plt.show()
# plot the mode of responsibilities
modes = model.map_grids[responsibilities.argmax(axis=1), :]
plt.figure()
# plt.figure(figsize=figure.figaspect(1))
plt.scatter(modes[:, 0], modes[:, 1], c=ytrain)
plt.colorbar()
plt.ylim(-1.1, 1.1)
plt.xlim(-1.1, 1.1)
plt.xlabel("z1 (mode)")
plt.ylabel("z2 (mode)")
plt.show()
# construct MLR model
model.mlr(Xtrain, ytrain)
# MLR prediction
predicted_ytest = np.ndarray.flatten(model.mlr_predict(Xtest))
# r2p, RMSEp, MAEp
print("r2p: {0}".format(float(1 - sum((ytest - predicted_ytest) ** 2) / sum((ytest - ytest.mean()) ** 2))))
print("RMSEp: {0}".format(float((sum((ytest - predicted_ytest) ** 2) / len(ytest)) ** (1 / 2))))
print("MAEp: {0}".format(float(sum(abs(ytest - predicted_ytest)) / len(ytest))))
# yy-plot
plt.figure(figsize=figure.figaspect(1))
plt.scatter(ytest, predicted_ytest)
YMax = np.max(np.array([np.array(ytest), predicted_ytest]))
YMin = np.min(np.array([np.array(ytest), predicted_ytest]))
plt.plot([YMin - 0.05 * (YMax - YMin), YMax + 0.05 * (YMax - YMin)],
[YMin - 0.05 * (YMax - YMin), YMax + 0.05 * (YMax - YMin)], 'k-')
plt.xlabel("z1 (mean)")
plt.ylabel("z2 (mean)")
plt.show()
# plot the mode of responsibilities
modes = model.map_grids[responsibilities.argmax(axis=1), :]
plt.figure()
# plt.figure(figsize=figure.figaspect(1))
plt.scatter(modes[:, 0], modes[:, 1], c=original_y)
plt.colorbar()
plt.ylim(-1.1, 1.1)
plt.xlim(-1.1, 1.1)
plt.xlabel("z1 (mode)")
plt.ylabel("z2 (mode)")
plt.show()
# construct MLR model
model.mlr(original_X, original_y)
# inverse analysis
estimated_x_mean, estimated_x_mode, responsibilities_inverse = model.inverse_gtm_mlr(autoscaled_target_y_value)
estimated_x_mean = estimated_x_mean * original_X.std(axis=0, ddof=1) + original_X.mean(axis=0)
estimated_x_mode = estimated_x_mode * original_X.std(axis=0, ddof=1) + original_X.mean(axis=0)
# print("estimated x-mean: {0}".format(estimated_x_mean))
print("estimated x-mode: {0}".format(estimated_x_mode))
estimated_x_mean_on_map = responsibilities_inverse.dot(model.map_grids)
estimated_x_mode_on_map = model.map_grids[np.argmax(responsibilities_inverse), :]
# print("estimated x-mean on map: {0}".format(estimated_x_mean_on_map))
print("estimated x-mode on map: {0}".format(estimated_x_mode_on_map))
