Question

我有以下代码用于绘制散点图：

plt.scatter(step1.values[step1['y_ocsvm'] == 1, 2], step1.values[step1['y_ocsvm'] == 1, 1], c = 'green', label = 'Normal')
plt.scatter(step1.values[step1['y_ocsvm'] == -1, 2], step1.values[step1['y_ocsvm'] == -1, 1], c = 'red', label = 'Outlier')
plt.scatter(step2.values[step2['y_ocsvm'] == 1, 2], step2.values[step2['y_ocsvm'] == 1, 1], c = 'green')
plt.scatter(step2.values[step2['y_ocsvm'] == -1, 2], step2.values[step2['y_ocsvm'] == -1, 1], c = 'red')

step1和step2是两个不同的数据帧。

假设我们有许多这样的不同数据帧，我想知道哪种更紧凑的方式编写上述代码？

我尝试过的事情：

我尝试创建可用数据帧名称的列表，并尝试遍历它们，但是会引发错误：

alldfs = [var for var in dir() if isinstance(eval(var), pd.core.frame.DataFrame)]
matches = [i for i in alldfs if 'step' in i]

for i,df in enumerate(matches):
    print(matches[i])
    plt.scatter(matches[i].values[matches[i]['y_ocsvm'] == 1, 2], matches[i].values[matches[i]['y_ocsvm'] == 1, 1], c = 'green', label = 'Normal')
    plt.scatter(matches[i].values[matches[i]['y_ocsvm'] == -1, 2], matches[i].values[matches[i]['y_ocsvm'] == -1, 1], c = 'red', label = 'Outlier')

它引发的错误：

alldfs = [var for var in dir() if isinstance(eval(var), pd.core.frame.DataFrame)]
matches = [i for i in alldfs if 'step' in i]

for i,df in enumerate(matches):
    print(matches[i])
    plt.scatter(matches[i].values[matches[i]['y_ocsvm'] == 1, 2], matches[i].values[matches[i]['y_ocsvm'] == 1, 1], c = 'green', label = 'Normal')
    plt.scatter(matches[i].values[matches[i]['y_ocsvm'] == -1, 2], matches[i].values[matches[i]['y_ocsvm'] == -1, 1], c = 'red', label = 'Outlier')
step1
Traceback (most recent call last):

  File "<ipython-input-51-e73fcf40a967>", line 3, in <module>
    plt.scatter(matches[i].values[matches[i]['y_ocsvm'] == 1, 2], matches[i].values[matches[i]['y_ocsvm'] == 1, 1], c = 'green', label = 'Normal')

AttributeError: 'str' object has no attribute 'values'

编辑1：

step1数据框

ContextID   BacksGas_Flow_sccm  StepID  y_ocsvm
7289973          0.5               1       1
7289973     0.333333333            1       1
7289973     0.416666667            1       1
7289973     0.416666667            1       1
7289999     0.333333333            1       1
7289999     0.333333333            1       1
7289999         0.25               1       1
7289999     0.416666667            1       1
7290650     0.333333333            1       1
7290650     0.166666667            1       -1
7290690     0.666666667            1       -1
7290690     0.583333333            1       1

step2数据框

ContextID   BacksGas_Flow_sccm  StepID  y_ocsvm
7289973       0.777777778         2       -1
7289973       0.777777778         2       -1
7289973       0.666666667         2       1
7289973       0.444444444         2       1
7289999       0.444444444         2       1
7289999       0.555555556         2       1
7290025       0.777777778         2       -1
7290025       0.666666667         2       1
7290650       0.444444444         2       1
7290650       0.888888889         2       -1
7290650       0.777777778         2       -1
7290650       0.555555556         2       1

如何遍历不同的数据帧以更紧凑的方式绘制散点图？

0 个答案: