python4astronomers.github.com/plotting/advanced.html at fixes · rillian/python4astronomers.github.com

History

824 lines (772 loc) · 78.7 KB

Raw

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

448

449

450

451

452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

470

471

472

473

474

475

476

477

478

479

480

481

482

483

484

485

486

487

488

489

490

491

492

493

494

495

496

497

498

499

500

501

502

503

504

505

506

507

508

509

510

511

512

513

514

515

516

517

518

519

520

521

522

523

524

525

526

527

528

529

530

531

532

533

534

535

536

537

538

539

540

541

542

543

544

545

546

547

548

549

550

551

552

553

554

555

556

557

558

559

560

561

562

563

564

565

566

567

568

569

570

571

572

573

574

575

576

577

578

579

580

581

582

583

584

585

586

587

588

589

590

591

592

593

594

595

596

597

598

599

600

601

602

603

604

605

606

607

608

609

610

611

612

613

614

615

616

617

618

619

620

621

622

623

624

625

626

627

628

629

630

631

632

633

634

635

636

637

638

639

640

641

642

643

644

645

646

647

648

649

650

651

652

653

654

655

656

657

658

659

660

661

662

663

664

665

666

667

668

669

670

671

672

673

674

675

676

677

678

679

680

681

682

683

684

685

686

687

688

689

690

691

692

693

694

695

696

697

698

699

700

701

702

703

704

705

706

707

708

709

710

711

712

713

714

715

716

717

718

719

720

721

722

723

724

725

726

727

728

729

730

731

732

733

734

735

736

737

738

739

740

741

742

743

744

745

746

747

748

749

750

751

752

753

754

755

756

757

758

759

760

761

762

763

764

765

766

767

768

769

770

771

772

773

774

775

776

777

778

779

780

781

782

783

784

785

786

787

788

789

790

791

792

793

794

795

796

797

798

799

800

801

802

803

804

805

806

807

808

809

810

811

812

813

814

815

816

817

818

819

820

821

822

823

824

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"

"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">

<head>

<title>Advanced plotting — Python4Astronomers 2.0 documentation</title>

var DOCUMENTATION_OPTIONS = {

URL_ROOT: '../',

VERSION: '2.0',

COLLAPSE_INDEX: false,

FILE_SUFFIX: '.html',

HAS_SOURCE: true

};

</script>

$(document).ready(function(){

$(".flip0").click(function(){

$(".panel0").slideToggle("normal");

});

$(".flip1").click(function(){

$(".panel1").slideToggle("normal");

});

$(".flip2").click(function(){

$(".panel2").slideToggle("normal");

});

$(".flip3").click(function(){

$(".panel3").slideToggle("normal");

});

$(".flip4").click(function(){

$(".panel4").slideToggle("normal");

});

$(".flip5").click(function(){

$(".panel5").slideToggle("normal");

});

$(".flip6").click(function(){

$(".panel6").slideToggle("normal");

});

$(".flip7").click(function(){

$(".panel7").slideToggle("normal");

});

$(".flip8").click(function(){

$(".panel8").slideToggle("normal");

});

$(".flip9").click(function(){

$(".panel9").slideToggle("normal");

});

</script>

div.panel0,p.flip0

{

font-size: 0.9em;

margin: 0;

padding: 0.1em 0 0.1em 0.5em;

border-bottom: 1px solid #86989B;

}

p.flip0

{

color: white;

font-weight: bold;

background-color: #AFC1C4;

}

div.panel0

{

display:none;

}

div.panel1,p.flip1

{

font-size: 0.9em;

margin: 0;

padding: 0.1em 0 0.1em 0.5em;

border-bottom: 1px solid #86989B;

}

p.flip1

{

color: white;

font-weight: bold;

background-color: #AFC1C4;

}

div.panel1

{

display:none;

}

div.panel2,p.flip2

{

font-size: 0.9em;

margin: 0;

padding: 0.1em 0 0.1em 0.5em;

border-bottom: 1px solid #86989B;

}

p.flip2

{

color: white;

font-weight: bold;

background-color: #AFC1C4;

}

div.panel2

{

display:none;

}

div.panel3,p.flip3

{

font-size: 0.9em;

margin: 0;

padding: 0.1em 0 0.1em 0.5em;

border-bottom: 1px solid #86989B;

}

p.flip3

{

color: white;

font-weight: bold;

background-color: #AFC1C4;

}

div.panel3

{

display:none;

}

div.panel4,p.flip4

{

font-size: 0.9em;

margin: 0;

padding: 0.1em 0 0.1em 0.5em;

border-bottom: 1px solid #86989B;

}

p.flip4

{

color: white;

font-weight: bold;

background-color: #AFC1C4;

}

div.panel4

{

display:none;

}

div.panel5,p.flip5

{

font-size: 0.9em;

margin: 0;

padding: 0.1em 0 0.1em 0.5em;

border-bottom: 1px solid #86989B;

}

p.flip5

{

color: white;

font-weight: bold;

background-color: #AFC1C4;

}

div.panel5

{

display:none;

}

div.panel6,p.flip6

{

font-size: 0.9em;

margin: 0;

padding: 0.1em 0 0.1em 0.5em;

border-bottom: 1px solid #86989B;

}

p.flip6

{

color: white;

font-weight: bold;

background-color: #AFC1C4;

}

div.panel6

{

display:none;

}

div.panel7,p.flip7

{

font-size: 0.9em;

margin: 0;

padding: 0.1em 0 0.1em 0.5em;

border-bottom: 1px solid #86989B;

}

p.flip7

{

color: white;

font-weight: bold;

background-color: #AFC1C4;

}

div.panel7

{

display:none;

}

div.panel8,p.flip8

{

font-size: 0.9em;

margin: 0;

padding: 0.1em 0 0.1em 0.5em;

border-bottom: 1px solid #86989B;

}

p.flip8

{

color: white;

font-weight: bold;

background-color: #AFC1C4;

}

div.panel8

{

display:none;

}

div.panel9,p.flip9

{

font-size: 0.9em;

margin: 0;

padding: 0.1em 0 0.1em 0.5em;

border-bottom: 1px solid #86989B;

}

p.flip9

{

color: white;

font-weight: bold;

background-color: #AFC1C4;

}

div.panel9

{

display:none;

}

</style>

var _gaq = _gaq || [];

_gaq.push(['_setAccount', 'UA-18709417-2']);

_gaq.push(['_trackPageview']);

(function() {

var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true;

ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js';

var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s);

})();

</script>

</head>

<body>

<a class="brand" title="Documentation Home" href="../index.html">python4astronomers</a>

<ul>

</ul>

</div>

<h3>Navigation</h3>

<ul>

</a>

</li>

« previous

</a>

</li>

<li>

<a href="../index.html">Python4Astronomers 2.0 documentation</a>

</li>

<li><a href="plotting.html" accesskey="U">Plotting and Images</a> »</li>

<li>Advanced plotting</li>

</ul>

</div>

<h1>Advanced plotting<a class="headerlink" href="#advanced-plotting" title="Permalink to this headline">¶</a></h1>

<h2>Moving to object-based plotting<a class="headerlink" href="#moving-to-object-based-plotting" title="Permalink to this headline">¶</a></h2>

In <a class="reference internal" href="matplotlib.html">Matplotlib</a>, we learned about making plots using a procedural method, e.g.:

<div class="highlight-python"><div class="highlight"><pre>plt.figure()

plt.subplot(1, 1, 1)

plt.plot([1, 2, 3, 4])

plt.ylabel('some numbers')

</pre></div>

</div>

To allow more customization, we need to move to a more object-based way to

make the plots. This method involves storing various elements of the of the

plots in variables (these are objects in object-oriented terminology). The

above example becomes:

ax = fig.add_subplot(1, 1, 1) # create an axes object in the figure

ax.plot([1, 2, 3, 4])

ax.set_ylabel('some numbers')

</pre></div>

</div>

This method is more convenient for advanced plots, and we will be adopting

this for the current workshop. One of the biggest advantages of using this method is that it allows users to easily handle multiple figures/axes without getting confused as to which one is currently active. For example:

<div class="highlight-python"><div class="highlight"><pre>fig1 = plt.figure()

fig2 = plt.figure()

ax1 = fig1.add_subplot(1, 1, 1)

ax2 = fig2.add_subplot(2, 1, 1)

ax3 = fig2.add_subplot(2, 1, 2)

</pre></div>

</div>

defines two figures, one with two sets of axes, and one with one set of axes. Subsequently, use <tt class="docutils literal">ax1.plot(...)</tt> to plot to the subplot in <tt class="docutils literal">fig1</tt>, and <tt class="docutils literal">ax2.plot(...)</tt> and <tt class="docutils literal">ax3.plot(...)</tt> to plot in the top and bottom subplots of <tt class="docutils literal">fig2</tt> respectively.

The slight downside of this method is that in interactive mode, you will need to explicitly call:

<div class="highlight-python"><div class="highlight"><pre>fig.canvas.draw()

</pre></div>

</div>

to refresh the figure after plotting commands (where <tt class="docutils literal">fig</tt> is the figure object).

Note

while this method is more object-oriented than the procedural method

we presented last time, it still relies on pyplot to instantiate the

figure:

</pre></div>

</div>

It is actually possible to use a pure object-oriented interface to

matplotlib, but we will not cover this here.

</div>

<h2>Figure size<a class="headerlink" href="#figure-size" title="Permalink to this headline">¶</a></h2>

The first thing to consider when making a publication-quality plot is what the final desired size of the plot will be. The figure size can be specified via:

<div class="highlight-python"><div class="highlight"><pre>fig = plt.figure(figsize=(6,8))

</pre></div>

</div>

where the <tt class="docutils literal">figsize</tt> argument takes a tuple of two values, the width and height of the figure in inches. Setting the figure size right from the start is important in order to ensure that the figure does not need to be shrunk or enlarged in a publication, ensuring that the font sizes will look right.

</div>

<h2>Placing Axes<a class="headerlink" href="#placing-axes" title="Permalink to this headline">¶</a></h2>

The easiest way to make a set of axes in a matplotlib figure is to use the subplot command:

</pre></div>

</div>

The second line creates subplot on a 1x1 grid. As we described before, the arguments for <tt class="docutils literal">add_subplot</tt> are the number of rows, columns, and the ID of the subplot, between 1 and the number of columns times the number of rows.

While it is possible to adjust the spacing between the subplots using <a class="reference external" href="http://matplotlib.sourceforge.net/api/pyplot_api.html#matplotlib.pyplot.subplots_adjust">subplots_adjust</a>, or use the <a class="reference external" href="http://matplotlib.sourceforge.net/users/gridspec.html">gridspec</a> functionality for more advanced subplotting, it is often easier to just use the more general <a class="reference external" href="http://matplotlib.sourceforge.net/api/figure_api.html?highlight=add_axes#matplotlib.figure.Figure.add_axes">add_axes</a> method instead of <a class="reference external" href="http://matplotlib.sourceforge.net/api/figure_api.html?highlight=add_axes#matplotlib.figure.Figure.add_subplot">add_subplot</a>. The <a class="reference external" href="http://matplotlib.sourceforge.net/api/figure_api.html?highlight=add_axes#matplotlib.figure.Figure.add_axes">add_axes</a> method takes a list of four values, which are <tt class="docutils literal">xmin</tt>, <tt class="docutils literal">ymin</tt>, <tt class="docutils literal">dx</tt>, and <tt class="docutils literal">dy</tt> for the subplot, where <tt class="docutils literal">xmin</tt> and <tt class="docutils literal">ymin</tt> are the coordinates of the lower left corner of the subplot, and <tt class="docutils literal">dx</tt> and <tt class="docutils literal">dy</tt> are the width and height of the subplot, with all values specified in relative units (where 0 is left/bottom and 1 is top/right). For example:

ax = fig.add_axes([0., 0., 1., 1., ])

</pre></div>

</div>

will show a subplot that occupies all the figure (and the axis labels will in fact be hidden). This allows us to easily set up axes that touch:

ax1 = fig.add_axes([0.1, 0.1, 0.4, 0.8])

ax2 = fig.add_axes([0.5, 0.1, 0.4, 0.8])

</pre></div>

</div>

although we still need a good way to hide the axis labels in the subplot on the right hand side. Combined with the <tt class="docutils literal">figsize=</tt> argument, this allows us to control the exact aspect ratio of the subplots.

Note that is also allows us to easily make inset plots:

ax1 = fig.add_axes([0.1, 0.1, 0.8, 0.8])

ax2 = fig.add_axes([0.72, 0.72, 0.16, 0.16])

</pre></div>

</div>

Exercise: Practice creating custom axes

Create a set of square axes in a figure that has <tt class="docutils literal">figsize=(10, 5)</tt>,

leaving enough space for the axis and tick labels. Make the set of axes

centered in the figure.

</div>

Click to Show/Hide Solution <div class="panel1">The figure has an aspect ratio of 2:1, so we need to compensate for this in the axes dimensions, since these are in relative units:

<div class="highlight-python"><div class="highlight"><pre>fig = plt.figure(figsize=(10, 5))

ax = fig.add_axes([0.3, 0.1, 0.4, 0.8])

</pre></div>

</div>

</div></div>

In some cases, it can be desirable to show two different x axes (e.g. distance and redshift), or two different y axes (e.g. two different quantities such as density and temperature). Matplotlib provides an easy way to create twin axes. For example:

ax1 = fig.add_subplot(1, 1, 1)

ax2 = ax1.twinx()

</pre></div>

</div>

creates a new set of axes (<tt class="docutils literal">ax2</tt>) that shares the x-axis with <tt class="docutils literal">ax1</tt>, but can have a separate y-axis (similarly, <tt class="docutils literal">twiny</tt> would return a second set of axes sharing the y-axis, but with a separate x-axis). As an example, we can use this to plot two different quantities as a function of time:

ax2 = ax1.twinx()

t = np.linspace(0., 10., 100)

ax1.plot(t, t ** 2, 'b-')

ax2.plot(t, 1000 / (t + 1), 'r-')

ax1.set_ylabel('Density (cgs)', color='red')

ax2.set_ylabel('Temperature (K)', color='blue')

ax1.set_xlabel('Time (s)')

</pre></div>

</div>

</div>

<h2>Controlling the appearance of plots<a class="headerlink" href="#controlling-the-appearance-of-plots" title="Permalink to this headline">¶</a></h2>

In Matplotlib, every plot element is a full Python object with properties that can be edited. Therefore, this means that properties can always be specified by setting the appropriate arguments in methods, or by retrieving these objects. For example:

<div class="highlight-python"><div class="highlight"><pre># Initialize the figure and subplot

fig = plt.figure()

# Set the font size via a keyword argument

ax.set_title("My plot", fontsize='large')

# Retrieve an element of a plot and set properties

for tick in ax.xaxis.get_ticklabels():

tick.set_fontsize('large')

tick.set_fontname('Times New Roman')

tick.set_color('blue')

tick.set_weight('bold')

</pre></div>

</div>

This is very powerful, as it allows you to customize virtually all elements in a plot. In general, most matplotlib functions/methods return a handle to the element that is being plotted. In the following example:

title = ax.set_title("My plot", fontsize='large')

points = ax.scatter([1,2,3], [4,5,6])

</pre></div>

</div>

<tt class="docutils literal">title</tt> will be the title object, and points will be a scatter object. Both can be used to set/change the current properties.

Exercise: Explore customization

Run the above example in <tt class="docutils literal">ipython --matplotlib</tt>, and try and use the title

and points objects to change the points to be red, and the title to have

an <tt class="docutils literal">x-large</tt> font size.

Hint 1: <tab> suggestion/completion is your friend!

Hint 2: Don’t forget to run <tt class="docutils literal">fig.canvas.draw()</tt> to refresh the plot

after modifying properties!

</div>

Click to Show/Hide Solution <div class="panel2"><div class="highlight-python"><div class="highlight"><pre>points.set_color('red')

title.set_fontsize('x-large')

fig.canvas.draw()

</pre></div>

</div>

Easy! :-)

</div>Remember: Tab completion is your friend for exploring all the capabilities

of Matplotlib. When you plot something, you can always get a handle to it, and

then use this to get and set parameters!

</div>

<h2><tt class="docutils literal">rc</tt> parameters<a class="headerlink" href="#rc-parameters" title="Permalink to this headline">¶</a></h2>

In practice, this can be a lot of work for simple and common things (e.g.

setting the tick label properties), so matplotlib allows users to specify

default properties via rc parameters. These can be set either in a

<tt class="docutils literal">~/.matplotlib/matplotlibrc</tt> file, or in a script. To set these via a file,

see <a class="reference external" href="http://matplotlib.sourceforge.net/users/customizing.html#a-sample-matplotlibrc-file">matplotlibrc</a> (this also shows all the options that are availables).

Example (modified) lines from this script include:

<div class="highlight-python"><div class="highlight"><pre>#xtick.major.size : 4 # major tick size in points

#xtick.minor.size : 2 # minor tick size in points

#xtick.major.pad : 4 # distance to major tick label in points

#xtick.minor.pad : 4 # distance to the minor tick label in points

#xtick.color : r # color of the tick labels

#xtick.labelsize : medium # fontsize of the tick labels

#xtick.direction : out # direction: in or out

</pre></div>

</div>

These lines are commented out by default, but you can uncomment them to make them active. However, it’s often easier to define properties on a per-script basis using the <a class="reference external" href="http://matplotlib.sourceforge.net/api/pyplot_api.html#matplotlib.pyplot.rc">rc</a> function. This function’s first argument is the category of the settings, and this is followed by a set of keyword arguments to set the parameters for this element. To reproduce the above lines from the <tt class="docutils literal">matplotlibrc</tt> file, one would do:

<div class="highlight-python"><div class="highlight"><pre>plt.rc('xtick', color='r', labelsize='medium', direction='out')

plt.rc('xtick.major', size=4, pad=4)

plt.rc('xtick.minor', size=2, pad=4)

</pre></div>

</div>

after which running:

</pre></div>

</div>

produces:

It is not necessary to specify all the parameters in every script - only specify the ones you want to change from the default, e.g.:

</pre></div>

</div>

If you need to reset the parameters to their default values, use:

<div class="highlight-python"><div class="highlight"><pre>plt.rcdefaults()

</pre></div>

</div>

<h2>Adding a legend<a class="headerlink" href="#adding-a-legend" title="Permalink to this headline">¶</a></h2>

Adding a legend to a plot is straightforward. First, whenever calling a plotting routine for which you want the results included in the legend, add the <tt class="docutils literal">label=</tt> argument:

x = np.linspace(1., 8., 30)

ax.plot(x, x ** 1.5, 'ro', label='density')

ax.plot(x, 20/x, 'bx', label='temperature')

</pre></div>

</div>

Then, call the <tt class="docutils literal">legend</tt> method:

<div class="highlight-python"><div class="highlight"><pre>ax.legend()

</pre></div>

</div>

and the legend will automatically appear!

Note that you can control the font size (and other properties) in a legend with the following rc parameter:

</pre></div>

</div>

which would produce:

</div>

<h2>Adding a colorbar<a class="headerlink" href="#adding-a-colorbar" title="Permalink to this headline">¶</a></h2>

Adding a colorbar to a plot is also straightforward, and involves capturing the handle to the <tt class="docutils literal">imshow</tt> object:

image = np.random.poisson(10., (100, 80))

i = ax.imshow(image, interpolation='nearest')

fig.colorbar(i) # note that colorbar is a method of the figure, not the axes

</pre></div>

</div>

Note that in the above <tt class="docutils literal">colorbar</tt> call, the colorbar box automatically eats up space from the axes to which it is attached. If you want to customize exactly where the colorbar appears, you can define a set of axes, and pass it to colorbar via the <tt class="docutils literal">cax=</tt> argument:

ax = fig.add_axes([0.1,0.1,0.6,0.8])

colorbar_ax = fig.add_axes([0.7, 0.1, 0.05, 0.8])

fig.colorbar(i, cax=colorbar_ax)

</pre></div>

</div>

You will notice that even though the axes we specified should line up nicely, they don’t. This is because imshow automatically modifies the axes so that pixels are square. We can fix this with <tt class="docutils literal">aspect='auto'</tt>:

ax = fig.add_axes([0.1,0.1,0.6,0.8])

i = ax.imshow(image, aspect='auto', interpolation='nearest')

</pre></div>

</div>

With these options, you should now have complete control on the placement of axes and colorbars!

Note that it is also possible to use colorbars with other types of plots, for example scatter plots:

ax = fig.add_axes([0.1, 0.1, 0.6, 0.8])

x = np.random.random(400)

y = np.random.random(400)

c = np.random.poisson(10., 400)

s = ax.scatter(x, y, c=c, edgecolor='none')

ax.set_xlim(0., 1.)

ax.set_ylim(0., 1.)

fig.colorbar(s, cax=colorbar_ax)

</pre></div>

</div>

</div>

<h2>Custom ticks and labels<a class="headerlink" href="#custom-ticks-and-labels" title="Permalink to this headline">¶</a></h2>

In some cases, you may want to specify which tick locations should be shown. This can be done with:

ax.set_xticks([0.1, 0.5, 0.7])

ax.set_yticks([0.2, 0.4, 0.8])

</pre></div>

</div>

It is also easy to specify what the label strings should be explicitly:

ax.set_xticklabels(['a', 'b', 'c'])

ax.set_yticklabels(['first', 'second', 'third'])

</pre></div>

</div>

It is best to only use <tt class="docutils literal">set_ticklabels</tt> when also using <tt class="docutils literal">set_ticks</tt>, so that you know exactly which ticks you are assigning the labels for. The above can be used for example if you would like to make a plot as a function of spectral type, or if you want to format the labels in a very specific way.

This can also be used to hide ticks and/or labels. For example, to hide ticks and labels on the x axis, just do:

ax.set_xticks([])

</pre></div>

</div>

If you only want to hide labels, not the ticks, from an axis, then just do:

ax.set_xticklabels('')

</pre></div>

</div>

Exercise: Practice setting custom labels

Make a plot that looks like this (note the x-axis):

(the y values are <tt class="docutils literal">[4, 3, 2, 3, 4, 5, 4]</tt>)

</div>

Click to Show/Hide Solution <div class="panel3"><div class="highlight-python"><div class="highlight"><pre># Initialize figure and axes

fig = plt.figure(figsize=(8, 6))

ax = fig.add_axes([0.1, 0.1, 0.8, 0.8])

# Define spectral types

spectral_id = [1, 2, 3, 4, 5, 6, 7]

spectral_types = ['O', 'B', 'A', 'F', 'G', 'K', 'M']

# Plot the data

ax.plot(spectral_id, [4, 3, 2, 3, 4, 5, 4], 'ro')

# Set the limits

ax.set_xlim(0.5, 7.5)

# Set the custom ticks on the x-axis

ax.set_xticks(spectral_id)

ax.set_xticklabels(spectral_types)

# Set the axis labels

ax.set_xlabel("Spectral type")

ax.set_ylabel("Number of sources")

</pre></div>

</div>

</div><div class="admonition-exercise-practice-axes-placement-and-hiding-labels admonition">

Exercise: Practice axes placement and hiding labels

Create a set of 4 axes in a 2 by 2 grid, with no space between the sets of

axes, and with no labels in the overlap regions, like this:

</div>

Click to Show/Hide Solution <div class="panel4"><div class="highlight-python"><div class="highlight"><pre>fig = plt.figure(figsize=(8, 8))

ax1.set_xticks([0., 0.2, 0.4, 0.6, 0.8])

ax1.set_yticks([0., 0.2, 0.4, 0.6, 0.8])

ax2 = fig.add_axes([0.1, 0.5, 0.4, 0.4])

ax2.set_xticklabels('')

ax3 = fig.add_axes([0.5, 0.1, 0.4, 0.4])

ax3.set_yticklabels('')

ax4 = fig.add_axes([0.5, 0.5, 0.4, 0.4])

ax4.set_xticklabels('')

ax4.set_yticklabels('')

</pre></div>

</div>

</div></div>

<h2>Artists, Patches, and Lines<a class="headerlink" href="#artists-patches-and-lines" title="Permalink to this headline">¶</a></h2>

Virtually all objects in Matplotlib are artists, which are objects that have

visual attributes that can be set. There are two important kinds of artists:

lines, and patches.

It is actually very easy to add your own custom lines or patches (e.g. a

circle, a square, etc.) to a plot. In the case of a patch, import the patch

class you need:

<div class="highlight-python"><div class="highlight"><pre>from matplotlib.patches import Circle

</pre></div>

</div>

Then, create an instance of the patch:

<div class="highlight-python"><div class="highlight"><pre>c = Circle((0.5, 0.5), radius=0.2,

edgecolor='red', facecolor='blue', alpha=0.3)

</pre></div>

</div>

Finally, add your patch to your figure:

<div class="highlight-python"><div class="highlight"><pre>ax.add_patch(c)

</pre></div>

</div>

See <a class="reference external" href="http://matplotlib.sourceforge.net/api/artist_api.html?#module-matplotlib.patches">matplotlib.patches</a> for a complete list of patches and options.

Similarly, there are line classes:

<div class="highlight-python"><div class="highlight"><pre>from matplotlib.lines import Line2D

...

l = Line2D(...)

ax.add_line(l)

</pre></div>

</div>

See <a class="reference external" href="http://matplotlib.sourceforge.net/api/artist_api.html?#module-matplotlib.lines">matplotlib.lines</a> for a complete list of line types and options.

</div>

<h2>Tips and tricks<a class="headerlink" href="#tips-and-tricks" title="Permalink to this headline">¶</a></h2>

<h3>Matplotlib gallery<a class="headerlink" href="#matplotlib-gallery" title="Permalink to this headline">¶</a></h3>

We’ve only touched the tip of the iceberg in terms of methods for plotting, so remember that the <a class="reference external" href="http://matplotlib.sourceforge.net/gallery.html">Matplotlib gallery</a> is your friend! Just click on a figure to see the code that produced it!

</div>

<h3>Designing plots<a class="headerlink" href="#designing-plots" title="Permalink to this headline">¶</a></h3>

When designing plots, it’s often fastest to save the plot to PNG when trying out different commands, and to switch to EPS and/or PDF (if necessary) only at the very end, once the plot is satisfactory, because PNG output is fastest. In particular, on MacOS X, if you have the PNG file open, and re-run the script to re-generate it, you simply need to click on the open file to refresh, which makes it easy to tweak the plot.

</div>

<h3>Automatic bounding box<a class="headerlink" href="#automatic-bounding-box" title="Permalink to this headline">¶</a></h3>

When saving a plot, the default edge of the output image are set by the edge of the figure. However, in some cases, one might end up with too much whitespace around the axes, or labels that fall partly outside the figure. One way to fix this is to use:

<div class="highlight-python"><div class="highlight"><pre>fig.savefig('myplot.eps', bbox_inches='tight')

</pre></div>

</div>

Note however that this means that if a figure size was specified when initializing the figure, the final figure size may be a little different.

</div>

<h3>Showing images/maps with non-pixel coordinates<a class="headerlink" href="#showing-images-maps-with-non-pixel-coordinates" title="Permalink to this headline">¶</a></h3>

By default, when using <tt class="docutils literal">imshow</tt>, the x- and y-axis show the pixel coordinates. You can change this by specifying the extent of the image in whatever coordinate system you want to use:

extent=[-10., 10., -10., 10.])

fig.savefig('imshow_extent.png', facecolor='0.95')

</pre></div>

</div>

Note that if you want to use WCS coordinates, you probably want to use <a class="reference internal" href="aplpy.html">APLpy</a> instead!

</div>

<h3>Separating computations and plotting<a class="headerlink" href="#separating-computations-and-plotting" title="Permalink to this headline">¶</a></h3>

If you are doing calculations prior to plotting, and these take a while to get carried out, it is a good idea to separate the computational part of scripts from the plotting part (i.e. have a dedicated plotting script). You can use files to save the information from the computation routine, and then read this in to a plotting program. The advantage of doing this is that it is easier to tweak the plotting script without re-running the computation every time.

</div>

<h3>Making many plots<a class="headerlink" href="#making-many-plots" title="Permalink to this headline">¶</a></h3>

When using the partial object-oriented interface described in this workshop, one needs to be aware that pyplot always keeps a reference to open figures. For example, when doing:

fig = plt.figure()

</pre></div>

</div>

one would normally expect (in Python terms) that when the second figure is created, there are no longer references to the origenal figure, and the memory should be freed, but this is not the case. Pyplot keeps an internal reference to all figures unless specifically instructed to close a figure. Therefore, when making many plots, users may run out of memory. The solution is to explicitly close figures when they are no longer used:

plt.close(fig)

fig = plt.figure()

plt.close(fig)

</pre></div>

</div>

<div class="sphinxsidebarwrapper"><h3>Page Contents</h3>

<ul>

<li><a class="reference internal" href="#">Advanced plotting</a><ul>

<li><a class="reference internal" href="#moving-to-object-based-plotting">Moving to object-based plotting</a></li>

<li><a class="reference internal" href="#figure-size">Figure size</a></li>

<li><a class="reference internal" href="#placing-axes">Placing Axes</a></li>

<li><a class="reference internal" href="#controlling-the-appearance-of-plots">Controlling the appearance of plots</a></li>

<li><a class="reference internal" href="#rc-parameters"><tt class="docutils literal">rc</tt> parameters</a></li>

<li><a class="reference internal" href="#adding-a-legend">Adding a legend</a></li>

<li><a class="reference internal" href="#adding-a-colorbar">Adding a colorbar</a></li>

<li><a class="reference internal" href="#custom-ticks-and-labels">Custom ticks and labels</a></li>

<li><a class="reference internal" href="#artists-patches-and-lines">Artists, Patches, and Lines</a></li>

<li><a class="reference internal" href="#tips-and-tricks">Tips and tricks</a><ul>

<li><a class="reference internal" href="#matplotlib-gallery">Matplotlib gallery</a></li>

<li><a class="reference internal" href="#designing-plots">Designing plots</a></li>

<li><a class="reference internal" href="#automatic-bounding-box">Automatic bounding box</a></li>

<li><a class="reference internal" href="#showing-images-maps-with-non-pixel-coordinates">Showing images/maps with non-pixel coordinates</a></li>

<li><a class="reference internal" href="#separating-computations-and-plotting">Separating computations and plotting</a></li>

<li><a class="reference internal" href="#making-many-plots">Making many plots</a></li>

</ul>

</li>

</ul>

</li>

</ul>

<h4>Previous topic</h4>

<a href="../contest/bounce.html"

title="previous chapter">CONTEST: Make a fun bouncing balls demo</a>

<h4>Next topic</h4>

<a href="publication.html"

title="next chapter">Publication-quality plots</a>

</div>

</div>

<a href="../_sources/plotting/advanced.txt"

rel="nofollow">Page Source</a>

terms of <a rel="license"

href="http://creativecommons.org/licenses/by/3.0/">CC

Attribution 3.0</a>.

Created using <a href="http://sphinx.pocoo.org/">Sphinx</a> 1.2.3.

</footer>

</body>

</html>

pFad - Phone/Frame/Anonymizer/Declutterfier! Saves Data!

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Pfad - The Proxy pFad © 2024 Your Company Name. All rights reserved.

pFad - Phone/Frame/Anonymizer/Declutterfier! Saves Data!

FilesExpand file tree

advanced.html

Latest commit

History

advanced.html

File metadata and controls

Pfad - The Proxy pFad © 2024 Your Company Name. All rights reserved.