我正在從 Udacity 課程中學(xué)習(xí)回歸，"Intro to Machine Learning"并且正在完成其中的迷你項(xiàng)目。我復(fù)制了項(xiàng)目的代碼，編寫了回歸部分，作為代碼的一部分，并嘗試在 Python 2 中的 Jupyter Notebook 上運(yùn)行它。不幸的是，我收到了以下錯(cuò)誤。    ImportError                               Traceback (most recent call last)    <ipython-input-8-433cee6c5e25> in <module>()         15 import pickle         16 sys.path.append("../tools/")    ---> 17 from feature_format import featureFormat, targetFeatureSplit         18 dictionary = pickle.load( open("../final_project/final_project_dataset_modified.pkl", "r") )         19     ImportError: No module named feature_format我正在處理的代碼如下：加載/格式化數(shù)據(jù)集的修改版本（為什么要修改？我們已經(jīng)刪除了一些麻煩點(diǎn)，你會(huì)發(fā)現(xiàn)自己在異常值迷你項(xiàng)目中）。繪制一個(gè)小散點(diǎn)圖訓(xùn)練/測(cè)試數(shù)據(jù)n您在指示處填寫回歸代碼：    #!/usr/bin/python    import sys    import pickle    sys.path.append("../tools/")    from feature_format import featureFormat, targetFeatureSplit    dictionary = pickle.load( open("../final_project/final_project_dataset_modified.pkl", "r") )    ### list the features you want to look at--first item in the     ### list will be the "target" feature    features_list = ["bonus", "salary"]    data = featureFormat( dictionary, features_list, remove_any_zeroes=True)    target, features = targetFeatureSplit( data )    ### training-testing split needed in regression, just like classification    from sklearn.cross_validation import train_test_split    feature_train, feature_test, target_train, target_test = train_test_split(features, target, test_size=0.5, random_state=42)    train_color = "b"    test_color = "r"    ### Your regression goes here!    ### Please name it reg, so that the plotting code below picks it up and     ### plots it correctly. Don't forget to change the test_color above from "b" to    ### "r" to differentiate training points from test points.    from sklearn import linear_model    reg = linear_model.LinearRegression()    reg.fit(feature_train,target_train)