Backprop working

functions.py

@@ -219,9 +219,6 @@ class Network:
             self.n_layers - 1: (dw, delta)
         }
 
-        # update weights and biases
-        self.update_w_b(self.n_layers - 1, dw, delta)
-
         # In case of three layer net will iterate over i = 2 and i = 1
         # Determine partial derivative and delta for the rest of the layers.
         # Each iteration requires the delta from the previous layer, propagating backwards.
@@ -244,7 +241,7 @@ class Network:
         self.w[index] -= self.learning_rate * np.mean(dw, 1)
         self.b[index] -= self.learning_rate * np.mean(np.mean(delta, 1), 0)
 
-    def fit(self, x, y_true, loss, epochs, batch_size, learning_rate=1e-3):
+    def fit(self, x, y_true, loss, epochs, batch_size, learning_rate=2e-2):
         """
         :param loss: Loss class (MSE, CrossEntropy etc.)
         """
@@ -267,9 +264,13 @@ class Network:
                 z, a = self.feed_forward(x_[k:l])
                 self.back_prop(z, a, y_[k:l])
 
-            if (i + 1) % 100 == 0:
+            if (i + 1) % 10 == 0:
                 print("Loss:", self.loss.loss(y_true, z[self.n_layers]))
 
+    def predict(self, x):
+        _, a = self.feed_forward(x)
+        return a[self.n_layers]
+
 if __name__ == "__main__":
     from sklearn import datasets
     #import sklearn.metrics
@@ -283,7 +284,21 @@ if __name__ == "__main__":
     # one hot encoding
     y = np.eye(3)[y]
 
-    nn = Network((4, 8, 3), (Relu, Sigmoid))
+    nn = Network((4, 8, 2, 3), (Relu, Relu, Sigmoid))
 
-    nn.fit(x[:2], y[:2], MSE, 1, batch_size=2)
-    #nn.fit(x, y, MSE, 10000, 16)
+    #nn.fit(x[:2], y[:2], MSE, 1, batch_size=2)
+    nn.fit(x, y, MSE, 1000, 16)
+    y_ = nn.predict(x)
+    a = np.argmax(y_, 1)
+
+    for i in range(a.size):
+        print(a[i], y[i])
+
+
+    # y_true = []
+    # y_pred = []
+    # for i in range(len(y)):
+    #     y_pred.append(np.argmax(y_[3][i]))
+    #     y_true.append(np.argmax(y[i]))
+    #
+    # print(sklearn.metrics.classification_report(y_true, y_pred))

simple_mlp.py

@@ -135,7 +135,7 @@ class NeuralNetwork:
 
         # partial derivative with respect to layer 1
         delta2 = np.dot(delta3, self.w[2].T) * diff_relu(self.z[2])
-        print(self.w[2].T)
+
         # dc_db1 = delta2
         dc_dw1 = np.dot(self.x.T, delta2)
 
@@ -194,7 +194,10 @@ if __name__ == "__main__":
     y = np.eye(3)[y]
 
     nn = NeuralNetwork(4, 8, 3, 2e-2)
-    nn.fit(x[:2], y[:2], 2, 1)
+    #nn.fit(x[:2], y[:2], 2, 1)
+    nn.fit(x, y, 16, 1000)
+    _, y_ = feed_forward(x, nn.w, nn.b)
+    print(y_)
 
     # # result
     # _, y_ = feed_forward(x, nn.w, nn.b)