MatplotLib

Basic

MatplotLib and Seaborn

Matplot Lib API Design

• plotting API - pyplot module
• Library - pylab module (useful function in Numpy and pyplot)
• plt.plot(x1,y1,x2,y2..., <attributes controls>)
  • plt.bar
  • plt.title(), .xlabel(),.ylabel()
  • plt.gca() - set axes
    • .setx_lim(), .set_ylim()
  • plt.figure(<figsize>,<dpi>
  • attributes control
    • color
      • b,g,m(megenta),w....
    • type:
      • \: solid,
      • _: dashed,
      • _. : dash_dot
      • , : - dotted,
      • None-nothing, ' '. "" - nothing_
    • marker:
      • o - circle
      • v -triangle_down
      • s-square
      • p -pentagon
      • * -star
      • h-hexagon1
      • +-plut
      • D-diamond
  • plt.xlable(), ylable(), title(), legend([loc])
• subplots
  • plt.subplot(<nrow><ncol><ind>, [color, maker..])

Pyplot Module: MATLAB-like interfaces

MatplotLib_Utils

• simpleTrainingCurves()
  • to visualize training

Pandas Ploting

• <dataframe/data column/ data series>.plot([kind])
  • return to plt directly
  • use plt to set up arguments
    • x, y
    • kind
      • bar, pie
    • subplots = True/False
      • default plot all columns together
  • autopct = <format string>
• <df>.boxplot()
  • 3 idioms returns different things
    • <df>.plot(kind='hist')
    • <df>.plot.hist()
    • <df>.hist()

Seaborn

• .set_style(<style>)
• displot(<dist>), box_plot(), heatmap()

PyLab

Boken

Plotly