Merge pull request #23 from UBC-MDS/r2_test

R2 test

docs/example.ipynb

@@ -2,28 +2,43 @@
     "cells": [
         {
             "cell_type": "markdown",
+            "metadata": {},
             "source": [
                 "# Example usage\n",
                 "\n",
                 "To use `matrics_calculator` in a project:"
-            ],
-            "metadata": {}
+            ]
         },
         {
             "cell_type": "code",
-            "execution_count": null,
+            "execution_count": 1,
+            "metadata": {},
+            "outputs": [
+                {
+                    "name": "stdout",
+                    "output_type": "stream",
+                    "text": [
+                        "0.1.0\n"
+                    ]
+                }
+            ],
             "source": [
                 "import matrics_calculator\n",
                 "\n",
                 "print(matrics_calculator.__version__)"
-            ],
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": null,
+            "metadata": {},
             "outputs": [],
-            "metadata": {}
+            "source": []
         }
     ],
     "metadata": {
         "kernelspec": {
-            "display_name": "Python 3",
+            "display_name": "matrics_calculator",
             "language": "python",
             "name": "python3"
         },
@@ -37,9 +52,9 @@
             "name": "python",
             "nbconvert_exporter": "python",
             "pygments_lexer": "ipython3",
-            "version": "3.8.5"
+            "version": "3.10.16"
         }
     },
     "nbformat": 4,
     "nbformat_minor": 4
-}
+}

src/matrics_calculator/r2.py

@@ -0,0 +1,57 @@
+# r-squared value calculation
+def r2(predictor, response): 
+    """
+    Calculates r-squared using linear regression.
+
+    Computes the r-squared value (coefficient of determination) using the provided predictor 
+    list and response list.
+    
+    Parameters
+    ----------
+    predictor : list
+        Predictor values to be used in calculating r-sqaured value.
+    response : list
+        Response values to be used in calculating r-sqaured value.
+
+    Returns
+    -------
+    float
+        r-sqaured value which is <= 1. 1 is the best score and a score below 0 is worse than 
+        using the mean of the target as predictions.
+
+    Examples
+    --------
+    data = {
+    'math_test': [80, 85, 90, 95],
+    'science_test': [78, 82, 88, 92],
+    'final_grade': [84, 87, 91, 94],
+    'absences': [3, 0, 1, 30]
+    }
+    >>> r2(data['math_test'],data['final_grade'])
+    0.997
+    >>> r2(data['math_test'],data['absences'])
+    0.541
+    """
+    from sklearn.linear_model import LinearRegression
+    import numpy as np
+
+    if not isinstance(predictor, list) or not isinstance(predictor, list):
+        print('Input must be lists')
+        return None
+
+    if len(predictor) == 0 or len(response) == 0:
+        print('Input cannot be empty')
+        return None
+
+    if isinstance(predictor,list):
+        predictor = np.array(predictor)
+    if isinstance(response,list):
+        response = np.array(response)
+
+    model = LinearRegression()
+    model.fit(predictor.reshape(-1,1),response)
+    response_predicted =  model.predict(predictor.reshape(-1,1))
+    response_mean = np.mean(response)
+    RSS = sum(((response-response_predicted) ** 2))
+    TSS = sum(((response-response_mean) ** 2))
+    return round(1 - (RSS/TSS),3)

tests/test_r2.py

@@ -0,0 +1,17 @@
+from matrics_calculator.r2 import r2
+
+def test_r2():
+    data = {
+    'math_test': [80, 85, 90, 95],
+    'science_test': [78, 82, 88, 92],
+    'final_grade': [84, 87, 91, 94],
+    'absences': [3, 0, 1, 30]
+    }
+
+    assert r2(1, 0) == None
+
+    assert r2([],[1,2,3]) == None
+
+    assert r2(data['math_test'], data['final_grade']) == 0.997
+
+    assert r2(data['math_test'], data['absences']) == 0.541