x_m, y_sd, X_train = ef.normalize(X_train) # Normalizes the data, extracting mean and SD of training data for normalizing the testing data later.
clf = SVC(kernel = 'linear', C = 0.14, random_state = seed, probability = True)
clf.fit(X_train, y_train)
# Finding the Training Accuracy:
correct = 0
total = 0
# ==================================
# TESTING PART:
# ------------
# Extracting Testing Data:
(X_test, y_test) = ef.fetch_test()
x_m, y_sd, X_test = ef.normalize(X_test, xm = x_m, ysd = y_sd) # Normalizes the testing data with the mean and SD of the training set.
# Fitting the classifier:
print '\nUsing %s estimators of depth %s.\n' % (str(num_est), str(tree_depth))
# Deriving the ROC curve:
y_check = clf.predict(X_test)
y_hat = clf.predict_proba(X_test)
y_hat = array([ entry[1] for entry in y_hat ])
fpr, tpr, thresholds = roc_curve(y_test, y_hat, pos_label=1)
# Plotting the result:
plt.plot(fpr, tpr)
plt.plot([0,1], [0,1], 'k--')
plt.xlabel('False Positive Rate')
print('Epoch:', epoch, 'Loss:', loss)
Losses.append(loss.item())
# Zero gradients, perform a backward pass, and update the weights.
optimizer.zero_grad()
# perform a backward pass (backpropagation)
for i in range(6):
list(model.parameters())[i].retain_grad()
loss.backward()
# Update the parameters
optimizer.step()
# Testing the model
x_test, y_test = ef.fetch_test()
x_ts = Variable(torch.Tensor(x_test).type(torch.FloatTensor))
y_ts = torch.Tensor([i for i in y_test]).type(torch.LongTensor)
y_pred = model(x_ts)
labels = y_ts.detach().numpy()
decs_1 = y_pred[:, 1].detach().numpy()
print('Labels:', labels.shape)
print('Decisions:', decs_1.shape)
fpr, tpr, thr = roc_curve(labels, decs_1)
# Plotting the result:
plt.plot(fpr, tpr)
plt.plot([0, 1], [0, 1], 'k--')
plt.xlabel('False Positive Rate')
plt.ylabel('True Positive Rate')