Reference Source: "Python Data Science Handbook," Chapter 4
Description: Notes
table of Contents:
1, Matplotlib General Tips
1.1 Import Matplotlib
1.2 provided drawing style
1.3 with no show ()? How to display graphics
1.4 Save the graphics file
2 Both interfaces Paint
2.1 MATLAB Interface style
2.2 Object-Oriented Interface
3, a simple line graph
3.1 Adapting Graphics: line color and style
3.2 adjusting pattern: the lower limit coordinate axes
3.3 Setting Graphic labels
4, simple scatter plot
4.1 Scatter painting with plt.plot
4.2 Scatter painting with plt.scatter
4.3 plot and scatter: Efficiency Comparison
5, exception handling Visualization
5.1 Basic error line
5.2 Continuous Error bars
6, and FIG density contour plots
Three-dimensional view visualization 6.1
7, frequency histogram, and the distribution density data interval into
8, configuration legend
8.1 Select the legend elements displayed
8.2 points of different sizes displayed in the legend
Simultaneously displaying a plurality of legends 8.3
9, a color bar configuration
9.1 Configuring color bar
Case 9.2: Digital Handwriting
10, the multi subgraph
10.1 plt.axes: Manually create a child diagram
10.2 plt.subplot: Simple Grid subgraph
10.3 plt.subplots: create a grid with a single line of code
10.4 plt.GridSpec: more complex arrangement
11, text and notes
11.1 Case: Holidays on the US birth rate
11.2 coordinate transformation and text position
11.3 arrows and notes
12, custom axis scale
12.1 major scale and a minor scale
12.2 hide the scale and label
12.3 increase or decrease the number of tick
12.4 fancy scale format
12.5 format generation with locator Summary
13, Matplotlib custom: the configuration file and stylesheets
13.1 Manually configure graphics
13.2 modify the default configuration: rcParams
13.3 stylesheet
14, a three-dimensional drawing Matplotlib
14.1 dimensional data points and lines
14.2 dimensional contour
FIG wireframe and surface 14.3
14.4 curved triangular part
15, with the geographic data visualization Basemap
15.1 Map Projection
15.2 Draw a map background
15.3 picture data on the map
15.4 Case: California city data
15.5 Case: Surface temperature data
16, for data visualization with Seaborn
16.1 Seaborn与Matplotlib
16.2 Seanborn graphical presentation
16.3 Case: Exploring marathon performance data
17 references
17.1 Matplotlib resources
17.2 Other Paint Python library
1, Matplotlib General Tips
1.1 Import Matplotlib
1.2 provided drawing style
1.3 with no show ()? How to display graphics
Three kinds of development environments: script, IPython shell and IPython Notebook
In the script , use matplotlib, display graphics must plt.show (), plt.show () will start an event loop (event loop), and find all the currently available graphical objects, and then open one or more interactive window display graphics, to note that, a python session can only be used once plt.show (), it is often regarded it placed at the end of the script.
In ipython shell in , you need to start matplotlib mode, matplotlib start by way of magic%. No longer need to use plt.show (). Can force an update by plt.draw ().
In ipython notebook in , we need to start matplotlib mode.
1.4 Save the graphics file
2 Both interfaces Paint
2.1 MATLAB Interface style
This interface is the most important feature of the state (stateful): it keeps track of the "current" graphics and axes, all plt command can be applied, available plt.gcf () to get the current graphics, plt.gca () get the current axis.
Disadvantages: inconvenient to switch the sub FIG.
2.2 Object-Oriented Interface
Object-oriented interfaces are no longer limited by the current "active" or pattern axis, it becomes an explicit method and Axes of Figure.
To draw more complex graphics, opposite the object method will be more convenient.
3, a simple line graph
3.1 Adapting Graphics: line color and style
3.2 adjusting pattern: the lower limit coordinate axes
3.3 Setting Graphic labels
matplotlib trap
4, simple scatter plot
4.1 Scatter painting with plt.plot
4.2 Scatter painting with plt.scatter
4.3 plot and scatter: Efficiency Comparison
5, exception handling Visualization
5.1 Basic error line
5.2 Continuous Error bars
To resolve the error displayed by a continuous variable and plt.plot plt.fill_between. The lower limit on the parameters passed plt.fill_between drawing, painting and graphics by plt.plot do comparison to visualize the error.
6, and FIG density contour plots
In the two-dimensional map represented by the three-dimensional contour data or color maps FIG.
With plt.contour drawing contour maps, contour maps (filled contourplot) filled with color by color plt.contourf Videos, graphics display with plt.imshow.
Three-dimensional view visualization 6.1
Plt.contour contour map can be used to create a function, it takes three parameters: x axis, y axis, z-axis grid data of the three axes. x-axis and y-axis represents the position of the pattern, and the z-axis level will be represented by contour lines.
When only one color graphics, dashed lines represent negative default, a solid line denotes a positive number.
matplotlib color scheme can view the information in the corresponding module plt.cm ipython using the tab key.
By analyzing the color bar: black area is the peak (peak), the red region is a valley (valley).
Description: The color change is a discrete rather than a continuous process, it does not look so clean
7, frequency histogram, and the distribution density data interval into
7.1 Frequency histogram
The above is a one-dimensional array is divided into sections to create a one-dimensional frequency histograms .
7.2 two-dimensional histogram and frequency data interval division
The two-dimensional array in accordance with the two-dimensional segmentation section, to create a two-dimensional frequency histograms.
7.3 Kernel Density Estimation
Kernel density estimation (kernel density estimation, KDE), a common method of multi-dimensional data distribution density assessment.
8, configuration legend
Legend represents a discrete graphical elements by discrete label.
8.1 Select the legend elements displayed
Legend want to use in a visual pattern, graphic elements may be assigned different labels.
The legend is displayed by default label all the elements, but ignores those elements without labels
Simultaneously displaying a plurality of legends 8.2
By creating a new scratch artists objects legend (legend artist), and () method adds a second legend on the view of the bottom (lower-level) ax.add_artist.
9, a color bar configuration
For drawing a colored dot, line, surface constituting the continuous label, color bar to indicate the effect is better.
In matplotlib, the color bar is an independent axis, can indicate the meaning of colors in the graphic.
9.1 Configuring color bar
Color bar itself can be seen as only a plt.Axes example, the format may be adapted and arranged about the axis scale values (e.g. plt.clim ()) .
10, the multi subgraph
10.1 plt.axes: Manually create a child diagram
Function plt.axes (), to create a default standard coordinate axis, to fill the entire FIG.
It has an optional parameter value pattern has four coordinate system configuration, respectively, is a graph showing a coordinate system [bottom, left, width, height] (the end coordinates, the left coordinate, width, height), the numerical value range lower-left corner (origin) to 0, the upper right corner is 1.
10.2 plt.subplot: Simple Grid subgraph
10.3 plt.subplots: create a grid with a single line of code
10.4 plt.GridSpec: more complex arrangement
11, text and notes
11.1 Case: Holidays on the US birth rate
11.2 coordinate transformation and text position
11.3 arrows and notes
12, custom axis scale
12.1 major scale and a minor scale
12.2 hide the scale and label
12.3 increase or decrease the number of tick
12.4 fancy scale format
12.5 format generation with locator Summary
13, Matplotlib custom: the configuration file and stylesheets
13.1 Manually configure graphics
13.2 modify the default configuration: rcParams
13.3 stylesheet
14, a three-dimensional drawing Matplotlib
14.1 dimensional data points and lines
14.2 dimensional contour
FIG wireframe and surface 14.3
14.4 curved triangular part
15, with the geographic data visualization Basemap
15.1 Map Projection
15.2 Draw a map background
15.3 picture data on the map
15.4 Case: California city data
15.5 Case: Surface temperature data
16, for data visualization with Seaborn
16.1 Seaborn与Matplotlib
16.2 Seanborn graphical presentation
16.3 Case: Exploring marathon performance data
17 references
17.1 Matplotlib resources
17.2 Other Paint Python library