scikit-learn:如何缩减'y'预测结果

Question

我正在尝试scikit-learn使用Boston Housing Data Set 学习和机器学习.

# I splitted the initial dataset ('housing_X' and 'housing_y')
from sklearn.cross_validation import train_test_split
X_train, X_test, y_train, y_test = train_test_split(housing_X, housing_y, test_size=0.25, random_state=33)

# I scaled those two datasets
from sklearn.preprocessing import StandardScaler
scalerX = StandardScaler().fit(X_train)
scalery = StandardScaler().fit(y_train)
X_train = scalerX.transform(X_train)
y_train = scalery.transform(y_train)
X_test = scalerX.transform(X_test)
y_test = scalery.transform(y_test)

# I created the model
from sklearn import linear_model
clf_sgd = linear_model.SGDRegressor(loss='squared_loss', penalty=None, random_state=42) 
train_and_evaluate(clf_sgd,X_train,y_train)

基于这个新模型clf_sgd,我试图y基于第一个实例预测X_train.

X_new_scaled = X_train[0]
print (X_new_scaled)
y_new = clf_sgd.predict(X_new_scaled)
print (y_new)

然而,结果对我来说很奇怪(1.34032174而不是20-30房子的价格范围)

[-0.32076092  0.35553428 -1.00966618 -0.28784917  0.87716097  1.28834383
  0.4759489  -0.83034371 -0.47659648 -0.81061061 -2.49222645  0.35062335
 -0.39859013]
[ 1.34032174]

我想这个1.34032174值应该缩减,但我试图找出如何做到这一点没有成功.欢迎任何提示.非常感谢你.

Answer 1

您可以使用inverse_transform您的scalery对象:

y_new_inverse = scalery.inverse_transform(y_new)

Answer 2

游戏有点晚了：只是不要缩放你的 y。通过缩放 y，您实际上会失去您的单位。回归或损失优化实际上是由特征之间的相对差异决定的。顺便说一句，对于房价（或任何其他货币价值），通常的做法是取对数。然后你显然需要做一个 numpy.exp() 来回到实际的美元/欧元/日元......