X = X_input.reshape((147,50))
#X_norm = feature_normalize(X_input).reshape((294,25))
#X = X_norm
'''
class_names = ['man','hand','eye']
X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size=0.3)
svm_model_linear = SVC(kernel = 'rbf', C = 1).fit(X_train, y_train)
svm_predictions = svm_model_linear.predict(X_test)
accuracy = svm_model_linear.score(X_test, y_test)
cm = confusion_matrix(y_test, svm_predictions)
matrix_plot.plot_confusion_matrix(y_test, svm_predictions, classes=class_names,
title='Confusion matrix, without normalization')
matrix_plot.plot_confusion_matrix(y_test, svm_predictions, classes=class_names, normalize=True,
title='confusion matrix of svm model generated from combined data of session 1 and session 2')
plt.show()
print("Results: ", svm_predictions)
print("True Values: ", y_test)
print("accuracy: ", accuracy*100)
# summarize history for accuracy
plt.plot(history.history['acc'])
plt.plot(history.history['val_acc'])
plt.title('CNN Accuracy on Augmented Data')
plt.ylabel('Accuracy')
plt.xlabel('Epoch')
plt.legend(['Train', 'Test'], loc='upper left')
plt.show()
# summarize history for loss
plt.plot(history.history['loss'])
plt.plot(history.history['val_loss'])
plt.title('CNN Accuracy on Augmented Data')
plt.ylabel('Loss')
plt.xlabel('Epoch')
plt.legend(['Train', 'Test'], loc='upper left')
plt.show()
y_predict = model.predict_classes(x_test)
for i in range(y_predict.shape[0]):
print("predicted Y: ", y_predict[i], " expected Y: ", np.argmax(y_test[i]))
class_names = ['eye', 'man', 'hand']
matrix_plot.plot_confusion_matrix(
np.argmax(y_test, axis=1),
y_predict,
classes=class_names,
normalize=True,
title='CNN Confusion Matrix on Augmented Data')
plt.show()
test_size=0.2,
random_state=4)
x_test, x_dev, y_test, y_dev = train_test_split(x_val,
y_val,
test_size=0.5,
random_state=4)
dtree_model = DecisionTreeClassifier(max_depth=6).fit(x_train, y_train)
dtree_predictions = dtree_model.predict(x_test)
cm = confusion_matrix(y_test, dtree_predictions)
# Plot normalized confusion matrix
matrix_plot.plot_confusion_matrix(
y_test,
dtree_predictions,
classes=class_names,
normalize=True,
title='Decision Tree Confusion Matrix on Original Data')
plt.show()
count = 0
for i in range(y_test.shape[0]):
if y_test[i] == dtree_predictions[i]:
count += 1
print("Train Count: ", y_train.shape)
print("============================")
print("Test Count: ", y_test.shape)
print("============================")
print("Results: ", dtree_predictions)
print("============================")
class_names = [0, 1, 2]
X_train, X_test, y_train, y_test = train_test_split(X, Y)
svm_model_linear = SVC(kernel='rbf', C=1).fit(X_train, y_train)
svm_predictions = svm_model_linear.predict(X_test)
accuracy = svm_model_linear.score(X_test, y_test)
cm = confusion_matrix(y_test, svm_predictions)
# Plot non-normalized confusion matrix
matrix_plot.plot_confusion_matrix(
y_test,
svm_predictions,
classes=class_names,
title='Confusion matrix, without normalization')
# Plot normalized confusion matrix
matrix_plot.plot_confusion_matrix(
y_test,
svm_predictions,
classes=class_names,
normalize=True,
title='confusion matrix of svm model generated from session 1 data')
plt.show()
print("Train Count: ", y_train.shape)
print("============================")
print("Test Count: ", y_test.shape)
sentence, label = line.split(".")
label = label.strip().lower()
detection = pass_act_detect(sentence)
if detection:
y_pred.append("P")
else:
y_pred.append("A") # false means active
if label == "(passive)":
y_true.append("P")
elif label == "(active)":
y_true.append("A") # false means active
else:
print("something went wrong")
print("prediction completed, generating confusion matrix")
cnf_matrix = confusion_matrix(y_true, y_pred)
np.set_printoptions(precision=2)
plt.figure()
plot_confusion_matrix(cnf_matrix,
classes=["Active", "Passive"],
title='Confusion matrix, without normalization')
plt.figure()
plot_confusion_matrix(cnf_matrix,
classes=["Active", "Passive"],
N=True,
title='Normalized confusion matrix')
plt.show()
x_test, x_dev, y_test, y_dev = train_test_split(x_val,
y_val,
test_size=0.5,
random_state=4)
svm_model_linear = SVC(kernel='rbf', C=1).fit(x_train, y_train)
svm_predictions = svm_model_linear.predict(x_test)
accuracy = svm_model_linear.score(x_test, y_test)
cm = confusion_matrix(y_test, svm_predictions)
matrix_plot.plot_confusion_matrix(
y_test,
svm_predictions,
classes=class_names,
normalize=True,
title='SVM Confusion Matrix on Augmented Data')
plt.show()
print("Train Count: ", y_train.shape)
print("============================")
print("Test Count: ", y_test.shape)
print("============================")
print("Results: ", svm_predictions)
print("============================")
print("True Values: ", y_test)
print("============================")
print("accuracy: ", accuracy * 100)
print("============================")
Y[50:] = Y[50:]*2
'''
class_names = ['man', 'hand', 'eye']
X_train, X_test, y_train, y_test = train_test_split(X, Y)
dtree_model = DecisionTreeClassifier(max_depth=4).fit(X_train, y_train)
dtree_predictions = dtree_model.predict(X_test)
cm = confusion_matrix(y_test, dtree_predictions)
# Plot normalized confusion matrix
matrix_plot.plot_confusion_matrix(y_test,
dtree_predictions,
classes=class_names,
normalize=True,
title='Decision Tree From Session 1 Data')
plt.show()
count = 0
for i in range(y_test.shape[0]):
if y_test[i] == dtree_predictions[i]:
#print(dtree_predictions[i])
#print(y_test[i])
count += 1
#print(count)
#print(count)
#print(y_test.shape[0])
