Uploaded public code and data

animals/users.csv

@@ -0,0 +1,24 @@
+initial_yn_questions,id
+879,6
+890,4
+889,15
+886,3
+880,8
+902,16
+911,13
+903,19
+924,18
+927,7
+908,11
+881,14
+922,27
+858,17
+895,23
+888,20
+885,21
+896,12
+920,2
+868,26
+906,10
+887,22
+940,5

catalina/users.csv

@@ -0,0 +1,20 @@
+initial_yn_questions,id
+868,19
+883,15
+899,18
+883,11
+879,10
+872,21
+913,2
+912,7
+899,5
+873,20
+885,9
+919,12
+925,8
+905,17
+883,16
+902,6
+880,4
+856,13
+880,14

custom_metrics.py

@@ -0,0 +1,74 @@
+import numpy as np
+from matplotlib import pyplot as plt
+from sklearn.metrics import confusion_matrix
+from sklearn.metrics import accuracy_score
+from sklearn.metrics import precision_score
+from sklearn.metrics import recall_score
+from sklearn.metrics import f1_score
+
+class ListTable(list):
+    def _repr_html_(self):
+        html = ["<table>"]
+        for row in self:
+            html.append("<tr>")
+
+            for col in row:
+                html.append("<td>{0}</td>".format(col))
+
+            html.append("</tr>")
+        html.append("</table>")
+        return ''.join(html)
+
+def draw_theta(theta, labels=None):
+    """Draw the confusion matrix received.
+    
+    Parameters
+    ----------
+    theta : np.array
+        confusion matrix drawn
+    """     
+
+    if labels is None:
+        labels = []
+
+    row_sums = theta.sum(axis=1)
+    new_matrix = theta / row_sums[:, np.newaxis]
+    print(np.round(new_matrix,3))
+
+    fig = plt.figure()
+    ax = fig.add_subplot(111)
+    res = ax.imshow(new_matrix,  cmap=plt.cm.Blues, interpolation='nearest')
+    plt.title('Confusion matrix\n')
+    ticks = [i for i in range(len(theta))]
+    ax.set_xticks(ticks)
+    ax.set_yticks(ticks)
+    ax.set_xticklabels(labels)
+    ax.set_yticklabels(labels)
+    plt.ylabel('True Class')
+    plt.xlabel('Predicted Class')
+    plt.show()
+
+def draw_metrics(y_true, y_pred, labels):
+    """
+    This method is only for to draw metrics
+    
+    """
+    draw_theta(confusion_matrix(y_true, y_pred) , labels)
+    table = ListTable()
+    print("Acurracy score",accuracy_score(y_true, y_pred))
+    table.append(["Metrics / Classes"]+list(labels))
+    table.append(["Precision score"]+list(np.round(precision_score(y_true, y_pred, average=None),3)))
+    table.append(["Recall score"]+list(np.round(recall_score(y_true, y_pred, average=None),3)))
+    table.append(["F score"]+list(np.round(f1_score(y_true, y_pred, average=None),3)))
+    return table
+
+def metrics(y_true, y_pred):
+
+    to_ret = {
+            'p':precision_score(y_true, y_pred, average=None),
+            'r':recall_score(y_true, y_pred, average=None),
+            'f':f1_score(y_true, y_pred, average=None),
+            }
+
+    return to_ret
+