数据可视化之美 -- 以Matlab、Python为工具

2 U: k6 O' U! P) B1 F' Q% Y5 [2 X

/ i9 }2 s) i+ l- s% l

在我们科研、工作中，将数据完美展现出来尤为重要。

- k" K S" ]& O$ p$ ^3 H* E

数据可视化是以数据为视角，探索世界。我们真正想要的是 — 数据视觉，以数据为工具，以可视化为手段，目的是描述真实，探索世界。

下面介绍一些数据可视化的作品（包含部分代码），主要是地学领域，可迁移至其他学科。

Example 1 ：散点图、密度图（Python）

, }( `( X! ~3 M8 q" e

import numpy as np

import matplotlib.pyplot as plt

# 创建随机数

% y% x, [- E' P# q

n = 100000

2 c+ @( h# G9 y: q4 w ]5 o

x = np.random.randn(n)

4 k5 A, ~! A! X& W

y = (1.5 * x) + np.random.randn(n)

fig1 = plt.figure()

$ [# D, }: v1 h! M9 g# p

plt.plot(x,y,.r)

1 s7 ~! ~9 _" }; c; Q3 ?1 W. {; e( W

plt.xlabel(x)

- q* l) ~1 _- l, _6 h4 y

plt.ylabel(y)

+ t- Z+ }* Y A8 M2 X$ v6 V

plt.savefig(2D_1V1.png,dpi=600)

nbins = 200

5 Q7 r: y+ P' [# X4 Q

H, xedges, yedges = np.histogram2d(x,y,bins=nbins)

# H needs to be rotated and flipped

4 g o H M4 H7 i& t

H = np.rot90(H)

H = np.flipud(H)

# 将zeros mask

# B+ g8 }+ Z* Y. b8 ~# c% f$ W

Hmasked = np.ma.masked_where(H==0,H)

' F1 S" z, m) |' f( i' F E

# Plot 2D histogram using pcolor

fig2 = plt.figure()

plt.pcolormesh(xedges,yedges,Hmasked)

1 ^1 J8 Z8 }0 m1 `5 i4 S1 V

plt.xlabel(x)

plt.ylabel(y)

, z8 S8 R3 C o' X/ x: O; v

cbar = plt.colorbar()

* a* u8 B: B1 r0 |" m( p

cbar.ax.set_ylabel(Counts)

plt.savefig(2D_2V1.png,dpi=600)

plt.show()

% e0 ~6 K2 I( K" U/ E% h3 {

打开凤凰新闻，查看更多高清图片

( B6 y- y/ g: H/ ?& G

# c% H' E/ W3 b$ m9 E% N2 p

7 _: w3 L$ ?3 F

Example 2 ：双Y轴（Python）

; O0 G c' T( h2 N# N! J( K

import csv

import pandas as pd

7 J W, ~6 x' h2 W* F8 ~4 Y

import matplotlib.pyplot as plt

from datetime import datetime

data=pd.read_csv(LOBO0010-2020112014010.tsv,sep=)

. K4 Y B @- { c0 O6 H

time=data[date [AST]]

sal=data[salinity]

tem=data[temperature [C]]

print(sal)

+ h; T* e- U) H/ f% V: D0 `" A

DAT = []

for row in time:

$ _# ], \% W0 _8 u" N" y2 i: s4 E3 d

DAT.append(datetime.strptime(row,"%Y-%m-%d %H:%M:%S"))

#create figure

fig, ax =plt.subplots(1)

# Plot y1 vs x in blue on the left vertical axis.

. x0 P& z1 t- l# G( v. {0 N- |

plt.xlabel("Date [AST]")

0 ^( h" J) \; Y$ o3 C) v

plt.ylabel("Temperature [C]", color="b")

plt.tick_params(axis="y", labelcolor="b")

1 k( z) e$ k S# w* D! H6 |5 x

plt.plot(DAT, tem, "b-", linewidth=1)

plt.title("Temperature and Salinity from LOBO (Halifax, Canada)")

fig.autofmt_xdate(rotation=50)

g( b" F3 ^' g+ ?9 {

# Plot y2 vs x in red on the right vertical axis.

plt.twinx()

$ T9 m* p! e' `. ]/ N3 M

plt.ylabel("Salinity", color="r")

! _, ~' |( l, e' D V# f+ B& ]8 {

plt.tick_params(axis="y", labelcolor="r")

% b) m* O \/ \' a( W

plt.plot(DAT, sal, "r-", linewidth=1)

#To save your graph

0 g9 Y7 d3 a, m, V) \

plt.savefig(saltandtemp_V1.png ,bbox_inches=tight)

plt.show()

5 m. ^0 A+ I0 B# X6 `

) U2 y& H3 [1 N, z+ ]3 D6 k

Example 3：拟合曲线（Python）

import csv

5 x- {% Y2 N, [* o; x# O0 m' n

import numpy as np

import pandas as pd

' x# |& L' h8 v$ M

from datetime import datetime

' `' @3 p5 R3 e! ~

import matplotlib.pyplot as plt

import scipy.signal as signal

6 ~& r/ O# @) v; U

data=pd.read_csv(LOBO0010-20201122130720.tsv,sep=)

time=data[date [AST]]

2 ^. w) Q; P, @ o* a

temp=data[temperature [C]]

datestart = datetime.strptime(time[1],"%Y-%m-%d %H:%M:%S")

- O3 M% v5 x3 l' v0 w/ i

DATE,decday = [],[]

for row in time:

+ l5 D8 S3 @3 o3 z. p9 I+ V T6 {' N; o

daterow = datetime.strptime(row,"%Y-%m-%d %H:%M:%S")

4 _" ]( j( v |! s5 O2 ?2 d) v

DATE.append(daterow)

decday.append((daterow-datestart).total_seconds()/(3600*24))

# First, design the Buterworth filter

N = 2 # Filter order

+ {8 ?0 k4 C* b+ ^: B& o3 {

Wn = 0.01 # Cutoff frequency

B, A = signal.butter(N, Wn, output=ba)

+ {/ p( u* V2 Y6 N1 g

# Second, apply the filter

tempf = signal.filtfilt(B,A, temp)

& E! }! J* b3 X0 Q2 M2 n

# Make plots

fig = plt.figure()

* e# r& C3 ~% z8 P6 S* I1 V2 o

ax1 = fig.add_subplot(211)

, ]# J0 ]0 l: C N1 R

plt.plot(decday,temp, b-)

plt.plot(decday,tempf, r-,linewidth=2)

plt.ylabel("Temperature (oC)")

plt.legend([Original,Filtered])

+ o0 Z/ \5 B" m0 ]8 h2 b" J4 |

plt.title("Temperature from LOBO (Halifax, Canada)")

% M. m G) `! I! i' T

ax1.axes.get_xaxis().set_visible(False)

ax1 = fig.add_subplot(212)

. d+ Z+ K9 Z1 ]1 f0 v% Y0 P; S$ G

plt.plot(decday,temp-tempf, b-)

* l% Y& u" X6 S0 s+ T) B. c, }

plt.ylabel("Temperature (oC)")

: Y; p6 j. O2 x

plt.xlabel("Date")

' d. w6 X' g+ w* \6 Y

plt.legend([Residuals])

plt.savefig(tem_signal_filtering_plot.png, bbox_inches=tight)

plt.show()

/ {6 e2 q" s( O# s8 t

* b2 f( z* K) l- ^

Example 4：三维地形（Python）

# This import registers the 3D projection

from mpl_toolkits.mplot3d import Axes3D

. }1 d. J' p/ p- E F" {1 O

from matplotlib import cbook

from matplotlib import cm

' z: ?3 F* R% f# ^9 W6 T+ x

from matplotlib.colors import LightSource

import matplotlib.pyplot as plt

import numpy as np

9 m6 V3 u1 @! g7 h: E5 ^7 u1 `9 k

filename = cbook.get_sample_data(jacksboro_fault_dem.npz, asfileobj=False)

8 p, x4 S. C3 o& e! h/ ~6 u5 m

with np.load(filename) as dem:

z = dem[elevation]

nrows, ncols = z.shape

0 e6 j# {/ p! {+ W% v$ h( E

x = np.linspace(dem[xmin], dem[xmax], ncols)

y = np.linspace(dem[ymin], dem[ymax], nrows)

$ m( i' m1 ?, g( M% t# ~

x, y = np.meshgrid(x, y)

region = np.s_[5:50, 5:50]

* q! [* ^! d$ Y% w7 _9 M. C

x, y, z = x[region], y[region], z[region]

" k7 T8 d3 l% ^9 g6 e

fig, ax = plt.subplots(subplot_kw=dict(projection=3d))

ls = LightSource(270, 45)

3 Z0 U9 {5 Q5 {; P0 {2 g" y8 l" M

rgb = ls.shade(z, cmap=cm.gist_earth, vert_exag=0.1, blend_mode=soft)

surf = ax.plot_surface(x, y, z, rstride=1, cstride=1, facecolors=rgb,

linewidth=0, antialiased=False, shade=False)

plt.savefig(example4.png,dpi=600, bbox_inches=tight)

plt.show()

( S' M- ?; z2 N* Q/ ~ 2 S7 l% a/ @. u1 w8 L* _/ g* p

Example 5：三维地形，包含投影（Python）

5 Y7 O. Z. u; d$ ] 6 k1 ]7 |+ m* e# L/ r1 W2 C

Example 6：切片，多维数据同时展现（Python）

+ V, X0 Q7 [( h , Y4 P9 }: C$ D+ g" k

Example 7：SSH GIF 动图展现（Matlab）

; V+ N6 C3 T4 s& h/ J # D( V3 \/ H3 _2 X, ?) y3 Y

Example 8：Glider GIF 动图展现（Python）

1 ^3 A9 P, _1 ^$ i

Example 9：涡度追踪 GIF 动图展现

' f! b, T, j6 U" ], d+ e

. [ a& ^" e+ O( K9 G6 A

( j; S! x/ s, ] R q0 ^" b

$ i, v2 K! N1 a+ J3 o( x# n9 P2 V* a - n; t) g4 Z% n' ^7 W ; M) h/ X; T, p/ _