F/implement-own-add-constant-func

.gitignore

@@ -2,4 +2,5 @@ __pycache__
 .pytest_cache
 .ruff_cache
 .vscode
-pytest-coverage.txt
+pytest-coverage.txt
+.coverage

CONTRIBUTING.md

@@ -1,40 +1,41 @@
 # Contributing
 
-## Development 
+## Development
 
 ### Setting up a development environment
 
 If you don't have a local development environment, you can follow these steps to set one up.
 
-Install [poetry](https://python-poetry.org/) and [task](https://taskfile.dev/).
+Install [poetry](https://python-poetry.org/).
 
 Now, initialize the project:
 
 ```bash
-task init
+make init
 ```
 
 ### Running tests
 
 You can run the tests with:
 
 ```bash
-task tests
+make tests
 ```
 
 This will run the tests with [pytest](https://docs.pytest.org/en/latest/) and show information about the coverage.
 
 ### Formatting the code
 
 To look for formatting issues:
+
 ```bash
-task check-formatting
+make check-formatting
 ```
 
 To format the code, you can use the command:
 
 ```bash
-task formatting
+make formatting
 ```
 
 ### Releasing a new version
@@ -45,4 +46,4 @@ To release a new version, you need to follow these steps:
 
 2. Create a Github release with the new version number.
 
-3. (Optional) Publish the new version to PyPI with `poetry publish --build`.
+3. (Optional) Publish the new version to PyPI with `poetry publish --build`.

Makefile

@@ -0,0 +1,16 @@
+
+.PHONY: init
+init:
+	poetry install -n
+
+.PHONY: tests
+tests:
+	poetry run pytest --cov-report=term-missing:skip-covered --cov=pier_ds_utils tests/ | tee pytest-coverage.txt
+
+.PHONY: check-formatting
+check-formatting:
+	poetry run ruff check .
+
+.PHONY: formatting
+formatting:
+	poetry run ruff format .

pier_ds_utils/estimator.py

@@ -5,6 +5,7 @@
 import statsmodels.api as sm
 
 from pier_ds_utils.transformer import BaseCustomTransformer
+from pier_ds_utils.prep import add_constant_column
 from sklearn.base import BaseEstimator
 
 
@@ -24,7 +25,9 @@ def get_params(self, deep=True):
 
     def fit(self, X, y, **fit_params):
         if self._add_constant:
-            X = sm.add_constant(X)
+            X = add_constant_column(
+                X, prepend=True, constant_value=1.0, column_name="const"
+            )
 
         self.model_ = sm.GLM(endog=y, exog=X, **self.init_params)
         fit_method = fit_params.pop("fit_method", "fit")
@@ -33,7 +36,9 @@ def fit(self, X, y, **fit_params):
 
     def predict(self, X, **predict_params):
         if self._add_constant:
-            X = sm.add_constant(X)
+            X = add_constant_column(
+                X, prepend=True, constant_value=1.0, column_name="const"
+            )
 
         return self.results_.predict(exog=X, **predict_params) * self.os_factor
 

pier_ds_utils/prep.py

@@ -0,0 +1,49 @@
+import pandas as pd
+import typing
+
+
+def add_constant_column(
+    df: typing.Union[pd.DataFrame, pd.Series],
+    prepend: bool = True,
+    column_name: str = "const",
+    constant_value: typing.Any = 1.0,
+) -> pd.DataFrame:
+    """
+    Add a constant column to a DataFrame.
+
+    Parameters
+    ----------
+    df : pd.DataFrame
+        The DataFrame to which the constant column will be added.
+    prepend : bool, default=True
+        If True, the constant column will be added as the first column.
+        If False, the constant column will be added as the last column.
+    column_name : str, default="const"
+        The name of the constant column.
+    constant_value : Any, default=1.0
+        The value to be used for the constant column.
+
+    Returns
+    -------
+    pd.DataFrame
+        The DataFrame with the constant column added.
+    """
+    if not isinstance(df, pd.DataFrame):
+        raise ValueError("df must be a DataFrame or Series")
+
+    df = df.copy()
+
+    original_column_names = df.columns.tolist()
+
+    df[column_name] = constant_value
+
+    if column_name in original_column_names:
+        original_column_names.remove(column_name)
+
+    column_names = (
+        [column_name] + original_column_names
+        if prepend
+        else original_column_names + [column_name]
+    )
+
+    return df[column_names]

pyproject.toml

@@ -1,12 +1,12 @@
 [tool.poetry]
 name = "pier-ds-utils"
-version = "0.3.0"
+version = "0.4.0"
 description = "A library for data science teams to avoid code duplication in common tasks involving scikit-learn transformers and estimators."
-authors = ["caiodallaqua <[email protected]>", "Gabriel Guarisa <[email protected]>"]
+authors = ["Gabriel Guarisa <[email protected]>"]
 readme = "README.md"
 
 [tool.poetry.dependencies]
-python = ">=3.8,<3.12"
+python = ">=3.9,<3.12"
 setuptools = "^69.0.3"
 scikit-learn = "*"
 pandas = "*"
@@ -17,6 +17,7 @@ pytest = "^7.4.4"
 boto3 = "^1.34.19"
 awswrangler = "^3.5.1"
 ruff = "^0.2.0"
+pytest-cov = "^6.0.0"
 
 [build-system]
 requires = ["poetry-core"]