：使用Twinx()进行通用图形绘制

Question

我正试图使用matplotlib开发一个相当通用的图表，但始终得到一个错误的ValueError: left cannot be >= right。我的代码是：

def perf_plot(x, y, data, title = ''):
    import matplotlib.pyplot as plt
    import statsmodels.api as sm
    plt.style.use('ggplot')

    df = data

    if y == 'slope':
        z = 'bp01'
    else: 
        z = 'slope'

    y = df[y].astype(float)
    x = df[x].astype(float)
    z = df[z].astype(float)

    blue = '#348EA9'
    orange = '#F48B37'
    green = '#52BA9B'
    red = '#EF4846'

    fig, ax1 = plt.subplots()
    fig.suptitle(title, fontsize=14, fontweight='bold')
    fig.subplots_adjust(top=0.85)
    ax1.scatter(x ,y, color = orange)

    lowess_y = sm.nonparametric.lowess(y, x ,frac=0.1)
    ax1.plot(lowess_y[:, 0], lowess_y[:, 1], color = blue)
    ax1.set_ylabel(y, color = blue)
    ax1.set_xlabel(x, color = 'b')

    ax2 = ax1.twinx()
    lowess_z = sm.nonparametric.lowess(z, x, frac=0.1)
    ax2.plot(lowess_z[:, 0], lowess_z[:, 1], color = green)
    ax2.set_ylabel(z, color = green)

    fig.tight_layout()
    plt.show()
    return

在googling之后，我发现这个错误通常与tight_layout()和标题的使用相关。移除标题并不能解决问题，但是如果我删除了tight_layout，图形就会打印，但两个轴都放在图形的左侧。因为重点是做一些像这，我不是真正的如何获得没有tight_layout的决斗轴。想法？

错误码：

perf_plot(x_, y_, df1)
Traceback (most recent call last):

  File "<ipython-input-149-36ff57804093>", line 1, in <module>
    perf_plot(x_, y_, df1)

  File "<ipython-input-148-ef2f325b64f9>", line 37, in perf_plot
    fig.tight_layout()

  File "C:\Users\user\AppData\Local\Continuum\Anaconda2\lib\site-packages\matplotlib\figure.py", line 1755, in tight_layout
    self.subplots_adjust(**kwargs)

  File "C:\Users\user\AppData\Local\Continuum\Anaconda2\lib\site-packages\matplotlib\figure.py", line 1620, in subplots_adjust
    self.subplotpars.update(*args, **kwargs)

  File "C:\Users\user\AppData\Local\Continuum\Anaconda2\lib\site-packages\matplotlib\figure.py", line 228, in update
    raise ValueError('left cannot be >= right')

ValueError: left cannot be >= right

Answer 1

这个错误是由

y = df[y].astype(float)
x = df[x].astype(float)
z = df[z].astype(float)

哪里

ax1.set_ylabel(y, color = blue)
ax1.set_xlabel(x, color = 'b')

导致标签成为数据格式。解决这一问题的办法是通过以下方式大幅增加数字大小：

fig, ax1 = plt.subplots(figsize=(8, 8)