to .github/workflows/publish.yml

@@ -0,0 +1,48 @@
+name: Unit tests
+
+on: [push, pull_request]
+
+jobs:
+  build:
+    runs-on: ${{ matrix.os }}
+    strategy:
+      matrix:
+        include:
+          - os: ubuntu-latest
+            python-version: 3.7
+            numpy-version: 1.18.5
+          - os: ubuntu-latest
+            python-version: 3.8
+            numpy-version: 1.19.5
+          - os: ubuntu-latest
+            python-version: 3.9
+            numpy-version: 1.20.3
+          - os: windows-latest
+            python-version: 3.7
+            numpy-version: 1.18.5
+          - os: windows-latest
+            python-version: 3.8
+            numpy-version: 1.19.5
+          - os: windows-latest
+            python-version: 3.9
+            numpy-version: 1.20.3
+    defaults:
+      run:
+        shell: bash -l {0}
+
+    steps:
+    - name: Git clone
+      uses: actions/checkout@v2
+    - name: Set up virtual environment
+      uses: conda-incubator/setup-miniconda@v2
+      with:
+        python-version: ${{ matrix.python-version }}
+        channels: defaults, conda-forge
+    - name: Install dependencies
+      run: conda install codecov mypy numpy=${{ matrix.numpy-version }} scikit-learn pandas pytest-cov
+    - name: Check static typing
+      run: mypy mapie examples --strict
+    - name: Test with pytest
+      run: pytest -vs --doctest-modules --cov-branch --cov=mapie --pyargs mapie
+    - name: Code coverage
+      run: codecov

.github/workflows/test.yml

@@ -2,6 +2,12 @@
 History
 =======
 
+0.2.2 (2021-06-XX)
+------------------
+
+* Switch to github actions for continuous integration of the code
+* Add image explaining MAPIE internals on the README
+
 0.2.1 (2021-06-04)
 ------------------
 

.travis.yml

@@ -1,12 +1,9 @@
 .. -*- mode: rst -*-
 
-|Travis|_ |AppVeyor|_ |Codecov|_ |CircleCI|_ |ReadTheDocs|_ |License|_ |PythonVersion|_ |PyPi|_
+|GitHubActions|_ |Codecov|_ |CircleCI|_ |ReadTheDocs|_ |License|_ |PythonVersion|_ |PyPi|_
 
-.. |Travis| image:: https://travis-ci.com/simai-ml/MAPIE.svg?branch=master
-.. _Travis: https://travis-ci.com/simai-ml/MAPIE
-
-.. |AppVeyor| image:: https://ci.appveyor.com/api/projects/status/js4d7km6ckr801nj/branch/master?svg=true
-.. _AppVeyor: https://ci.appveyor.com/project/gmartinonQM/mapie
+.. |GitHubActions| image:: https://github.com/simai-ml/MAPIE/actions/workflows/test.yml/badge.svg
+.. _GitHubActions: https://github.com/simai-ml/MAPIE/actions
 
 .. |Codecov| image:: https://codecov.io/gh/simai-ml/MAPIE/branch/master/graph/badge.svg?token=F2S6KYH4V1
 .. _Codecov: https://codecov.io/gh/simai-ml/MAPIE
@@ -96,7 +93,7 @@ and two standard deviations from the mean.
 
 
 
-MAPIE returns a ``np.ndarray`` of shape (n_samples, 3, len(alpha)) giving the predictions,
+MAPIE returns a ``np.ndarray`` of shape ``(n_samples, 3, len(alpha))`` giving the predictions,
 as well as the lower and upper bounds of the prediction intervals for the target quantile
 for each desired alpha value.
 The estimated prediction intervals can then be plotted as follows. 
@@ -146,7 +143,18 @@ The effective coverage is the actual fraction of true labels lying in the predic
 📘 Documentation
 ================
 
-The documentation can be found `on this link <https://mapie.readthedocs.io/en/latest/>`_.
+How does **MAPIE** works ? It is basically based on cross-validation and relies on:
+
+- Residuals on the whole trainig set obtained by cross-validation,
+- Perturbed models generated during the cross-validation.
+
+**MAPIE** then combines all these elements in a way that provides prediction intervals on new data with strong theoretical guarantees [1].
+
+.. image:: https://github.com/simai-ml/MAPIE/raw/master/doc/images/mapie_internals.png
+    :width: 400
+    :align: center
+
+The full documentation can be found `on this link <https://mapie.readthedocs.io/en/latest/>`_.
 It contains the following sections:
 
 - `Quickstart <https://mapie.readthedocs.io/en/latest/quick_start.html>`_
@@ -193,10 +201,10 @@ with the financial support from Région Ile de France.
 🔍  References
 ==============
 
-MAPIE methods are based on the work by `Foygel-Barber et al. (2020) <https://www.stat.uchicago.edu/~rina/jackknife.html>`_.
+MAPIE methods are based on the work by `Foygel-Barber et al. (2021) <https://doi.org/10.1214/20-AOS1965>`_.
 
 [1] Rina Foygel Barber, Emmanuel J. Candès, Aaditya Ramdas, and Ryan J. Tibshirani.
-Predictive inference with the jackknife+. Ann. Statist., 49(1):486–507, 022021
+"Predictive inference with the jackknife+." Ann. Statist., 49(1):486–507, February 2021.
 
 📝 License
 ==========

HISTORY.rst

@@ -1,5 +1,6 @@
 # Release checklist
 
+- [ ] Update the version number with `bump2version major|minor|patch`
 - [ ] Edit HISTORY.rst and AUTHORS.rst to make sure it’s up-to-date and add release date
 - [ ] Check whether any new files need to go in MANIFEST.in
 - [ ] Make sure tests run, pass and cover 100% of the package:
@@ -16,7 +17,6 @@
     * `rm -rf build dist MAPIE.egg-info`
     * `python setup.py sdist bdist_wheel`
 - [ ] Check that your package is ready for publication: `twine check dist/*`
-- [ ] Update the version number with `bump2version major|minor|patch`
 - [ ] Make sure everything is committed and pushed: `git push origin master`
 - [ ] Push new tag to your commit: `git push --tags`
 - [ ] Upload it to TestPyPi: `twine upload --repository-url https://test.pypi.org/legacy/ dist/*`

README.rst

@@ -215,4 +215,4 @@ References
 ==========
 
 [1] Rina Foygel Barber, Emmanuel J. Candès, Aaditya Ramdas, and Ryan J. Tibshirani.
-Predictive inference with the jackknife+. Ann. Statist., 49(1):486–507, 022021
+"Predictive inference with the jackknife+." Ann. Statist., 49(1):486–507, February 2021.