Back Matter

Search in book...
Toggle Font Controls
Create new playlist

Name your new playlist

Playlist description (optional)
Sign In

Email address

Password

Forgot Password?

or

Continue with Facebook

Continue with Google
Sign Up

Full Name

Email address

Confirm Email Address

Password

or

Continue with Facebook

Continue with Google

Appendix A: MATLAB/Python Recipe Index

The following recipes show hybrid MATLAB/Python solutions to problems the core MATLAB product cannot solve on its own. Other solutions are, of course, possible using toolboxes, posts to the File Exchange,¹ projects on GitHub, and so on.

Several solutions use the py2mat.m Python-to-MATLAB variable converter (Appendix D), so this utility should be placed on the MATLAB search path.

Section	Title
7.2	Read YAML files
7.3	Write YAML files
7.4	Read an ini file
7.5	Write an ini file
7.8	Write an .xlsx file
7.12	CRUD with a SQLite database
7.18	TCP server
7.20	CRUD with a PostgreSQL database
7.21	CRUD with a MongoDB database
7.22	Interact with Redis
11.6	Curve fitting with differential_evolution()
11.8	Weighted least squares
11.11	Solving simultaneous nonlinear equations
11.15	Using SymPy in MATLAB
11.16	Compute Laplace transforms
11.18	Using pint in MATLAB
12.6	Overlay contours on globe with Cartopy
12.7	Shade map regions by value with Cartopy
13.13	Maps with GeoPandas
14.12	Accelerating MATLAB with Python on a single computer
14.14	Accelerating MATLAB with Python on multiple computers

Appendix B: Generating Sample Data with Faker

Data-intensive algorithms need data, and lots of it, to test algorithmic correctness, assess performance, track memory use, estimate scale limits, and so on. Some algorithms need nothing more than large random matrices (Section 11.1.6.14), while others require realistic structured data.

The Python module faker² is a great resource for creating structured data. It has functions that can generate endless streams of names, addresses, phone numbers, nonsensical text, and many others. Table B-1 lists many of these functions.

Table B-1

A subset of data creation methods in faker

address( )	file_extension( )	mac_processor( )
am_ pm( )	file_ name( )	military_ ship( )
bank_ country( )	file_ path( )	mime_ type( )
bban( )	first_ name( )	month( )
boolean( )	fixed_ width( )	month_name( )
bs( )	format( )	msisdn( )
building_number( )	free_ email( )	name( )
catch_ phrase( )	free_email_domain( )	paragraph( )
city( )	future_date( )	password( )
color( )	future_datetime( )	past_date( )
company( )	hex_color( )	phone_number( )
company_email( )	hexify( )	port_number( )
coordinate( )	hostname( )	postalcode( )
country( )	http_method( )	pricetag( )
country_code( )	image_ url( )	provider( )
credit_card_expire( )	ios_platform_token( )	sentence( )
credit_card_full( )	ipv4( )	ssn( )
credit_card_number( )	ipv4_network_class( )	state( )
credit_card_provider( )	ipv4_private( )	street_address( )
cryptocurrency( )	ipv4_public( )	street_name( )
csv( )	ipv6( )	street_suffix( )
currency( )	isbn10( )	suffix( )
date( )	isbn13( )	swift( )
date_between( )	iso8601( )	text( )
date_of_birth( )	items( )	time( )
date_this_century( )	itin( )	timezone( )
date_this_decade( )	job( )	tld( )
date_this_month( )	language_name( )	unix_device( )
date_this_year( )	last_name( )	unix_partition( )
day_of_month( )	latitude( )	unix_time( )
day_of_week( )	latlng( )	uri( )
dga( )	lexify( )	url( )
domain_name( )	license_plate( )	user_name( )
ean( )	linux_processor( )	uuid4( )
ean13( )	locale( )	word( )
ean8( )	longitude( )	year( )
email( )	mac_address( )	zipcode( )

This brief example shows how to create five names and phone numbers:

Python:

from faker import Faker

fake = Faker()

Faker.seed(123)

for i in range(5):

name = fake.name()

phone = fake.phone_number()

print(f'{name:30s} {phone}')

# output:

Brandon Russell 068-855-0225

Mia Martin 001-196-105-7101x220

Jeffrey Woods (045)873-9895x068

Joshua Hogan 786.580.2174x255

Stacey Sharp 503-915-0536x8804

Setting the random generator seed to a fixed value with Faker.seed() enables repeatable values—although the same seed value used on different Python installations will not necessarily yield the same output.

Faker is used in the Python and MATLAB programs that demonstrate interaction with the MongoDB database engine in Sections 7.19.2 and 7.21.

Appendix D: Python-to-MATLAB and MATLAB-to-Python Variable Converters

The MATLAB function py2mat() converts native Python variables returned by calls to Python functions via MATLAB’s py module into native MATLAB variables. Such a conversion is necessary if you want to perform MATLAB operations on values returned from Python.

mat2py() does the opposite and can be useful for converting MATLAB variables into equivalents that can be understood by Python functions.

Conversions from Python native to MATLAB native are demonstrated with the Python file demo_py2mat.py which has two functions. simple_vars() returns an integer, a floating-point value, a string, a list, a dictionary, a tuple, and a 2D NumPy array. nested_var() returns a list containing scalars and a dictionary which itself has compound values such as a NumPy array and a tuple.

Python:

# file: code/matlab_py/demo_py2mat.py

import numpy as np

def simple_vars():

I = 42

F = np.e

S = 'Red'

L = [1, 2.2, 'green']

D = {'A' : 65, 'B' : 66, 'Z' : 90}

T = (225, 'blue')

N = F*np.arange(12).reshape(3,4)

return I, F, S, L, D, T, N

def nested_var():

N = np.arange(12).reshape(3,4)

return [1, 2.2, 'green',

{'A' : 65, 'B' : 66,

'N' : N,

'Z' : (225, 'blue')},

27.8]

def arrays():

arr_int8 = np.array([1, 2], dtype=np.int8)

arr_uint8 = np.array([1, 2], dtype=np.uint8)

arr_int16 = np.array([1, 2], dtype=np.int16)

arr_uint16 = np.array([1, 2], dtype=np.uint16)

arr_int32 = np.array([1, 2], dtype=np.int32)

arr_uint32 = np.array([1, 2], dtype=np.uint32)

arr_int64 = np.array([1, 2], dtype=np.int64)

arr_uint64 = np.array([1, 2], dtype=np.uint64)

arr_float16 = np.array([1, 2], dtype=np.float16)

arr_float32 = np.array([1, 2], dtype=np.float32)

arr_float64 = np.array([1, 2], dtype=np.float64)

arr_complex64 = np.array([1-1j, 2-2j], dtype=np.complex64)

arr_complex128 = np.array([1-1j, 2-2j], dtype=np.complex128)

return arr_int8 ,

arr_uint8 ,

arr_int16 ,

arr_uint16 ,

arr_int32 ,

arr_uint32 ,

arr_int64 ,

arr_uint64 ,

arr_float16 ,

arr_float32 ,

arr_float64 ,

arr_complex64,

arr_complex128

First, let’s see what happens if we call these functions from MATLAB and then try to use the Python return values without conversion:

MATLAB 2020b:

>> x = py.demo_py_to_mat.simple_vars();

>> x

x =

Python tuple with no properties.

(42, 2.718281828459045, 'Red', [1, 2.2, 'green'],

{'A': 65, 'B': 66, 'Z': 90}, (225, 'blue'),

array([[ 0. , 2.71828183, 5.43656366, 8.15484549],

[10.87312731, 13.59140914, 16.30969097, 19.0279728 ],

[21.74625463, 24.46453646, 27.18281828, 29.90110011]]))

>> I = x{1}; F = x{2}; S = x{3}; L = x{4}; D = x{5}; T = x{6}; N = x{7};

>> whos

Name Size Bytes Class Attributes

D 1x1 8 py.dict

F 1x1 8 double

I 1x1 8 py.int

L 1x3 8 py.list

N 1x1 8 py.numpy.ndarray

S 1x3 8 py.str

T 1x2 8 py.tuple

x 1x7 8 py.tuple

The double-precision floating-point variable F is the only returned value that comes over directly as a native MATLAB variable. The Python integer, string, and NumPy array mostly behave like normal MATLAB variables or can be easily coerced into MATLAB-native variables with calls to int64(), double(), and string().

MATLAB 2020b:

>> I

I =

Python int with properties:

denominator: [1×1 py.int]

imag: [1×1 py.int]

numerator: [1×1 py.int]

real: [1×1 py.int]

>> I3 = I * 3

I3 =

126

>> S

S =

Python str with no properties.

Red

>> fprintf('S = %s ', S)

S = R

S = e

S = d

>> fprintf('S = %s ', string(S))

S = Red

>> N

N =

Python ndarray:

0 2.7183 5.4366 8.1548

10.8731 13.5914 16.3097 19.0280

21.7463 24.4645 27.1828 29.9011

Use details function to view the properties of the Python object.

Use double function to convert to a MATLAB array.

>> sin(N)

Check for missing argument or incorrect argument data type in call to function 'sin'.

>> sin(double(N))

ans =

0 0.4108 -0.7490 0.9551

-0.9925 0.8547 -0.5661 0.1775

0.2425 -0.6196 0.8873 -0.9984

The list and dictionary can be converted to a MATLAB cell array and struct via cell() and struct():

MATLAB 2020b:

>> C = cell(L)

C =

1×3 cell array

{1×1 py.int} {[2.2000]} {1×5 py.str}

>> Str = struct(D)

Str =

struct with fields:

A: [1×1 py.int]

B: [1×1 py.int]

Z: [1×1 py.int]

These conversions are problematic though; although the containers are MATLAB native, their contents are not. Another wrinkle is the restrictive set of rules governing field names in a MATLAB struct. A MATLAB struct’s field name must be a valid MATLAB variable name which means it must begin with a letter followed by letters, digits, and/or underscores. However, Python dictionary keys, which struct() function will try to use for MATLAB field names, have no such restrictions; dictionary keys may be numeric, strings with arbitrary contents, even tuples of other items.

Let’s look at the more complex example:

MATLAB 2020b:

>> x

x =

Python list with no properties.

[1, 2.2, 'green', {'A': 65, 'B': 66, 'N': array([[ 0, 1, 2, 3],

[ 4, 5, 6, 7],

[ 8, 9, 10, 11]]), 'Z': (225, 'blue')}, 27.8]

Superficially, it looks like x comes across cleanly. Internally, though, these are all native Python entities. To access the numeric array, for example, we have to use a Python dictionary accessor: x4.get(’N’).

My solution to this problem is the function py2mat(). Within MATLAB, it recursively descends a Python-native variable until it reaches primitives (integers, floats, strings, datetimes) and then applies MATLAB’s conversion functions to build up a corresponding MATLAB variable. Care is taken to map Python dictionary keys to valid MATLAB field names; key names that violate MATLAB variable name rules are replaced by the string K<i><name> where <i> is a six-digit, zero-padded index of key position and <name> is as much of the original key name as can be preserved.

The returned value from py2mat.m is a pure MATLAB variable, cell, or struct.

MATLAB 2020b:

% file code/matlab_py/py2mat.m

function [x_mat] = py2mat(x_py)

Im = @py.importlib.import_module;

np = Im('numpy');

% Convert a native Python variable to an equivalent

% native MATLAB variable.

switch class(x_py)

% Python dictionaries

case 'py.dict'

try

% the obvious dict -> struct conversion only works

% if the key string can be a MATLAB variable

x_mat = struct(x_py);

catch EO

% A failure most likely means a Python key is not

% a proper MATLAB variable name. Go through them

% individually

key_index = int64(0);

for key = cell(py.list(x_py.keys()))

key_index = key_index + 1;

v_name = string(key);

if isvarname(v_name)

x_mat.(v_name) = x_py.get(v_name);

else

% this key can't be used; replace it

fixed = sprintf('K%06d_', key_index) + ...

regexprep(string(key),'W','_');

x_mat.(fixed) = x_py.get(v_name);

end

fields = fieldnames(x_mat);

for i = 1:length(fields)

new_var = py2mat(x_mat.(fields{i}));

x_mat = setfield(x_mat,fields{i}, new_var);

end

% NumPy arrays and typed scalars

case 'py.numpy.ndarray'

switch string(x_py.dtype.name)

case "float64"

x_mat = x_py.double;

case "float32"

x_mat = x_py.single;

case "float16"

% doesn't exist in matlab, upcast to float32

x_mat = single(x_py.astype(np.float32));

% only reals and logicals can be cast to arrays so

% have to cast the rest to either single or double

case "uint8"

x_mat = uint8(x_py.astype(np.float32));

case "int8"

x_mat = int8(x_py.astype(np.float32));

case "uint16"

x_mat = uint16(x_py.astype(np.float32));

case "int16"

x_mat = int16(x_py.astype(np.float32));

case "uint32"

x_mat = uint32(x_py.astype(np.float32));

case "int32"

x_mat = int32(x_py.astype(np.float32));

case "uint64"

x_mat = uint64(x_py.astype(np.float64));

case "int64"

x_mat = int64(x_py.astype(np.float64));

% Complex types require a math operation to coerce the

% real and imaginary components to contiguous arrays.

% Use "+0" for minimal performance impact. Without this,

% complex creation fails with

% Python Error: ValueError: ndarray is not contiguous

case "complex64"

x_mat = complex(single(x_py.real+0), single(x_py.imag+0));

case "complex128"

x_mat = complex(double(x_py.real+0), double(x_py.imag+0));

case "complex256"

fprintf('py2mat: MATLAB does not support quad precision complex ')

x_mat = [];

return

otherwise

% gets here with np.float16, custom dtypes

fprintf('py2mat: %s not recognized ', ...

string(x_py.dtype.name));

x_mat = [];

return

end

% Scipy sparse matrices

case {'py.scipy.sparse.coo.coo_matrix', ...

'py.scipy.sparse.csr.csr_matrix', ...

'py.scipy.sparse.csc.csc_matrix', ...

'py.scipy.sparse.dok.dok_matrix', ...

'py.scipy.sparse.bsr.bsr_matrix', ...

'py.scipy.sparse.dia.dia_matrix', ...

'py.scipy.sparse.lil.lil_matrix', }

ndims = x_py.get_shape();

if length(ndims) ˜= 2

fprintf('py2mat: can only convert 2D sparse matrices ')

x_mat = [];

return

end

nR = int64(ndims{1});

nC = int64(ndims{2});

x_py = x_py.tocoo();

values = py2mat(x_py.data);

if isempty(values)

% gets here if trying to convert a complex256 sparse matrix

return

end

% add 1 to row & col indices to go from 0-based to 1-based

x_mat = sparse(single(x_py.row)+1, single(x_py.col)+1, values, nR, nC);

% Python sets, tuples, and lists

case {'cell', 'py.tuple', 'py.list'}

[nR, nC] = size(x_py);

x_mat = cell(nR,nC);

for r = 1:nR

for c = 1:nC

x_mat{r,c} = py2mat(x_py{r,c});

end

% Python strings

case 'py.str'

x_mat = string(x_py);

% Python integers

case 'py.int'

x_mat = x_py.int64;

% Python floats (float64) -- same as matlab double

case 'double'

x_mat = x_py;

case 'py.datetime.datetime'

x_mat = datetime(int64(x_py.year), ...

int64(x_py.month), ...

int64(x_py.day), ...

int64(x_py.hour), ...

int64(x_py.minute),...

int64(x_py.second),...

int64(x_py.microsecond));

% punt

otherwise

% return the original item? nothing?

fprintf('py2mat: type "%s" not recognized ', ...

string(x_py.dtype.name));

x_mat = [];

% x_mat = x_py;

end

The MATLAB-to-Python converter, mat2py.m, is as follows:

MATLAB 2020b:

% code/matlab_py/mat2py.m

% Convert a MATLAB variable to an equivalent Python-native variable.

function [x_py] = mat2py(x_mat)

Im = @py.importlib.import_module;

np = Im('numpy');

sp = Im('scipy.sparse');

dt = Im('datetime');

tz = Im('dateutil.tz');

x_py = np.array({});

switch class(x_mat)

case 'char'

x_py = py.bytes(x_mat,'ASCII');

case 'string'

x_py = py.str(x_mat);

case 'datetime'

int_sec = int64(floor(x_mat.Second));

frac_sec = x_mat.Second - double(int_sec);

micro_sec = int64(round(1e6 * frac_sec));

if length(x_mat.TimeZone)

tzinfo = tz.gettz(x_mat.TimeZone);

else

tzinfo = py.None;

end

x_py = dt.datetime(int64(x_mat.Year), int64(x_mat.Month), ...

int64(x_mat.Day) , int64(x_mat.Hour) , ...

int64(x_mat.Minute), int64(x_mat.Second), ...

micro_sec, tzinfo)

case {'double', 'single'}

if issparse(x_mat)

if ndims(x_mat) ˜= 2

fprintf('mat2py: can only convert 2D sparse matrices ')

return

end

[nR,nC] = size(x_mat);

[i,j,vals] = find(x_mat);

% subtract 1 to go from 1-based to 0-based indices

py_I = np.array(int64(i)-1);

py_J = np.array(int64(j)-1);

py_vals = mat2py(vals);

py_dims = py.tuple({int64(nR), int64(nC)});

py_IJ = py.tuple({py_I, py_J});

V_IJ = py.tuple({py_vals, py_IJ});

x_py = sp.coo_matrix(V_IJ,py_dims);

elseif ismatrix(x_mat)

if isreal(x_mat)

x_py = np.array(x_mat);

else

x_py = np.array(real(x_mat)) + 1j*np.array(imag(x_mat));

end

otherwise

fprintf('mat2py: %s conversion is not implemented ', class(x_mat))

end % switch

end

Appendix E: A Utility to Patch Cartopy to Use requests

Run the following program in a matpy conda environment to modify Cartopy’s img_tiles.py file to use the requests module to download image tiles. This allows MATLAB calls to Cartopy to download map image tiles without errors. Additionally, on Windows it modifies geos.py to set a library path correctly and avoid a DLL error.

Python:

#!/usr/bin/env python

# code/bin/patch_cartopy.py

# Change the active cartopy installation so that:

# 1. Replace urllib with requests to download image tiles in

# cartopy/io/img_tiles.py

# 2. [Windows only] Replace variable sys.prefix with expanded

# string in shapely/geos.py when searching for geos_c.dll

# so it can be found by MATLAB.

import sys

import pathlib

import shutil

import re

import cartopy.io

import shapely

def fix_geos(): # {{{

"""

On Windows, modify shapely/geos.py when the CONDA_PREFIX

string is empty so that

sys.prefix

is hardcoded to the correct virtual environment path

to load the geos_c.dll.

"""

if sys.platform != 'win32':

print('fix_geos() exit since this is not Windows')

return

shapely_dir = shapely.__path__

if not shapely_dir:

print(f"Unable to determine shapely's location, exit")

return

P = pathlib.Path(shapely_dir[0]) / 'geos.py'

if not P.exists():

print(f"{str(P)} doesn't exist, exit")

return

lines_in = P.read_text().split(' ')

found_win = False

for L in lines_in:

if re.search(r'_lgeoss*=.*?sys.prefix,.*?geos_c.dll', L):

found_win = True

if not found_win:

print(f"{str(P)} does not use sys.prefix to determine "

f"geos_c.dll path, nothing done.")

return

shutil.copy(str(P), f'{str(P)}.bak') # backup the original version

print(f'Wrote backup {str(P)}.bak')

lines_out = []

found_win = False

did_fix = False

for L in lines_in:

if did_fix:

lines_out.append( L )

continue

if re.search(r"s*==s*'win32':", L):

found_win = True

lines_out.append( L )

elif found_win and L.endswith("'geos_c.dll'))"):

did_fix = True

lines_out.append( f'#{L}' )

lines_out.append( L.replace('sys.prefix', f"r'{sys.prefix}'") )

else:

lines_out.append(L)

P.write_text(' '.join(lines_out))

print(f'Wrote updated {str(P)}')

# }}}

def fix_img_tiles(): # {{{

"""

Modify cartopy/io/img_tiles.py to use

requests.get()

instead of

urllib.request.Request()

to download an image tile.

"""

cartopy_io_dir = cartopy.io.__path__

if not cartopy_io_dir:

print(f"Unable to determine cartopy.io's location, exit")

return

P = pathlib.Path(cartopy_io_dir[0]) / 'img_tiles.py'

if not P.exists():

print(f"{str(P)} doesn't exist, exit")

return

lines_in = P.read_text().split(' ')

if 'import requests' in lines_in:

print(f'{str(P)} already uses requests module, nothing done')

return

shutil.copy(str(P), f'{str(P)}.bak') # backup the original version

print(f'Wrote backup {str(P)}.bak')

New = [

'request = requests.get(url, stream=True)',

'request.raw.decode_content = True',

'img = Image.open(request.raw)',

]

lines_out = []

request_start = 0

for L in lines_in:

if re.search(r'ˆimports+cartopys*$', L):

lines_out.append('import requests')

lines_out.append(L)

elif re.search(r'ˆs+requests+=s*', L):

request_start = 1

lines_out.append(f'#{L}')

spaces = ' ' * L.find('r') # 12

lines_out.append(f'{spaces}{New[0]}')

lines_out.append(f'{spaces}{New[1]}')

lines_out.append(f'{spaces}{New[2]}')

elif 0 < request_start < 5:

request_start += 1

lines_out.append(f'#{L}')

else:

lines_out.append(L)

P.write_text(' '.join(lines_out))

print(f'Wrote updated {str(P)}')

# }}}

def main():

fix_geos()

fix_img_tiles()

pass

if __name__ == "__main__": main()

Index

Altair

Anaconda

Animations

ball bouncing

key elements

matplotlib

Application program interface (API)

arange() function

Area plots

imagesc()

imshow()

bilinear interpolation

prefixes

astimezone() method

Austin

Basic linear algebra subroutines (BLAS)

Block compressed row (BSR)

Bokeh

Bridge modules

built-in scope

Cartopy

cartographic transformation

drawing lines, map

MATLAB

overlay contours, globe

projection functions

shade map regions, value

Cell arrays

Class

circle

constructor method

custom exceptions

custom printers

__init__() function

instances

private vs. public

variables

Cloud service

Coiled’s cloud

performance

running

Command-line input

argparse

function arguments

sys module

Comma-separated value (CSV)

Comments

block comment

definition

docstrings

Communication-intensive parallel program

Compressed Sparse Column (CSC)

Compressed Sparse Row (CSR)

Computer Algebra System (CAS)

conda package manager

Conda environment

conda-meta directory

conda update

Converters

create_dataset() function

Create, read, update, delete (CRUD)

Curve fitting

differential_evolution()

bridge module

compute_ABCD() function

cost function

matlab.engine

py module

solution

fit models

curve_fit()

differential_evolution()

r2_score() function

scipy.optimize module

scipy.optimize.curve_fit()

test data

linear regression

polynomials

Cython

MATLAB

parallel for loop

performance

Python

Dask

execution paradigm

Mandelbrot set

multiple computer

parallel MATLAB

prime factor

scheduler

single computer

Data

cleaning

columns

changing

data types

renaming

timestamps

Data classes

convenience mechanism

dynamic modification

entity relationship

field values

calling mag()

MATLAB vs. Python

point

traversing

type enforcement

type validation, Pydantic

Data containers

MATLAB

nonscalars

Python

Dataframes.

SeeTables

Dates

datetime

string

time deltas

datestr() function

datetime functions

datetime module

datetime object

definition

ISO 8601 time string

Julian date

Unix epoch seconds

Debugging

Python code

diary mode

do_work()

functions

logging

steps

trace module

deepcopy() function

Dictionaries

add keys

creation

datasets

iterating over keys

for key in dict

sorting by key

sorting by values

key existence

.get() method

key collision

KeyError

.setdefault() method

test_key

limitation

vs. lists

Python

country-to-capital city relationship

subscript

struct

tuples

values

Dictionary of keys (DOK)

Differential equations

Direct frequency

Directories

contents

creation

deletion

expressions

trees

.csv files

dir()

find() functions

generators

os.walk()

output

printing

traversing

.xlsx files

Duffing oscillator equation

eig() eigenvalue function

enclosing scope

enumerate() function

Enumerations

env command

Excel .xls, .xlsx

definition

extract columns

extract rectangular block

extract rows

get worksheet names

reading files

writing

writing file

Exceptions

executemany() function

External executables

bytes-Like Object

CalledProcessError

ffmpeg

os.environ object

Result

Result.returncode

Status

subprocess module

subprocess.CalledProcessError

subprocess.run()

subsystem.run()

system() function

f2py command

File system

deletion

finding

MATLAB

Python

Finite element (FE)

FlameGraphs

For loops

definition

examples

exits

nested loops

variables

Fortran-compiled module

Functions

comparison operators

decorators

keyword arguments

left-hand side argument count

lists/dicts/arrays

pass by values/pass by reference

type annotation/argument validation

variable arguments

Generators

definition

range()

yield, next()

geo* functions

GeoPandas

create plot

dataframes

data type

geographic data

join

maps

bridge functions

bridge module

issues

iterrows()

plotting

read_file() function

shapefile

Zillow research

zip code

global keyword

global scope

h5dump command-line utility

HDF5

datasets

netcdf4

reading files

reading/writing dataset attributes

writing files

Hierarchical Data File (HDF)

High performance computing (HPC)

baseline performance

Mandelbrot set

multiprocessing

profiling tools

austin

drawbacks

flameGraphs

scalene

reference hardware/OS

reference problems

Mandelbrot set

If statements

Boolean expressions/operators

range tests

variables

imshow() function

Indentation

brackets/parentheses

index chaining

indexing methods

index strides

MATLAB and Python

start/end/negative indices

Tabs

zero-based indexing/index ranges

Index masks

creation

mathematical operations

uses

input() function

Installation, MATLAB

creating/running Python program

downloads

package

post-install configuration/checkout

toolboxes

Interpolation

one-dimensional

extrapolation

linear

spline

two-dimensional

cubic/linear methods

evaluation

function

grid

griddata()

meshgrid()

np.meshgrid()

testing

values

Ints

ipython, IDEs

application navigator

autoload modules

version

iterables

iterrows()

JavaScript Object Notation (JSON)

complex data structures

MATLAB to Python

Python to MATLAB

Just-in-time (JIT)

Keyword arguments

Lambda function

Linear algebra

eigenvalue problems

linear equations

MATLAB vs.Python

SVD

Linear indexing

Linear programming (LP) problems

integer-valued

linprog()

bridge module

coefficients

MATLAB

Python solution

relationships

sol object

solution status

objective function

PuLP module

server properties

Linux

list() function

List of lists (LIL)

Lists

local scope

Local variables, then enclosing, global, and built-in (LEGB)

Long expressions

macOS

Mandelbrot set

MATLAB .mat files

data containers

definition

reading files

v7.3

writing files

MATLAB

Anaconda’s Python distribution

call functions

existing session

new session

cell arrays

configuration

Linux/macOS

matlab command

path

pyenv command

Windows

container.Map

definition

eigensolvers

environment variables

errors

functions

horzcat() function

hybrid environment

incompatible library function

install engine

install module

libraries

lower-level libraries

maps

matpy

py module

Python

pyversion()

software

software engineering

toolboxes

MATLAB-native variable

MATLAB performance

MATLAB/Python interaction

file I/O

system calls

TCP/IP exchange

matplotlib module

Matrix slices

maxsplit keyword

MB() function

Mixed-type mathematical expressions

Modules/packages

definition

def main()

dependency conflicts/virtual environments

installation

namespace

search path

MongoDB

bridge module

delete

read

updates

mset function

multiprocessing module

Natural Earth project

ncinfo command-line tool

Network communication protocols

HTTP, HTTPS

MATLAB client

Python as a web server

TCP/IP

TCP server

Numba

keyword argument

limitations

Mandelbrot program

MATLAB

parallel for loop

performance

Python

Numeric arrays

constants

conversion

MATLAB

commands

functions

single-precision floats

type

NumPy

definition

eigensolvers

NumPy and SciPy

NumPy arrays

decimal to binary

hexadecimal to decimal

MATLAB

scalars

operations

addition/subtraction

elementwise

Hermitian

transpose

terms of interest

find() function

smallest term

sorted data

text files, reading

typecasting

NumPy arrays

binary files

endian conversions

metadata

reading

binary to decimal

broadcasting

definition

distance matrix

memory

normalize vectors

rules

complex scalars

creation

diagonal matrices

identity matrices

inflating matrices

linspace

logspace

meshgrid

nan() function

np.array() function

ones

random

range() function

stacking

test matrices

triangle matrices

zeros() function

data types

decimal to hexadecimal

definition

dimensions

dtype argument

formatting

import

MATLAB

vs. MATLAB matrices

column major

dimensions

numeric literals

.resize() method

scalars

np.array() function

NumPy module

operations

bitwise

elementwise

Hermitian

multiplication

transpose

submatrices

extraction

updation

terms of interest

closest term

find() function

longest term

sorted data

values

text files

reading

writing

typecasting

whos command

Object-oriented programming (OOP)

class method

collisions

computational performance

performance implications

simulation

vectorized solution

open() function

Operating system

computer’s hardware

CPU cycles

environment variables

independent method

MATLAB

memory

numerical analyses

program iterates

Python

random vector

setenv() function

simulations

TypeError error

pack() function

Pandas

Parallel program

pathlib.Path() object

Pickle files

pinned

Pivot tables

multilevel aggregation

single-level aggregation

Plotly

Plots

batch mode

editing

line

point

stack

Plotting

areas

Cartopy

exploration

lines

points

Point/line plots

date and time

histograms

MATLAB

matplotlib commands

metric vs. US customary

Python

saving

command

formats

MATLAB

matplotlib’s function

Python

subplots() function

Young’s modulus

PostgreSQL

prange() function

Prime factors

print() function

purchase_orders

py2mat() function

Python

deployment

distributions

list

.append() method

comprehension

.count() method

creation

empty

.extend () method

extract range

find() function

IndexError

indexing

.insert() method

length

list comprehension

membership test

merging

methods

nested containers

.pop() method

.remove() method

.reverse() method

sorting

subset of items

unmerging

ValueError exception

zipped

modules, importing

multiple computers

substantial cost

sys.path

alias

append()

insert()

tuples

creation

hashable

parentheses

vs. Python lists

wrapping

variables

Python Package Index (PyPI)

Python-specific aspects

Pythran

comments

MATLAB

parallel for loop

performance

Qt library

range() function

.ravel() method

Read-evaluate-print-loop (REPL)

readmatrix() function

read_text() method

Redis

Regression

confidence intervals

least squares

ordinary

weighted

prediction intervals

Regular expressions

Remote computer

re.split()

Roots

multivariate

nonlinear equations

univariate

rstrip() method

Satellite

save() command

Scalars

Scalene

SciPy

coordinate format (COO)

definition

eigensolvers

functions

optimization module

scipy.interpolate module

scipy.optimize module

sparse module

sparse matrix

scipy.io.loadmat()

Seaborn

Sea surface temperature (SST)

CARTOPY_USER_BACKGROUNDS

data

earth_veg_index.jpeg

orthographic projection

Second-order differential equation

.send() and .recv() functions

Set operations

MATLAB, unique() function

Python, .issuperset() method

iteration/casting

set() function

Simulated annealing

Annealer-derived class

automatic estimator

compute_distances() function

concept

.distance_matrix attribute

energy() method

parameters

requirements

.state attribute

TSP

Singular value decomposition (SVD)

sleeper() and add_numbers()

Solver code

Solver function

South American wheat production

add_geometries() function

borders

Cartopy

cases

country names

database

low resolution file

polygonal regions

Sparse matrices

BSR

creation

COO

csr_matrix()

dense matrix

dense representation

COO

dimension

index values

indices/values

row/columns

sparse() function

.sum_duplicates() method

CSC

CSR

definition

diagonal (DIA)

DOK

formats/capabilities

input and output (I/O)

component parts

mat files

pickle files

LIL

linear algebra

eigenvalue problem

finite element–based symmetric

linear equations

matrix summary

matrix-vector product

scipy.sparse.linalg module

Sturm sequence number

MATLAB

SciPy

.split() method

SQLite

Stack plots

start_new_order() function

startup directory

startup.m file

strftime() function

Strings

byte arrays

character arrays

formatting

operations

append

commands

conversion

length

<TAB> key

method chaining

repeat

replace characters

regular expressions

purposes

replacing

separation

substrings

splitting

templates

testing

equality

starting characters

substring

trailing characters

struct() function

struct’s pack() function

Structured data

classes

hasattr() function

isfield() function

namedtuple method

SimpleNamespace method

Structured Query Language (SQL)

Symbolic mathematics

char attribute

collisions

derivatives

integrals

laplace_transform() function

linear algebra

s.doit()

series

solving equations

SymPy code generators

variables

sys.exc_info() function

Tables

columns

adding

deletion

names

types

creation

.describe() method

groupsummary() method

head() function

indexing

.info() method

joins

flavors

food costs

food type

how= keyword

wxp column

MATLAB’s table

pets.csv file

read_csv()

string/floating-point types

rows

iteration

sorting

size

subsets

all columns, selected rows

all rows, selected columns

filter rows, conditional operations

selected rows, selected columns

summary() function

tail() function

.value_counts() method

Test matrices

arrow matrix

random sparse

arrays

spy() function

Text files

assignments

binary files

CSV

definition

NumPy’s genfromtxt() function

pandas

python module

ini files

I/O exceptions

JSON

MATLAB/Python

parsing text

operations

regex’s

re.split()

splitting string, substring

string search

stripping whitespace

pathlib, reading corrupted files

reading/writing numeric arrays

read ini file

STDOUT

write ini file

write YAML files

XML

lxml modules

with namespaces

pretty-printing

without namespaces

writing XML

YAML

Time

current

definition

module

string formats

tic/toc/%timeit

time.strftime() function

time.time() function

Timezones

.tobytes() method

to_datetime() function

.to_excel() method

.tofile() method

Traveling Salesman Problem (TSP)

Traversing generators

Tuples, np.ones() function

Two-dimensional array

Unit systems

defining units

MATLAB

metric values

NumPy array

pint/Symbolic Toolbox

Python

Unix epoch seconds

urlread() function

Variable assignment

in-place update operators

languages

print expressions

walrus operator

Vectorization

Virtual environments

Anaconda

commands

definition

matpy

open source languages

which command

While loops

whosmat() function

write() method

$max aggregation

yield keyword

Young’s modulus

zoneinfo

zoneinfo module

..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.

Table of Contents for Back Matter

Create new playlist

Sign In

Sign Up

Table of Contents for
Back Matter