数据可视化之美 -- 以Matlab、Python为工具

! W$ w X# F6 j8 a

- e! m6 m9 {: D: n$ M3 V5 e

7 f- `9 p. x* G3 {

O. W4 c7 c$ D0 G" T

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。

Example 1 ：散点图、密度图（Python）

import numpy as np

import matplotlib.pyplot as plt

8 j8 F# ~+ Q% Y) `& ?* W

# 创建随机数

% O$ P( H8 j8 v& W }

n = 100000

x = np.random.randn(n)

4 H1 S+ x4 D3 z/ }

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

fig1 = plt.figure()

plt.plot(x,y,.r)

2 P/ ~" b. @5 T4 }1 t% {, w

plt.xlabel(x)

9 ~/ f: \& A; @% p6 ]4 Y

plt.ylabel(y)

4 i1 A/ S+ o. I0 M! I T' X) Z; u

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

. O2 z. F( `9 J4 \) a" [

nbins = 200

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

( P5 p- g1 F8 ^

# H needs to be rotated and flipped

- V" \6 K6 K0 ?8 j/ G

H = np.rot90(H)

H = np.flipud(H)

) i. Q, D, L+ P' c+ a+ ~5 B0 l5 U& Z

# 将zeros mask

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

# Plot 2D histogram using pcolor

$ d. h" ?* r+ ^1 v; C. f+ M

fig2 = plt.figure()

3 ]- s9 r/ P& b! @% A! M }

plt.pcolormesh(xedges,yedges,Hmasked)

plt.xlabel(x)

plt.ylabel(y)

' b) b' U4 E9 a, E" [/ e# X( j

cbar = plt.colorbar()

3 w! a; I- A# n$ b" k

cbar.ax.set_ylabel(Counts)

+ r0 V1 ]6 B! p0 i, g

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

* B. i$ ^( Q7 A6 Y% q4 h

plt.show()

# X$ f0 ~. p3 \3 s4 q# L1 I2 _4 @

打开凤凰新闻，查看更多高清图片

) U4 p! ?$ D# \7 P. O3 ~1 e2 U+ \

1 M* }- y3 {0 l, r, P3 n2 E

Example 2 ：双Y轴（Python）

" Z- h- H: g C% Q" T5 Q) v. {4 o, V

import csv

import pandas as pd

import matplotlib.pyplot as plt

5 A; \- _- J, }

from datetime import datetime

- t7 O9 F' z( p8 ^

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

time=data[date [AST]]

sal=data[salinity]

$ l6 t- G8 p2 e

tem=data[temperature [C]]

+ c, I% p+ ?* u1 Q: q4 K1 C

print(sal)

! R& n; J/ e& E5 L/ H" W" b# L* A

DAT = []

& F; M) \6 J$ @% W6 @! u! r

for row in time:

0 ^" K. {( {& ^% E; h

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

+ h: _7 `0 b+ ]; n( l- U0 @

#create figure

fig, ax =plt.subplots(1)

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

+ c, P; L6 F; F3 ^

plt.xlabel("Date [AST]")

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

- N m. I8 ^ S

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

1 f/ V ~1 i" g! t2 O

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

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

fig.autofmt_xdate(rotation=50)

# }% h v& ^+ p. K/ I4 J3 a

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

plt.twinx()

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

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

& K: W2 J1 \/ X

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

#To save your graph

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

plt.show()

Example 3：拟合曲线（Python）

' w, L' C9 L# O. {4 v

import csv

import numpy as np

import pandas as pd

from datetime import datetime

# @' e; R" t- c8 e8 K

import matplotlib.pyplot as plt

8 W7 ]4 ^1 \/ N( V! n7 h( z

import scipy.signal as signal

, F9 U+ Y% y5 K4 g9 c4 l

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

4 @/ T/ m/ r' {- t4 ?+ ]. \. X

time=data[date [AST]]

temp=data[temperature [C]]

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

DATE,decday = [],[]

3 P+ a" ~0 F5 Q, e U

for row in time:

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

2 s5 I( T, L9 P; Y, K* q

DATE.append(daterow)

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

! X' Q, _9 w" A/ {

# First, design the Buterworth filter

/ z3 h, N9 A0 N/ l

N = 2 # Filter order

Wn = 0.01 # Cutoff frequency

# p- G v- c' d" v" p

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

0 ^, Q# t! u$ H5 T4 B" m8 h

# Second, apply the filter

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

# Make plots

fig = plt.figure()

ax1 = fig.add_subplot(211)

plt.plot(decday,temp, b-)

; u1 _2 I" [- l, ^0 ^

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

plt.ylabel("Temperature (oC)")

plt.legend([Original,Filtered])

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

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

( k8 k2 ]6 d" h5 U

ax1 = fig.add_subplot(212)

* r: h$ k' R* ` @3 `6 C

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

plt.ylabel("Temperature (oC)")

plt.xlabel("Date")

plt.legend([Residuals])

4 p% O5 i# v( V( a' J+ E, U% e

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

plt.show()

Example 4：三维地形（Python）

# This import registers the 3D projection

( C2 m: S# Z5 e

from mpl_toolkits.mplot3d import Axes3D

2 i9 t; a# C' X$ o' K

from matplotlib import cbook

6 I# Y( r- s# n3 z" R

from matplotlib import cm

' O' K# L# W( _

from matplotlib.colors import LightSource

: Q" E5 S8 ^' X; t

import matplotlib.pyplot as plt

2 U" s n# Y1 `; m. g) c) l

import numpy as np

; D: T- A1 K- z+ y2 b. @9 l/ O

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

with np.load(filename) as dem:

" Z+ h/ ?% L C& J6 A" o& l

z = dem[elevation]

nrows, ncols = z.shape

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

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

' F" b7 c' I' V* P

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

0 J: v2 X0 _' P% r$ J* h

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))

# {8 Z& v# j% Y9 p3 A, D

ls = LightSource(270, 45)

+ E: `' O7 x) k" N$ h6 O- q

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

$ x8 _: b& d4 e3 ~' P+ B+ e( G5 b

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

7 x6 C6 `& ~, b" c

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

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

plt.show()

8 N& u7 c4 C# I7 W1 }

Example 5：三维地形，包含投影（Python）

( o& ^4 n3 K6 }/ A, C' }9 ? 9 u7 e8 j- N8 I" ]

Example 6：切片，多维数据同时展现（Python）

# L- F' D+ ~5 L4 z% l5 L 1 J7 u9 c7 A, e

Example 7：SSH GIF 动图展现（Matlab）

Example 8：Glider GIF 动图展现（Python）

5 U- E t$ Z$ |+ L( t N

Example 9：涡度追踪 GIF 动图展现

& I$ z+ o7 O6 x, c" {* x

3 x: _ V$ D' q: r# j z% `1 h: G

; C# A" z2 \8 A0 T- x- m