数据可视化之美 -- 以Matlab、Python为工具

5 T% i; M- M" u* d3 E. U% o. C

/ M4 V) d% j' {4 Z6 n' K

% n0 k# x/ E5 j, O( Z+ u

6 H! J8 S1 V2 r9 _; ]& f

在我们科研、工作中，将数据完美展现出来尤为重要。

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。

' v5 o: n# L! \0 b0 A K

Example 1 ：散点图、密度图（Python）

import numpy as np

import matplotlib.pyplot as plt

# y2 {2 ]$ v" O* n+ }8 l0 |

# 创建随机数

n = 100000

x = np.random.randn(n)

$ E: R9 c$ h: E! m" X

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

fig1 = plt.figure()

+ \* u' J+ @- l q6 Z# B

plt.plot(x,y,.r)

. Q4 C/ u/ c2 N, J

plt.xlabel(x)

2 {7 M9 k6 ]0 M

plt.ylabel(y)

: I* Y7 t) C" v6 S) s8 k' J$ W) k

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

nbins = 200

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

2 Y& P! l2 @, g* r

# H needs to be rotated and flipped

$ i2 Y7 U5 h/ v

H = np.rot90(H)

8 I6 ^7 E U. n* c4 U' I

H = np.flipud(H)

) s& G" f3 h: B: ^3 g+ d8 {

# 将zeros mask

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

6 f+ A: I7 E# @" _0 X4 y

# Plot 2D histogram using pcolor

% }0 s' Q7 g3 S

fig2 = plt.figure()

plt.pcolormesh(xedges,yedges,Hmasked)

plt.xlabel(x)

; z; D& d2 T6 j! X; X

plt.ylabel(y)

cbar = plt.colorbar()

cbar.ax.set_ylabel(Counts)

" t' ^) z- q9 M3 v5 c& l1 b

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

5 `1 K- i9 N, R7 W2 ?

plt.show()

7 O! {7 S8 C0 K+ {6 j6 w+ T1 R* i

打开凤凰新闻，查看更多高清图片

* A" l7 O, [3 A7 w. H

/ O) }- s' m) m

6 V. u& r: S1 ]# V5 d% K* P

Example 2 ：双Y轴（Python）

9 \0 s! ~. B0 B# P3 s: R

import csv

! U8 `4 M5 ~0 a+ N I1 x

import pandas as pd

import matplotlib.pyplot as plt

from datetime import datetime

/ o( c/ q* I6 A5 i

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

time=data[date [AST]]

: M' k) W9 w. m3 F& {& q( }

sal=data[salinity]

3 e8 {! e x' p( K! L

tem=data[temperature [C]]

7 U* |, N& }$ n' b

print(sal)

DAT = []

for row in time:

) d9 B# L% |$ ^: a" N# E

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

#create figure

. D! x- e h8 f Z2 e7 C* d: E

fig, ax =plt.subplots(1)

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

plt.xlabel("Date [AST]")

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

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

- C" z/ I1 b: [, R0 m- Z

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

4 X* l# D& M0 A% }

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

: z, v0 X, E3 Y- m% K9 {' J

fig.autofmt_xdate(rotation=50)

1 B/ \; ~4 ^5 m' S) A7 Q

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

plt.twinx()

# M! |# T; Z3 {' [

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

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

5 x7 a: S( D8 L$ O7 w, E9 B, ~

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

3 Q$ i' g+ `6 ]+ u( A3 g: y6 O9 y4 ?

#To save your graph

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

plt.show()

7 h0 n. d o) g

Example 3：拟合曲线（Python）

6 z6 f3 a% c7 j# Q

import csv

3 [0 K# P, O7 A! _0 ]9 s. x

import numpy as np

! s5 W; y G& Z2 A) h

import pandas as pd

from datetime import datetime

import matplotlib.pyplot as plt

import scipy.signal as signal

+ F# d! D2 a5 i2 v) Y

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

9 L9 ~7 K ]+ [7 i" B7 G, |; W/ M3 b

time=data[date [AST]]

7 a3 t: D, ]9 M

temp=data[temperature [C]]

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

DATE,decday = [],[]

for row in time:

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

/ v* Y# K% N" D0 w( j

DATE.append(daterow)

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

# First, design the Buterworth filter

5 L5 ?& H$ T9 X

N = 2 # Filter order

Wn = 0.01 # Cutoff frequency

w5 \/ R3 ^9 L! ?

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

# Second, apply the filter

9 D3 d) r- x6 e, y7 S

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

# Make plots

. h3 B* e( ]# j+ q

fig = plt.figure()

ax1 = fig.add_subplot(211)

0 R: x; v8 ]: H$ S' [+ d3 c

plt.plot(decday,temp, b-)

" W+ q# c2 x8 P. S; _( h

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

& F0 o- k6 w# c5 V! U2 k

plt.ylabel("Temperature (oC)")

: m# \) r" Y" G( J1 h6 n' `

plt.legend([Original,Filtered])

! f; \* j1 @. ~/ H7 ?3 d* P h

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

^2 h' l- c5 Q' t8 l- v! ~

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

ax1 = fig.add_subplot(212)

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

3 Y' R: `$ C6 C% o* \# W0 H

plt.ylabel("Temperature (oC)")

& `9 g |) @4 `" I4 b& L1 e6 M

plt.xlabel("Date")

plt.legend([Residuals])

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

: P, x. N. g- w1 s

plt.show()

7 u1 Q/ c. ?" o4 z; o5 m% R+ Q+ ^

7 b6 `( r- {: f1 s8 h u+ L

Example 4：三维地形（Python）

# This import registers the 3D projection

from mpl_toolkits.mplot3d import Axes3D

5 D( @6 f. v S' C

from matplotlib import cbook

: D" r7 q: e2 v @; r- b' r6 ]4 i

from matplotlib import cm

& N+ {4 |- K4 {) U1 W0 W: V

from matplotlib.colors import LightSource

import matplotlib.pyplot as plt

import numpy as np

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

6 n' U e4 `" K

with np.load(filename) as dem:

% M" D% W" \, w$ p% L

z = dem[elevation]

nrows, ncols = z.shape

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

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

6 Z! W3 \; o. H# {# S& q

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

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

0 W! [7 @8 g- e; b# n2 I

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

8 P. U/ i8 D# s

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

8 K: i' |3 {0 S" F8 W' p

ls = LightSource(270, 45)

4 z) h& e1 ?- a

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,

8 n" H( Z1 M" W* s

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

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

plt.show()

Example 5：三维地形，包含投影（Python）

" y' h Y3 N) d9 B

Example 6：切片，多维数据同时展现（Python）

$ h' c" ?% Q* G: F# _& b+ d8 l

Example 7：SSH GIF 动图展现（Matlab）

Example 8：Glider GIF 动图展现（Python）

p3 U# Y* ]- |$ ]3 I1 s) j- ]

Example 9：涡度追踪 GIF 动图展现

# h) k! X4 I# L5 | {

% y( Z5 v- ?. S ~ ?7 _, K % n5 O4 h* Y# K( A1 S