y_data = np.zeros((m, 1))
x_data[:, 0] = 1
for i in range(m):
line_tempt = line_list[i].split(",")
for j in range(len(line_tempt)):
if j != len(line_tempt) - 1:
x_data[i, j + 1] = line_tempt[j]
else:
y_data[i, 0] = line_tempt[j]
# print(x_data)
# print(y_data)
file.close()
# description:x_data is a m*3 matrix, y_data is a vector
# loading end
# ignore the feature x0
x_data, mu, std = featureNormalize.feature_normalize(x_data)
# trick!!: normalize data before adding x0
# =======================Part2. Gradient Descent ===========================
# initialize some params
alpha = 0.01
num_iters = 5000
theta = np.zeros((3, 1))
# my gradientDescent support multi-variables
theta, cost_history = gradientDescent.gradient_descent(x_data, y_data, theta,
alpha, num_iters)
# print(theta)
# =======================Part3. Predict ================
input_data = np.array([[1, 1650, 3]], dtype=np.float64)
input_data[0, 1:] -= mu[1:]
input_data[0, 1:] /= std[1:]
print("predict price is {}".format(np.dot(input_data, theta)))
plt.axis([0, 13, 0, 150])
#plt.xticks(list(range(0,13,2)))
#plt.yticks(list(range(0,200,50)))
input('Program paused. Press ENTER to continue')
# ===================== Part 6 : Feature Mapping for Polynomial Regression =====================
# One solution to this is to use polynomial regression. You should now
# complete polyFeatures to map each example into its powers
#
p = 8
# Map X onto Polynomial Features and Normalize
X_poly = pf.poly_features(X, p)
X_poly, mu, sigma = fn.feature_normalize(X_poly)
X_poly = np.c_[np.ones(m), X_poly]
# Map X_poly_test and normalize (using mu and sigma)
X_poly_test = pf.poly_features(Xtest, p)
X_poly_test -= mu
X_poly_test /= sigma
X_poly_test = np.c_[np.ones(X_poly_test.shape[0]), X_poly_test]
# Map X_poly_val and normalize (using mu and sigma)
X_poly_val = pf.poly_features(Xval, p)
X_poly_val -= mu
X_poly_val /= sigma
X_poly_val = np.c_[np.ones(X_poly_val.shape[0]), X_poly_val]
print('Normalized Training Example 1 : \n{}'.format(X_poly[0]))
# plt.figure()
# plt.scatter(X[:, 0], X[:, 1], facecolors='none', edgecolors='b', s=20)
# plt.axis('equal')
# plt.axis([0.5, 6.5, 2, 8])
# plt.show()
input('Program paused. Press ENTER to continue')
# ===================== Part 2: Principal Component Analysis =====================
# You should now implement PCA, a dimension reduction technique. You
# should complete the code in pca.py
#
print('Running PCA on example dataset.')
# Before running PCA, it is important to first normalize X
X_norm, mu, sigma = fn.feature_normalize(X)
# Run PCA
U, S = pca.pca(X_norm)
# rk.draw_line(mu, mu + 1.5 * S[0] * U[:, 0])
# rk.draw_line(mu, mu + 1.5 * S[1] * U[:, 1])
# plt.show()
print('Top eigenvector: \nU[:, 0] = {}'.format(U[:, 0]))
print('You should expect to see [-0.707107 -0.707107]')
input('Program paused. Press ENTER to continue')
# ===================== Part 3: Dimension Reduction =====================
# You should now implement the projection step to map the data onto the
# first k eigenvectors. The code will then plot the data in this reduced
print('Loading data ...\n')
# Load Data
data = load_data('ex1data2.txt')
data = np.split(data, [2], axis=1)
X = data[0]
y = data[1]
m = len(y)
# Print out some data points
print('First 10 examples from the dataset: \n')
for i in range(10):
print(' x = [%.0f %.0f], y = %.0f \n' % (X[i][0], X[i][1], y[i]))
# pause_func()
# Scale features and set them to zero mean
print('Normalizing Features ...\n')
X, mu, sigma = feature_normalize(X)
# Add intercept term to X
X = np.append(np.ones((m, 1)), X, axis=1)
# ================ Part 2: Gradient Descent ================
print('Running gradient descent ...\n')
# Number of iterations (loops)
num_iters = 400
# Try some other values of alpha
alpha = 1
theta = np.zeros((3, 1))
theta, J_history_0 = gradient_descent_multi(X, y, theta, alpha, num_iters)
print('theta is \n', theta, '\n')
alpha = 0.3
theta = np.zeros((3, 1))
data = np.loadtxt('ex1data2.txt', delimiter=',')
X = data[:, :2]
y = data[:, 2]
m = y.size
# Print out some data points
print('First 10 examples from the dataset:')
for i in range(10):
print('x = {}, y = {}'.format(X[i], y[i]))
input('Program paused. Press enter to continue.\n')
# Scale features and set them to zero mean
print('Normalizing Features ...')
(X, mu, sigma) = featureNormalize.feature_normalize(X)
# Add intercept term to X
X = np.c_[np.ones(m), X]
# ================ Part 2: Gradient Descent ================
# ====================== YOUR CODE HERE ======================
# Instructions: We have provided you with the following starter
# code that runs gradient descent with a particular
# learning rate (alpha).
#
# Your task is to first make sure that your functions -
# computeCost and gradientDescent already work with
# this starter code and support multiple variables.
#
data = scio.loadmat('ex7data1.mat')
X = data['X'] #两个特征
# 可视化
plt.figure()
plt.scatter(X[:, 0], X[:, 1], facecolors='none', edgecolors='b', s=20)
plt.axis('equal')
plt.axis([0.5, 6.5, 2, 8])
input('Program paused. Press ENTER to continue')
'''第2部分 实现PCA 进行数据压缩'''
print('Running PCA on example dataset.')
# 在PCA之前 要对特征进行缩放
X_norm, mu, sigma = fn.feature_normalize(X)
# 执行PCA 返回U矩阵 和S矩阵
U, S = pca.pca(X_norm)
#对比两个不同的特征向量 U[:,0]更好 投影误差最小 U中的各个特征向量（列）都是正交的 2D->1D 取前1个特征向量 作为Ureduce
rk.draw_line(mu, mu + 1.5 * S[0] * U[:, 0])
rk.draw_line(mu, mu + 1.5 * S[1] * U[:, 1])
print('Top eigenvector: \nU[:, 0] = {}'.format(
U[:, 0])) #利用PCA得到的特征向量矩阵Ureduce（降维后子空间的基）
print('You should expect to see [-0.707107 -0.707107]')
input('Program paused. Press ENTER to continue')
'''第3部分 得到降维后的样本点 再进行压缩重放'''
print('Dimension reductino on example dataset.')