将另一个类导入python类

Question

我希望能够将类A导入到类B中，并使用类B中的函数。如下所示：

from scipy import stats
import numpy as np

class TestLearner(object):

    def __init__(self, param1 = 2):
        self.param = param1

    def train_model(self, X, y):
        slope, intercept, r_value, p_value, std_err = stats.linregress(X,y)
        self.slope = slope
        self.intercept = intercept

    def predict_new(self, X):
        y = self.intercept + self.slope * X
        return y

class Test(object):

    def __init__(self, learner):
        self.learner = learner

    def add_evidence(self, X, y):
        learner = self.learner
        slope, intercept = learner.train_model(X,y)
        self.slope = slope
        self.intercept = intercept

    def predict(self, X):
        y= self.intercept * self.slope * X
        return y

将TestLearner作为t导入

x_train = np.random.randn(10,3)
y_train = np.random.randn(10,)

x_test = np.random.randn(5,3)

testing = Test(learner = t.TestLearner)
testing.add_evidence(X=x_train,y= y_train)

我得到以下错误：

Traceback (most recent call last):
  File "<input>", line 40, in <module>
  File "<input>", line 26, in add_evidence
TypeError: train_model() missing 1 required positional argument: 'y'

我不明白。火车模型似乎在寻找“自我”作为输入，我认为这在TestLearner类中是自我包含的

Answer 1

为了在类中调用方法，您需要将该类初始化为实例，您可以在测试构造函数中初始化实例，但这仅在您将另一个类而不是实例传递到Test（）时有效

class Test(object):

    def __init__(self, learner):
        self.learner = learner()

    def add_evidence(self, X, y):
        learner = self.learner
        slope, intercept = learner.train_model(X,y)
        self.slope = slope
        self.intercept = intercept

    def predict(self, X):
        y= self.intercept * self.slope * X
        return y