Tensorflow线性回归结果与Numpy / SciKit-Learn不匹配

时间:2018-02-05 00:16:59

标签: python numpy tensorflow scikit-learn linear-regression

我正在研究Aurelien Geron的“动手机器学习”一书中的Tensorflow示例。但是,我无法复制this supporting notebook中的简单线性回归示例。为什么Tensorflow不匹配Numpy / SciKit-Learn结果?

据我所知,没有优化(我们使用的是正规方程式,所以它只是矩阵计算),而且答案似乎与精度误差不同。

import numpy as np
import tensorflow as tf
from sklearn.datasets import fetch_california_housing

housing = fetch_california_housing()
m, n = housing.data.shape
housing_data_plus_bias = np.c_[np.ones((m, 1)), housing.data]

X = tf.constant(housing_data_plus_bias, dtype=tf.float32, name="X")
y = tf.constant(housing.target.reshape(-1, 1), dtype=tf.float32, name="y")
XT = tf.transpose(X)
theta = tf.matmul(tf.matmul(tf.matrix_inverse(tf.matmul(XT, X)), XT), y)

with tf.Session() as sess:
    theta_value = theta.eval()

theta_value

答案:

array([[ -3.74651413e+01],
       [  4.35734153e-01],
       [  9.33829229e-03],
       [ -1.06622010e-01],
       [  6.44106984e-01],
       [ -4.25131839e-06],
       [ -3.77322501e-03],
       [ -4.26648885e-01],
       [ -4.40514028e-01]], dtype=float32)

######与纯NumPy比较

X = housing_data_plus_bias
y = housing.target.reshape(-1, 1)
theta_numpy = np.linalg.inv(X.T.dot(X)).dot(X.T).dot(y)

print(theta_numpy)

答案:

[[ -3.69419202e+01]
 [  4.36693293e-01]
 [  9.43577803e-03]
 [ -1.07322041e-01]
 [  6.45065694e-01]
 [ -3.97638942e-06]
 [ -3.78654265e-03]
 [ -4.21314378e-01]
 [ -4.34513755e-01]]

######与Scikit-Learn比较

from sklearn.linear_model import LinearRegression
lin_reg = LinearRegression()
lin_reg.fit(housing.data, housing.target.reshape(-1, 1))

print(np.r_[lin_reg.intercept_.reshape(-1, 1), lin_reg.coef_.T])

答案:

[[ -3.69419202e+01]
 [  4.36693293e-01]
 [  9.43577803e-03]
 [ -1.07322041e-01]
 [  6.45065694e-01]
 [ -3.97638942e-06]
 [ -3.78654265e-03]
 [ -4.21314378e-01]
 [ -4.34513755e-01]]

更新:我的问题听起来与this one类似,但根据建议并没有解决问题。

1 个答案:

答案 0 :(得分:3)

我只是按tensorflownumpy比较结果。由于您对dtype=tf.float32X使用了y,因此我会将np.float32用于numpy示例,如下所示:

X_numpy = housing_data_plus_bias.astype(np.float32)
y_numpy = housing.target.reshape(-1, 1).astype(np.float32)

现在让我们尝试按tf.matmul(XT, X)tensorflow)和X.T.dot(X)numpy)比较结果:

with tf.Session() as sess:
    XTX_value = tf.matmul(XT, X).eval()
XTX_numpy = X_numpy.T.dot(X_numpy)

np.allclose(XTX_value, XTX_numpy, rtol=1e-06) # True
np.allclose(XTX_value, XTX_numpy, rtol=1e-07) # False

所以这是float精度的问题。如果您将精度更改为tf.float64np.float64,则theta的结果会相同。