数据可视化之美 -- 以Matlab、Python为工具

{. B+ \* q4 d# |

; |9 C% W7 x( I; M1 |

在我们科研、工作中，将数据完美展现出来尤为重要。

3 T, k6 W, f, @- g: h

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。

* v+ R5 \% _8 d- n8 s! S/ _: }

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）

& p. i- g$ y$ {" R1 O1 I

import numpy as np

import matplotlib.pyplot as plt

# 创建随机数

n = 100000

S; U6 A* R8 z& X: m O; M4 t

x = np.random.randn(n)

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

fig1 = plt.figure()

plt.plot(x,y,.r)

plt.xlabel(x)

plt.ylabel(y)

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

nbins = 200

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

# H needs to be rotated and flipped

6 N5 r- v' Z( U& P, S/ M

H = np.rot90(H)

H = np.flipud(H)

" b$ t, e: ^* W7 I; ^

# 将zeros mask

5 Q9 {; I0 k& m' C0 Y( Q5 [1 @

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

# Plot 2D histogram using pcolor

% d) z3 r; x& m6 |7 Z

fig2 = plt.figure()

plt.pcolormesh(xedges,yedges,Hmasked)

/ S) D: s' ]* ~: e) B) @

plt.xlabel(x)

, L9 G3 ?. l1 C' C

plt.ylabel(y)

cbar = plt.colorbar()

cbar.ax.set_ylabel(Counts)

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

$ s0 e$ Q! Q* C# a0 i

plt.show()

4 m m0 Z& |" y: B6 C

打开凤凰新闻，查看更多高清图片

6 a, v1 _0 q7 p* V/ n

, M% q9 i4 i# O$ Z

0 }, o. c) S& p7 d% h6 H

Example 2 ：双Y轴（Python）

& b8 g$ o/ Q0 v" R% L# ]3 v

import csv

- h3 y" f* r" M6 q

import pandas as pd

import matplotlib.pyplot as plt

+ M6 |! U+ W' K- h v# y1 }

from datetime import datetime

: \' g3 E# j! r H1 {2 q/ a% @2 u

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

time=data[date [AST]]

B) B% R1 o4 W' T5 I7 h

sal=data[salinity]

' p7 M; y* j! D3 y7 l3 I

tem=data[temperature [C]]

7 {9 O4 N( m4 m8 G- e

print(sal)

9 C9 r5 v7 O: ^; v5 \0 S: W

DAT = []

for row in time:

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

#create figure

; z2 B' a% ?% R1 X. K' x+ i

fig, ax =plt.subplots(1)

4 Q7 T, ^6 F# D; O# f8 K, a" k* b( S

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

& I, m0 a: W$ O

plt.xlabel("Date [AST]")

. Z( s4 ~ I: P, d3 P3 |2 m

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

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

2 S' k! a: n+ Y2 h% E. h

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

8 n6 X- h6 s. f0 R' S

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

fig.autofmt_xdate(rotation=50)

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

' d8 j0 a1 V m' y" ?) ]7 Q- X

plt.twinx()

. B/ x; n: z, i

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

7 ?# w( [/ |1 ~0 ^) W$ a. R$ O

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

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

7 e+ I/ g7 ?* ~2 m# O$ i! |7 ?0 ]

#To save your graph

g: P W+ }+ }2 x2 |

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

plt.show()

Example 3：拟合曲线（Python）

import csv

import numpy as np

import pandas as pd

from datetime import datetime

9 J; n3 \- V3 D1 J: Q

import matplotlib.pyplot as plt

! w& C8 P* a6 Q

import scipy.signal as signal

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

. u+ Y1 d3 y& u' y

time=data[date [AST]]

temp=data[temperature [C]]

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

DATE,decday = [],[]

+ U, t4 Y* o4 z$ F

for row in time:

: X. h. a! ?6 X5 y% P

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

DATE.append(daterow)

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

/ U5 x/ a/ _5 P3 Y) p6 ^

# First, design the Buterworth filter

. b+ o& x; {; T# v% V0 r" {

N = 2 # Filter order

0 S1 w6 S) t4 `' C2 K! w+ c. [

Wn = 0.01 # Cutoff frequency

- V' i) Q0 l Q% W

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

# Second, apply the filter

1 X3 y8 K" S% R6 Z {

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

# Make plots

fig = plt.figure()

' b: \2 r+ _# y \, X3 I

ax1 = fig.add_subplot(211)

& v2 v+ e5 _: [* k( @

plt.plot(decday,temp, b-)

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

# }6 n9 x* c$ U+ P) o7 c8 B

plt.ylabel("Temperature (oC)")

9 J2 H9 w+ |9 V+ z7 ?

plt.legend([Original,Filtered])

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

& ^# q: \" P6 J/ M" q3 y

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

ax1 = fig.add_subplot(212)

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

) [! ]& L4 T! Q0 S

plt.ylabel("Temperature (oC)")

plt.xlabel("Date")

plt.legend([Residuals])

5 T6 B* ?9 b3 p: q! z4 L& U* A

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

plt.show()

5 c1 j/ X8 v. y1 Z) U4 |

Example 4：三维地形（Python）

! y* D% A0 A6 I- E

# This import registers the 3D projection

' i3 g9 F1 _3 \' `- v$ p6 V- b* w

from mpl_toolkits.mplot3d import Axes3D

' j+ `( e7 P+ l+ n6 u5 W

from matplotlib import cbook

from matplotlib import cm

from matplotlib.colors import LightSource

) G! N$ T! H: @

import matplotlib.pyplot as plt

import numpy as np

4 S# X$ B+ Q/ t$ y5 j0 v# s2 ]3 a

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

with np.load(filename) as dem:

& ?- N D6 V, x8 O3 j9 M

z = dem[elevation]

$ c( b: ?; J2 m% Z" ?1 P

nrows, ncols = z.shape

. x: e" R( M# W) K1 u+ B! N F r

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

5 L+ D2 c2 H$ T

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

% C6 F, m% B3 k& O4 W

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

0 T) g G2 K5 R

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

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

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

( m9 i+ ?: k0 Q8 L- W8 S

ls = LightSource(270, 45)

2 I2 d; b8 i i0 C5 l* w

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)

- w( z) V+ A6 u8 A" w) e. X& ]

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

4 W( a/ B# z5 _: W* u

plt.show()

2 I/ D0 }7 e! M( J+ L

Example 5：三维地形，包含投影（Python）

) q0 {# w! }, S1 j. j- v. r

Example 6：切片，多维数据同时展现（Python）

! W+ z. H% _! _$ i

Example 7：SSH GIF 动图展现（Matlab）

Example 8：Glider GIF 动图展现（Python）

Example 9：涡度追踪 GIF 动图展现

7 |; o8 s/ }* ^1 _! e: f 3 v" q, X# O) c' z, a x* C6 k

' p' A% c0 P* ^5 a3 j

( @7 {- ~! M! c2 E( y3 ]

% c- i$ {" U0 C- \# Y( w2 P 6 D( [5 u: e g6 c8 h" w / S7 E* T1 G& z4 e3 U2 |, w# w