removed random_weights and substituted with np.random.uniform

Author: dmeoli

deep_learning4e.py

@@ -82,7 +82,7 @@ def __init__(self, in_size=3, out_size=3, activation=Sigmoid):
         self.activation = activation()
         # initialize weights
         for node in self.nodes:
-            node.weights = random_weights(-0.5, 0.5, in_size)
+            node.weights = np.random.uniform(-0.5, 0.5, in_size)
 
     def forward(self, inputs):
         self.inputs = inputs

learning4e.py

@@ -567,7 +567,7 @@ def __init__(self, l_rate=0.01, epochs=1000, optimizer='bfgs'):
 
     def fit(self, X, y):
         loss = MeanSquaredError(X, y)
-        self.w = minimize(fun=loss.function, x0=np.zeros((X.shape[1], 1)),method=self.optimizer, jac=loss.jacobian).x
+        self.w = minimize(fun=loss.function, x0=np.zeros((X.shape[1], 1)), method=self.optimizer, jac=loss.jacobian).x
         return self
 
     def predict(self, example):

tests/test_learning4e.py

@@ -123,16 +123,6 @@ def test_random_forest():
     assert grade_learner(rf, tests) >= 1 / 3
 
 
-def test_random_weights():
-    min_value = -0.5
-    max_value = 0.5
-    num_weights = 10
-    test_weights = random_weights(min_value, max_value, num_weights)
-    assert len(test_weights) == num_weights
-    for weight in test_weights:
-        assert min_value <= weight <= max_value
-
-
 def test_ada_boost():
     iris = DataSet(name='iris')
     iris.classes_to_numbers()

utils4e.py

@@ -357,10 +357,6 @@ def normalize(dist):
     return [(n / total) for n in dist]
 
 
-def random_weights(min_value, max_value, num_weights):
-    return [random.uniform(min_value, max_value) for _ in range(num_weights)]
-
-
 def conv1D(x, k):
     """1D convolution. x: input vector; K: kernel vector."""
     return np.convolve(x, k, mode='same')