Merge branch 'joss' into paper

.github/actions/publish/action.yaml

@@ -65,12 +65,23 @@ inputs:
 runs:
   using: composite
   steps:
+    - name: display inputs
+      run: |
+        echo ${{ inputs.BRANCH }}
+      shell: bash
+
     - name: Checkout Repository
       uses: actions/checkout@v3
       with:
         ref: ${{ inputs.BRANCH }}
         token: ${{ inputs.ACCESS_TOKEN }}
 
+    - name: check install
+      run: |
+        git status
+        git branch
+      shell: bash
+
     - name: Install poetry
       run: pip install poetry${{ inputs.POETRY_VERSION != '' && format('=={0}', inputs.POETRY_VERSION) || '' }} poetry-core${{ inputs.POETRY_CORE_VERSION != '' && format('=={0}', inputs.POETRY_CORE_VERSION) || '' }}
       shell: bash
@@ -84,8 +95,14 @@ runs:
 
     - name: Set GitHub Tag as Package Version
       run: |
-        sed -i -r 's/__version__ *= *".*"/__version__ = "${{ github.event.release.tag_name }}"/g' ${{ inputs.PACKAGE_DIRECTORY }}__init__.py
-        sed -i '0,/version =.*/s//version = "'"${{ github.event.release.tag_name }}"'"/' ./pyproject.toml
+        if ${{ inputs.UPDATE_CODE_VERSION }}
+        then
+          vname="${{ github.event.release.tag_name }}"
+          vname=${vname:1}
+          echo $vname
+          sed -i -r 's/__version__ *= *".*"/__version__ = "'"$vname"'" /g' ${{ inputs.PACKAGE_DIRECTORY }}__init__.py
+          sed -i '0,/version =.*/s//version = "'"$vname"'"/' ./pyproject.toml
+        fi
       shell: bash
 
     - name: Add and Commit Version

.github/workflows/release_pypi.yaml

@@ -19,9 +19,20 @@ jobs:
           python-version: '3.10' 
 
       - name: Install from TestPyPI & run tests with installed package
+        id: install
         run: |
           python -m pip install --index-url https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple/ choice-learn
-          python tests/manual_run.py
+          cd ../
+          echo ${{ github.event.release.tag_name }}
+          python choice-learn/tests/manual_run.py
+          cd choice-learn
+          echo "b"
+          git fetch --all
+          BRANCH=$(git branch --list -r "origin/release_*" | tr '*' ' ')
+          echo $BRANCH
+          BRANCH="${BRANCH:9}"
+          echo $BRANCH
+          echo "BRANCH=$BRANCH" >> $GITHUB_OUTPUT
 
       - name: publish to PyPI
         uses: ./.github/actions/publish
@@ -32,5 +43,5 @@ jobs:
           PUBLISH_REGISTRY_PASSWORD: ${{ secrets.PYPI_PASSWORD }}
           PUBLISH_REGISTRY_USER: ${{ secrets.PYPI_USERNAME }}
           UPDATE_CODE_VERSION: false
-          BRANCH:  release_${{ github.event.release.tag_name }}
+          BRANCH: ${{ steps.install.outputs.BRANCH }}
 

README.md

@@ -1,9 +1,9 @@
 <div align="center">
 
 
-<img src="docs/illustrations/logos/logo_choice_learn.png" width="300">
+<img src="https://github.com/artefactory/choice-learn/blob/main/docs/illustrations/logos/logo_choice_learn.png" width="300">
 
-*Large-scale choice modeling through the lens of machine learning*
+*Large-scale choice modelling through the lens of machine learning*
 
 [![CI status](https://github.com/artefactory/choice-learn/actions/workflows/ci.yaml/badge.svg)](https://github.com/artefactory/choice-learn/actions/workflows/ci.yaml?query=branch%3Amain)
 [![Linting , formatting, imports sorting: ruff](https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/charliermarsh/ruff/main/assets/badge/v2.json)](https://github.com/astral-sh/ruff)

choice_learn/__init__.py

@@ -1,4 +1,4 @@
 """Choice-Learn library for choice modeling in Python."""
 
-__version__ = "0.0.2"
+__version__ = "0.0.4"
 __author__ = "Vincent Auriau"

choice_learn/data/storage.py

@@ -52,10 +52,67 @@ def __str__(self):
         return f"FeatureStorage with name {self.name}"
 
 
-class FeaturesStorage(Storage):
+class FeaturesStorage(object):
+    """Base FeaturesStorage class that redirects toward the right one."""
+
+    def __new__(
+        cls,
+        ids=None,
+        values=None,
+        values_names=None,
+        name=None,
+        as_one_hot=False,
+    ):
+        """Redirects toward the right object.
+
+        Parameters
+        ----------
+        ids : Iterable, optional
+            IDs to references features with, by default None
+        values : Iterable, optional
+            Features to be stored, by default None
+        values_names : list, optional
+            List of names for the features to be stored, by default None
+        name : str, optional
+            Name of the FeaturesStorage, to be matched in ChoiceDataset, by default None
+        as_one_hot: bool
+            Whether features are OneHot representations or not.
+
+        Returns
+        -------
+        FeaturesStorage
+            One of ArrayStorage, DictStorage or OneHotStorage
+        """
+        if as_one_hot:
+            return OneHotStorage(ids=ids, values=values, name=name)
+
+        if ids is None and (isinstance(values, np.ndarray) or isinstance(values, list)):
+            return ArrayStorage(values=values, values_names=values_names, name=name)
+
+        if ids is not None:
+            check_ids = np.unique(ids) == np.arange(len(ids))
+            if isinstance(check_ids, np.ndarray):
+                check_ids = check_ids.all()
+            if check_ids:
+                values = [values[np.where(np.array(ids) == i)[0][0]] for i in np.arange(len(ids))]
+                return ArrayStorage(values=values, values_names=values_names, name=name)
+
+        return DictStorage(
+            ids=ids, values=values, values_names=values_names, name=name, indexer=StorageIndexer
+        )
+
+
+class DictStorage(Storage):
     """Function to store features with ids."""
 
-    def __init__(self, ids=None, values=None, values_names=None, name=None, indexer=StorageIndexer):
+    def __init__(
+        self,
+        ids=None,
+        values=None,
+        values_names=None,
+        name=None,
+        indexer=StorageIndexer,
+    ):
         """Build the store.
 
         Parameters
@@ -70,6 +127,7 @@ def __init__(self, ids=None, values=None, values_names=None, name=None, indexer=
         name: string, optional
             name of the features store
         """
+        print("DictStorage")
         if isinstance(values, dict):
             storage = values
             lengths = []
@@ -169,6 +227,91 @@ def batch(self):
         return self.indexer
 
 
+class ArrayStorage(Storage):
+    """Function to store features with ids as NumPy Array."""
+
+    def __init__(self, values=None, values_names=None, name=None):
+        """Build the store.
+
+        Parameters
+        ----------
+        values : array_like
+            list of values of features to store
+        values_names : array_like
+            Iterable of str indicating the name of the features. Must be same length as values.
+        name: string, optional
+            name of the features store
+        """
+        if isinstance(values, list):
+            storage = np.array(values)
+        elif not isinstance(values, np.ndarray):
+            raise ValueError("ArrayStorage Values must be a list or a numpy array")
+
+        # self.storage = storage
+        self.values_names = values_names
+        self.name = name
+
+        self.shape = storage.shape
+        self.indexer = storage
+
+    def get_element_from_index(self, index):
+        """Getter method over self.sequence.
+
+        Returns the features stored at index index. Compared to __getitem__, it does take
+        the index-th element of sequence but the index-th element of the store.
+
+        Parameters
+        ----------
+        index : (int, list, slice)
+            index argument of the feature
+
+        Returns
+        -------
+        array_like
+            features corresponding to the index index in self.store
+        """
+        return self.batch[index]
+
+    def __len__(self):
+        """Return the length of the sequence of apparition of the features."""
+        return self.shape[0]
+
+    def __getitem__(self, id_keys):
+        """Subset FeaturesStorage, keeping only features which id is in keys.
+
+        Parameters
+        ----------
+        id_keys : Iterable
+            List of ids to keep.
+
+        Returns
+        -------
+        FeaturesStorage
+            Subset of the FeaturesStorage, with only the features whose id is in id_keys
+        """
+        if not isinstance(id_keys, list):
+            id_keys = [id_keys]
+        return ArrayStorage(
+            values=self.batch[id_keys], values_names=self.values_names, name=self.name
+        )
+
+    def get_storage_type(self):
+        """Functions to access stored elements dtypes.
+
+        Returns
+        -------
+        tuple
+            tuple of dtypes of the stored elements, as returned by np.dtype
+        """
+        element = self.get_element_from_index(0)
+        return element.dtype
+
+    @property
+    def batch(self):
+        """Indexing attribute."""
+        return self.indexer
+
+
 class OneHotStorage(Storage):
     """Specific Storage for one hot features storage.
 

docs/paper/paper.bib

@@ -89,7 +89,6 @@ @Article{Harris:2020
  volume        = {585},
  number        = {7825},
  pages         = {357--362},
- doi           = {10.1038/s41586-020-2649-2},
  publisher     = {Springer Science and Business Media {LLC}},
 }