Blanche Mahe assignment

numpy_questions.py

@@ -41,6 +41,18 @@ def max_index(X):
     j = 0
 
     # TODO
+    if not isinstance(X, np.ndarray):
+        raise ValueError
+    if X.ndim != 2:
+        raise ValueError
+
+    max = X[i, j]
+    for index_row in range(X.shape[0]):
+        for index_col in range(X.shape[1]):
+            if X[index_row, index_col] > max:
+                max = X[index_row, index_col]
+                i = index_row
+                j = index_col
 
     return i, j
 
@@ -64,4 +76,10 @@ def wallis_product(n_terms):
     """
     # XXX : The n_terms is an int that corresponds to the number of
     # terms in the product. For example 10000.
-    return 0.
+    if n_terms == 0:
+        return 1
+
+    pi_over_2 = 1
+    for n in range(1, n_terms+1):
+        pi_over_2 *= (4*(n**2))/(4*(n**2)-1)
+    return 2*pi_over_2

sklearn_questions.py

@@ -29,28 +29,50 @@
 
 
 class OneNearestNeighbor(BaseEstimator, ClassifierMixin):
-    "OneNearestNeighbor classifier."
+    """OneNearestNeighbor classifier."""
 
     def __init__(self):  # noqa: D107
         pass
 
     def fit(self, X, y):
-        """Write docstring.
-
-        And describe parameters
+        """Fit the OneNearestNeighbor classifier on training data.
+
+        Parameters
+        ----------
+        self : defines the instance of the class OneNearestNeighbor we are
+        working on
+        X : ndarray of the training data,
+        with shape (n_observations, p_features)
+        y : 1-darray of the labels associated with each dimension of X,
+        with shape (n_observations)
+
+        Returns
+        -------
+        self : maintains the instance
         """
         X, y = check_X_y(X, y)
         check_classification_targets(y)
         self.classes_ = np.unique(y)
         self.n_features_in_ = X.shape[1]
 
         # XXX fix
+        self.X_ = X
+        self.y_ = y  # store X and y as "learned" data,
+        #  ensure we have trained on X and y
         return self
 
     def predict(self, X):
-        """Write docstring.
+        """Predict y label for input data X using the OneNearestNeighbor rule.
+
+        Parameters
+        -------
+        self : still maintain the instance of the class
+        X : ndarray of test data, with shape (n_observations, p_features)
 
-        And describe parameters
+        Returns
+        -------
+        y_pred : 1-darray of labels predicted for the test data X,
+        with shape (n_observations,)
         """
         check_is_fitted(self)
         X = check_array(X)
@@ -60,15 +82,31 @@ def predict(self, X):
         )
 
         # XXX fix
+        idx = 0
+        for x in X:
+            euclidean_distances = np.sqrt(np.sum((self.X_ - x) ** 2, axis=1))
+            NN_index = np.argmin(euclidean_distances)
+            NN = self.y_[NN_index]
+            y_pred[idx] = NN
+            idx += 1
         return y_pred
 
     def score(self, X, y):
-        """Write docstring.
-
-        And describe parameters
+        """Score model performance by evaluating proportion of accurate y_pred.
+
+        Parameters
+        -------
+        X : ndarray of test data, with shape (n_observations, p_features)
+        y : 1d-array of the true labels associated with test samples X,
+        with shape (n_observations,)
+
+        Returns
+        -------
+        a score : float in [0,1]
+        reflecting the proportion of accurate predictions
         """
         X, y = check_X_y(X, y)
         y_pred = self.predict(X)
 
         # XXX fix
-        return y_pred.sum()
+        return sum(y_pred == y)/len(y)

students.txt

@@ -60,7 +60,7 @@ Liard	Eléanor
 Liu	Guangyue    X 
 Liu	Yunxian     
 Lucille	Maximilien X     
-Mahé	Blanche     
+Mahé	Blanche     X
 Martin	Justin     X
 Massias	Mathurin     
 Massoud	Alexandre.....X