具有 2d 坐标列表 + 分类值的决策树分类器答案

【问题标题】：Decision tree classifier with list of 2d coordinates + categorical value具有 2d 坐标列表 + 分类值的决策树分类器
【发布时间】：2018-04-06 16:01:57
【问题描述】：

我有一个带有特征和标签的 numpy 数组。

每个数据点由 10 个二维坐标 (x,y) 和一个指示方向的字符串（“左”或“右”）组成。

例子：

[array([[-19.24181754,  -0.6933614 ],
       [-17.39631579,  -0.84320702],
       [-14.57501754,  12.99707368],
       [ -8.6202386 ,   4.90138246],
       [  0.82478596,  20.01929825],
       [  4.79946667, -10.70312982],
       [  7.10694035,  17.47812632],
       [ 11.06254737,  14.17312982],
       [ 17.04467368,   0.19169825],
       [ 18.94181053,   6.92687018]])
  'left']]

标签是 12 个不同的字符串 ('4-4-2', '5-3-2', ...)。我想用这些数据尝试不同的算法来比较它们的性能。第一个算法是决策树分类器。

我看到了两个潜在的问题：

以 10 个点为特征的数组
分类数据

对于第二点，one-hot-coding 应该可以解决问题。在 numpy 数组中是否有一种非常简单的方法可以做到这一点？

对于第一点，我不确定这是否是一个问题，因为我到目前为止还没有尝试过。

编辑：

我的编码：

#Separate features and labels
X = test[:, [0, 4]]
Y = test[:,10]

zeros = np.zeros((len(X), 2), dtype=int)

X = np.append(X, zeros, axis=1)

for datapoint in X:
    if(datapoint[1] == 'left'):
        datapoint[2] = 1
    else:
        datapoint[3] = 1

X = np.delete(X, 1, 1)

#Divide into test and training data: 80% training, 20% test
X_train, X_test, y_train, y_test = train_test_split(X,Y, test_size=0.2, random_state=100)

#Initialize classifier
clf_gini = DecisionTreeClassifier(criterion = "gini", random_state = 100,
                               max_depth=3, min_samples_leaf=5)

#Train classifier
clf_gini.fit(X_train, y_train)

错误：

  "This module will be removed in 0.20.", DeprecationWarning)
Backend TkAgg is interactive backend. Turning interactive mode on.
Traceback (most recent call last):
  File "C:\Program Files\JetBrains\PyCharm Community Edition 2017.3.2\helpers\pydev\pydevd.py", line 1668, in <module>
    main()
  File "C:\Program Files\JetBrains\PyCharm Community Edition 2017.3.2\helpers\pydev\pydevd.py", line 1662, in main
    globals = debugger.run(setup['file'], None, None, is_module)
  File "C:\Program Files\JetBrains\PyCharm Community Edition 2017.3.2\helpers\pydev\pydevd.py", line 1072, in run
    pydev_imports.execfile(file, globals, locals)  # execute the script
  File "C:\Program Files\JetBrains\PyCharm Community Edition 2017.3.2\helpers\pydev\_pydev_imps\_pydev_execfile.py", line 18, in execfile
    exec(compile(contents+"\n", file, 'exec'), glob, loc)
  File "C:/Users/D071947/PycharmProjects/Formation/DecisionTree.py", line 41, in <module>
    clf_gini.fit(X_train, y_train)
  File "C:\Users\D071947\PycharmProjects\Formation\venv\lib\site-packages\sklearn\tree\tree.py", line 790, in fit
    X_idx_sorted=X_idx_sorted)
  File "C:\Users\D071947\PycharmProjects\Formation\venv\lib\site-packages\sklearn\tree\tree.py", line 116, in fit
    X = check_array(X, dtype=DTYPE, accept_sparse="csc")
  File "C:\Users\D071947\PycharmProjects\Formation\venv\lib\site-packages\sklearn\utils\validation.py", line 433, in check_array
    array = np.array(array, dtype=dtype, order=order, copy=copy)
ValueError: setting an array element with a sequence.

我猜它与二维坐标数组有关？

【问题讨论】：

您可以将 2D 特征的 10 维数组表示为 20 个特征。

标签： python machine-learning decision-tree

【解决方案1】：

问题确实是嵌套的坐标数组。就像 Cardelling 所建议的那样，将 10 个 2d 坐标表示为 20 个功能起作用。

为了展平我使用的列表numpy.ravel()

【讨论】：