epochs logic changed - old commit

classifier/linear_model.py

@@ -81,7 +81,7 @@ class MultiClassPerceptron:
 
      This model also shuffles the training data after each epoch.
     """
-    def __init__(self, epochs: int = 5000, learning_rate: float = 1.0, random_state: int = 42):
+    def __init__(self, epochs: int = 5, learning_rate: float = 1.0, random_state: int = 42):
         """
         :type epochs: int
         :type learning_rate: float
@@ -127,20 +127,15 @@ class MultiClassPerceptron:
             sample_weights = get_sample_weights_with_features(theta_bias=-0.25, random_state=self.random_state)
             self.perceptron_dict[label] = Perceptron(label, sample_weights, theta_bias=-0.25)
 
-        next_print = int(self.epochs/10)
-
-        random.seed(self.random_state)
-        random_list = [random.randint(0, train_len-1) for i in range(self.epochs)]
-
         # Training Iterations
         for epoch in range(self.epochs):
 
-            if epoch >= next_print:
+            print('Training Epoch :: (', (epoch+1), '/', self.epochs, ')')
-                print('Training Multi-Class Perceptron Classifier..... (', epoch, '/', self.epochs, ')')
+
-                next_print = next_print + int(self.epochs/10)
+            for i in range(train_len):
 
                 # Pick a number from random list
-            inst = X_train[random_list[epoch]]
+                inst = X_train[i]
 
                 perceptron_scores = []  # list for storing perceptron scores for each label
                 for label, perceptron in self.perceptron_dict.items():

testing/model_testing.py

@@ -18,10 +18,10 @@ labels = set([inst.true_label for inst in X_train_inst])
 X_test_inst = read_csv_file(test_file_path, '\t')
 
 # number of training iterations
-epochs = int(len(X_train_inst)*1.5)
+epochs = 50
 
 # create MultiClassPerceptron classifier object
-clf = MultiClassPerceptron(epochs=epochs, learning_rate=0.75, random_state=101)
+clf = MultiClassPerceptron(epochs=epochs, learning_rate=0.5, random_state=101)
 
 # train the model
 clf.fit(X_train=X_train_inst, labels=list(labels))