def test_model(self,X,Y):
""" Test using specific model's eval function. """
if hasattr(self.model, "evals"):
labels = np.array(Y)
p_labels = self.model.evals(X)
success =np.sum(labels == p_labels)
total_count = len(X)
else:
labels = [] # List of actual labels
p_labels = [] # List of model's predictions
success = 0 # Number of correct predictions
total_count = 0 # Number of images
for i in range(len(X)):
x = X[i] # Test input
y = Y[i] # Actual label
y_ = self.model.eval(x) # Model's prediction
labels.append(y)
p_labels.append(y_)
if y == y_:
success += 1
total_count +=1
return 1.0*success/total_count, getConfusionMatrix(labels,p_labels)
def test_model(self, X, Y):
""" Test using specific model's eval function. """
labels = [] # List of actual labels
p_labels = [] # List of model's predictions
success = 0.0 # Number of correct predictions
total_count = 0.0 # Number of images
for i in range(len(X)):
x = X[i] # Test input
y = Y[i] # Actual label
y_ = self.model.eval(x) # Model's prediction
labels.append(y)
p_labels.append(y_)
if y == y_:
success += 1.0
total_count += 1.0
return success / total_count, getConfusionMatrix(labels, p_labels)