Code update and results

preprocessing.py

@@ -9,6 +9,12 @@ def get_data(data_file):
     return data
 
 
+def get_data_with_id(data_file):
+    data = pd.read_csv(data_file)
+
+    return data
+
+
 # imputation
 def get_average_column_value(data, col):
     total = 0

random_forest.py

@@ -14,7 +14,7 @@ def get_train_test(data):
     labels = np.where(labels == 2, 0, labels)
     labels = np.where(labels == 4, 1, labels)
     train, test, train_labels, test_labels = train_test_split(data, labels,
-                                                              test_size=0.3, random_state=42)
+                                                              test_size=0.3, random_state=1)
     return [train, test, train_labels.astype(int), test_labels.astype(int)]
 
 
@@ -259,13 +259,44 @@ def all_combos_accuracy(removed_data, average_data):
 
         print(test_accuracy)
 
+
+def get_results(model, X, y, result_doc, ids):
+    result = open(result_doc, 'w+')
+    result.write('samplecodenumber,trueclass,predclass\n')
+
+    loaded_model = pickle.load(open(model, 'rb'))
+    predictions = loaded_model.predict(X)
+
+    for i in range(len(predictions)):
+        if y[i] == 2:
+            Y = 0
+        else:
+            Y = 1
+
+        result.write(str(ids[i]))
+        result.write(',')
+        result.write(str(Y))
+        result.write(',')
+        result.write(str(predictions[i]))
+        result.write('\n')
+
+    result.close()
+
 
 if __name__ == '__main__':
     original, removed, average = preprocessing.preprocess()
+    with_id = preprocessing.get_data_with_id('breast-cancer-wisconsin.data')
     # train_all_combos(removed, average)
     # all_combos_accuracy(removed, average)
     # all_combos_roc(removed, average)
-    investigate(removed, average)
+    # investigate(removed, average)
+
+    X = average[average.columns[:-1]]
+    y = average[average.columns[-1]].values
+
+    get_results('models/average_gini_normal_80.model', X, y, 'randomforest_result.txt', with_id['ID'].values)
+
+