Data analysis notes 09

Visual chart There is a trend Distribution constitute Compare and practice

matplotlib

Trends show how data changes over time
Through a line chart Or histogram
Distribution Distribution law of data Use a scatter chart to show different distribution characteristics
constitute Used to represent The proportion of each part of the data in the total If Want to show the proportion of a period of time , It is more suitable to display with percentage stacked histogram
Compare Used for data comparison in a certain dimension The relationship between comparative data is usually reflected by cluster histogram and parallel subgraph
contact Measure the relationship between two dimensional data double y Axis overlay

matplotlib The composition of the image Install and import matplotlib modular Set up matplotlib typeface
Elements canvas (Figure) Coordinate image (Axes) Image title （Title) Line (line) Mark (marker) Legend (legend）
Set up Chinese
macOS plt.rcParams[“font.sans-serif”]=“Arial Unicode MS”
Windows plt.rcParams[“font.sans-serif”]=“SimHei”
Show the image plt.show()
Broken line diagram plot Histogram bar The line chart has linechar
Emphasize the change law of data over time Acquisition is It is very suitable for showing the journey overview at the same time

plt.plot() Set the required parameter x Axis y The value corresponding to the axis is passed into the function in turn
y The axis is time type data y The axis is the data corresponding to a time point
Integer is continuous data str It belongs to discontinuous data
TODO Use plt.plot() function
With data[“month”] by x The sum of the values of the axis data[“sum”] by y The value of the shaft , Draw line chart

plt.plot(data[“month”],data[“sum”],color=“orange”,marker=“o”,label=" Total monthly sales )
legend() Show Legend
TODO Use plt.show() Function to display the image
Add axis title Add image title plt.xlabel() plt.ylabel() plt.title()
Draw a histogram Show the data changes over a period of time or the comparison between various data x The axis is a time type Or category data y The axis is numerical data
plt.bar The parameters inside are the same as the line chart
label Used to set images and titles

Scatter plot Show the correlation and comparison values between the two data
The scatter chart can provide the following key information Whether there is a correlation trend If there is a trend Is it linear or nonlinear Whether there are outliers
color legend xlabel ylabel Nibbling is the same as before

Adjust the legend position In order to solve the problem of double y The legend of the axis overlay diagram overlaps

plt.twinx() The function represents drawing double x Axis overlay

Draw a composite image

Percentage stacked histogram
Clustered histogram
Compare multiple data on one dimension Draw a clustered histogram Just for one DataFrame object
Dyadic subgraph
x Axis and y The specific operation of the shaft is Calling plot.bar() Function time Index columns Or the list of fake case column indexes is passed into this function
data Variable usage plot.bar() function
data.plot.bar
Percentage stacked histogram
Show the relationship between a single item and the whole
data.plot.bar(“month”,[“ On the first floor ”,“ On the second floor ”,“ Third floor ”],stacked=True)
figure Is a coordinate graph Axes
adopt subplot() Row number of subgraphs Number of columns in a subgraph Sequence number of subgraph
Avoid axis occlusion rotate x Axis readable plt.xticks() Assign degrees as integers to rotation Parameters Passed in the function To rotate x Axis scale
plt.tight_layout() Function to adjust the layout of the subgraph
ax=plt.gca() It needs to be used in the drawing to return to the current
ax Parameters are used to specify coordinate graphics
pandas Modules cannot be like matplotlib.pyplot By default, the image will be drawn on the coordinate axis of the current sub image