How to move tick labels off left spine

前端 未结 2 534
时光说笑
时光说笑 2020-12-22 09:57

In matplotlib, how do I move the y tick labels to the far left side of the charting areas when I have moved the left spine?

Code:

import matplotlib.p         


        
相关标签:
2条回答
  • 2020-12-22 10:36

    Changing the transformation of the y-ticklabels back to the original y-axis transform would give you the desired ticks on the left side of the axes.

    plt.setp(ax.get_yticklabels(), transform=ax.get_yaxis_transform())
    

    Complete code for reproducibility

    import matplotlib.pyplot as plt
    
    X = range(-10,5)
    y = [i**2 for i in X]
    
    fig, ax = plt.subplots(1,1, figsize=(10,8))
    plt.plot(X, y)
    ax.spines['top'].set_visible(False)
    ax.spines['right'].set_visible(False)
    ax.spines['left'].set_position('zero')
    
    plt.setp(ax.get_yticklabels(), transform=ax.get_yaxis_transform())
    
    plt.show()
    
    0 讨论(0)
  • 2020-12-22 10:37

    You can exploit the set_position attribute of your tick labels by iterating over them. Here you have to pass the (xy) value to be set as the new location of your tickl-labels. We know the x-position is -10. What I found is that it does not matter what y-position you specify, the ticks are offset by the value of x which kind of makes sense too because the tick labels have to correspond to the ticks on the y-axis. However, one can't leave the y-value empty. So specifying (-10) won't work and (-10, 0), (-10, -100) or (-10, 1000) have no effect on the y-position of the ticks. The reason is as @ImportanceOfBeingErnest clarified: The y position of the label will be determined by the y-values of your data and is set at draw time after applying the changes.

    for tick in ax.yaxis.get_major_ticks():
        print (tick.label.set_position((-10, 0)))
    

    0 讨论(0)
提交回复
热议问题