Hide keyboard shortcuts

Hot-keys on this page

r m x p   toggle line displays

j k   next/prev highlighted chunk

0   (zero) top of page

1   (one) first highlighted chunk

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

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

""" 

Visualizing the Output of LDA Models 

************************************ 

 

Functions and classes of this module are for visualizing LDA models. This is, \ 

except one function (:func:`plot_wordcloud`), based on the document-topics \ 

distribution DataFrame. 

 

Contents 

******** 

* :func:`plot_wordcloud` plots the top ``n`` words for a specific topic. \ 

The higher their weight, the bigger the word. 

* :clas:`PlotDocumentTopics` is basically the core of this module. Construct \ 

this class, if you want to plot the content of the document-topics DataFrame. 

* :meth:`static_heatmap` plots a static, :module:`matplotlib`-based heatmap of \ 

the document-topics distribution. 

* :meth:`interactive_heatmap` plots an interactive, :module:`bokeh`-based \ 

heatmap of the document-topics distribution. 

* :meth:`static_barchart_per_topic` plots a static, :module:`matplotlib`-based \ 

barchart of the document proportions for a specific topic. 

* :meth:`interactive_barchart_per_topic` plots an interactive, :module:`bokeh`-based \ 

barchart of the document proportions for a specific topic. 

* :meth:`static_barchart_per_document` plots a static, :module:`matplotlib`-based \ 

barchart of the topic proportions for a specific document. 

* :meth:`interactive_barchart_per_document` plots an interactive, :module:`bokeh`-based \ 

barchart of the topic proportions for a specific document. 

* :meth:`topic_over_time` plots a static, :module:`matplotlib`-based \ 

line diagram of the development of a topic over time, based on metadata. 

* :meth:`to_file` saves either a :module:`matplotlib` or a :module:`bokeh` figure \ 

object to disk. 

 

""" 

 

 

import logging 

from dariah_topics import postprocessing 

import matplotlib 

matplotlib.use('Agg') 

import matplotlib.pyplot as plt 

from matplotlib import cm 

import numpy as np 

import os 

import pandas as pd 

from bokeh.plotting import figure 

from bokeh import palettes 

from bokeh.models import ( 

ColumnDataSource, 

HoverTool, 

LinearColorMapper, 

BasicTicker, 

ColorBar 

) 

 

from collections import Counter 

 

 

log = logging.getLogger('dariah_topics') 

 

 

class PlotDocumentTopics: 

""" 

Class to visualize document-topic matrix. 

""" 

def __init__(self, document_topics): 

self.document_topics = document_topics 

 

 

def static_heatmap(self, figsize=(1000 / 96, 600 / 96), dpi=None, 

labels_fontsize=13, cmap='Blues', ticks_fontsize=12, 

xlabel='Document', ylabel='Topic', xticks_bottom=0.1, 

xticks_rotation=50, xticks_ha='right', colorbar=False): 

"""Plots a static heatmap. 

 

Args: 

figsize (tuple), optional: Size of the figure in inches. Defaults to 

``(1000 / 96, 500 / 96)``. 

dpi (int), optional: Dots per inch. Defaults to None. 

labels_fontsize (int), optional: Fontsize of the figure labels. Defaults 

to 13. 

cmap (str), optional: Colormap for the figure. Defaults to ``Blues``. 

ticks_fontsize (int), optional: Fontsize of axis ticks. Defaults to 12. 

xlabel (str), optional: Label of x-axis. Defaults to ``Document``. 

ylabel (str), optional: Label of y-axis. Defaults to ``Topic``. 

xticks_bottom (str), optional: Distance to bottom of x-ticks. Defaults 

to 0.1. 

xticks_rotation (int), optional: Rotation degree of x-ticks. Defaults 

to 50. 

xticks_ha (str), optional: The horizontal alignment of the x-tick labels. 

Defaulst to ``right``. 

colorbar (bool), optional: If True, include colorbar. Defaults to True. 

 

Returns: 

Figure object. 

""" 

fig, ax = plt.subplots(figsize=figsize, dpi=dpi) 

heatmap = ax.pcolor(self.document_topics, cmap=cmap) 

ax.set_xlabel(xlabel, fontsize=labels_fontsize) 

ax.set_ylabel(ylabel, fontsize=labels_fontsize) 

ax.set_xticks(np.arange(self.document_topics.shape[1]) + 0.5) 

ax.set_yticks(np.arange(self.document_topics.shape[0]) + 0.5) 

ax.set_xticklabels(list(self.document_topics.columns), fontsize=ticks_fontsize) 

ax.set_yticklabels(list(self.document_topics.index), fontsize=ticks_fontsize) 

fig.autofmt_xdate(bottom=xticks_bottom, rotation=xticks_rotation, ha=xticks_ha) 

if colorbar: 

cax = ax.imshow(self.document_topics, interpolation='nearest', cmap=cmap) 

cbar = fig.colorbar(cax, ticks=np.arange(0, 1, 0.1)) 

return fig 

 

def __static_barchart(self, index, describer, figsize=(11, 7), color='#053967', 

edgecolor=None, linewidth=None, alpha=None, labels_fontsize=15, 

ticks_fontsize=14, title=True, title_fontsize=17, 

dpi=None, transpose_data=False): 

"""Plots a static barchart. 

 

Args: 

index Union(int, str): Index of document-topics matrix column or 

name of column. 

describer (str): Describer of what the plot shows, e.g. either document 

or topic. 

title (bool), optional: If True, figure will have a title in the format 

``describer: index``. 

title_fontsize (int), optional: Fontsize of figure title. 

transpose_data (bool): If True. document-topics matrix will be transposed. 

Defaults to False. 

color (str), optional: Color of the bins. Defaults to ``#053967``. 

edgecolor (str), optional: Color of the bin edges. Defaults to None. 

lindewidth (float), optional: Width of bin lines. Defaults to None. 

alpha (float): Alpha value used for blending. Defaults to None. 

figsize (tuple), optional: Size of the figure in inches. Defaults to 

``(1000 / 96, 500 / 96)``. 

dpi (int), optional: Dots per inch. Defaults to None. 

labels_fontsize (int), optional: Fontsize of the figure labels. Defaults 

to 15. 

ticks_fontsize (int), optional: Fontsize of axis ticks. Defaults to 14. 

 

Returns: 

Figure object. 

""" 

fig, ax = plt.subplots(figsize=figsize, dpi=dpi) 

if isinstance(index, int): 

if transpose_data: 

proportions = self.document_topics.T.iloc[index] 

else: 

proportions = self.document_topics.iloc[index] 

if title: 

plot_title = '{0}: {1}'.format(describer, proportions.name) 

ax.set_title(plot_title, fontsize=title_fontsize) 

elif isinstance(index, str): 

if transpose_data: 

proportions = self.document_topics.T.loc[index] 

else: 

proportions = self.document_topics.loc[index] 

if title: 

plot_title = '{}: {}'.format(describer, index) 

ax.set_title(plot_title, fontsize=title_fontsize) 

else: 

raise ValueError("{} must be int or str.".format(index)) 

 

y_axis = np.arange(len(proportions)) 

x_axis = proportions 

y_ticks_labels = proportions.index 

ax.barh(y_axis, x_axis, color=color, edgecolor=edgecolor, linewidth=linewidth, alpha=alpha) 

ax.set_xlabel('Proportion', fontsize=labels_fontsize) 

ax.set_ylabel(describer, fontsize=labels_fontsize) 

ax.set_yticks(y_axis) 

ax.set_yticklabels(y_ticks_labels, fontsize=ticks_fontsize) 

ax.tick_params(axis='x', labelsize=ticks_fontsize) 

return fig 

 

def static_barchart_per_topic(self, **kwargs): 

"""Plots a static barchart per topic. 

 

Args: 

index Union(int, str): Index of document-topics matrix column or 

name of column. 

describer (str): Describer of what the plot shows, e.g. either document 

or topic. 

title (bool), optional: If True, figure will have a title in the format 

``describer: index``. 

title_fontsize (int), optional: Fontsize of figure title. 

transpose_data (bool): If True. document-topics matrix will be transposed. 

Defaults to False. 

color (str), optional: Color of the bins. Defaults to ``#053967``. 

edgecolor (str), optional: Color of the bin edges. Defaults to None. 

lindewidth (float), optional: Width of bin lines. Defaults to None. 

alpha (float): Alpha value used for blending. Defaults to None. 

figsize (tuple), optional: Size of the figure in inches. Defaults to 

``(1000 / 96, 500 / 96)``. 

dpi (int), optional: Dots per inch. Defaults to None. 

labels_fontsize (int), optional: Fontsize of the figure labels. Defaults 

to 15. 

ticks_fontsize (int), optional: Fontsize of axis ticks. Defaults to 14. 

 

Returns: 

Figure object. 

""" 

return self.__static_barchart(**kwargs) 

 

def static_barchart_per_document(self, **kwargs): 

"""Plots a static barchart per document. 

 

Args: 

index Union(int, str): Index of document-topics matrix column or 

name of column. 

describer (str): Describer of what the plot shows, e.g. either document 

or topic. 

title (bool), optional: If True, figure will have a title in the format 

``describer: index``. 

title_fontsize (int), optional: Fontsize of figure title. 

transpose_data (bool): If True. document-topics matrix will be transposed. 

Defaults to False. 

color (str), optional: Color of the bins. Defaults to ``#053967``. 

edgecolor (str), optional: Color of the bin edges. Defaults to None. 

lindewidth (float), optional: Width of bin lines. Defaults to None. 

alpha (float): Alpha value used for blending. Defaults to None. 

figsize (tuple), optional: Size of the figure in inches. Defaults to 

``(1000 / 96, 500 / 96)``. 

dpi (int), optional: Dots per inch. Defaults to None. 

labels_fontsize (int), optional: Fontsize of the figure labels. Defaults 

to 15. 

ticks_fontsize (int), optional: Fontsize of axis ticks. Defaults to 14. 

 

Returns: 

Figure object. 

""" 

return self.__static_barchart(transpose_data=True, **kwargs) 

 

def interactive_heatmap(self, palette=palettes.Blues[9], reverse_palette=True, 

tools='hover, pan, reset, save, wheel_zoom, zoom_in, zoom_out', 

width=1000, height=550, x_axis_location='below', toolbar_location='above', 

sizing_mode='fixed', line_color=None, grid_line_color=None, axis_line_color=None, 

major_tick_line_color=None, major_label_text_font_size='9pt', 

major_label_standoff=0, major_label_orientation=3.14/3, colorbar=True): 

"""Plots an interactive heatmap. 

 

Args: 

palette (list), optional: A list of color values. Defaults to ``palettes.Blues[9]``. 

reverse_palette (bool), optional: If True, color values of ``palette`` will 

be reversed. Defaults to True. 

tools (str), optional: Tools, which will be includeded. Defaults to ``hover, 

pan, reset, save, wheel_zoom, zoom_in, zoom_out``. 

width (int), optional: Width of the figure. Defaults to 1000. 

height (int), optional: Height of the figure. Defaults to 550. 

x_axis_location (str), optional: Location of the x-axis. Defaults to 

``below``. 

toolbar_location (str), optional: Location of the toolbar. Defaults to 

``above``. 

sizing_mode (str), optional: Size fixed or width oriented. Defaults to ``fixed``. 

line_color (str): Color for lines. Defaults to None. 

grid_line_color (str): Color for grid lines. Defaults to None. 

axis_line_color (str): Color for axis lines. Defaults to None. 

major_tick_line_color (str): Color for major tick lines. Defaults to None. 

major_label_text_font_size (str): Font size for major label text. Defaults 

to ``9pt``. 

major_label_standoff (int): Standoff for major labels. Defaults to 0. 

major_label_orientation (float): Orientation for major labels. Defaults 

to ``3.14 / 3``. 

colorbar (bool): If True, colorbar will be included. 

 

Returns: 

Figure object. 

""" 

if reverse_palette: 

palette = list(reversed(palette)) 

 

x_range = list(self.document_topics.columns) 

y_range = list(self.document_topics.index) 

 

stacked_data = pd.DataFrame(self.document_topics.stack()).reset_index() 

stacked_data.columns = ['Topics', 'Documents', 'Distributions'] 

mapper = LinearColorMapper(palette=palette, 

low=stacked_data.Distributions.min(), 

high=stacked_data.Distributions.max()) 

source = ColumnDataSource(stacked_data) 

 

fig = figure(x_range=x_range, 

y_range=y_range, 

x_axis_location=x_axis_location, 

plot_width=width, plot_height=height, 

tools=tools, toolbar_location=toolbar_location, 

sizing_mode=sizing_mode, 

logo=None) 

fig.rect(x='Documents', y='Topics', source=source, width=1, height=1, 

fill_color={'field': 'Distributions', 'transform': mapper}, 

line_color=line_color) 

 

fig.grid.grid_line_color = grid_line_color 

fig.axis.axis_line_color = axis_line_color 

fig.axis.major_tick_line_color = major_tick_line_color 

fig.axis.major_label_text_font_size = major_label_text_font_size 

fig.axis.major_label_standoff = major_label_standoff 

fig.xaxis.major_label_orientation = major_label_orientation 

 

if 'hover' in tools: 

fig.select_one(HoverTool).tooltips = [('x-Axis', '@Documents'), 

('y-Axis', '@Topics'), 

('Score', '@Distributions')] 

 

if colorbar: 

feature = ColorBar(color_mapper=mapper, major_label_text_font_size=major_label_text_font_size, 

ticker=BasicTicker(desired_num_ticks=len(palette)), 

label_standoff=6, border_line_color=None, location=(0, 0)) 

fig.add_layout(feature, 'right') 

return fig 

 

 

def __interactive_barchart(self, index, describer, tools='hover, pan, reset, save, wheel_zoom, zoom_in, zoom_out', 

width=1000, height=400, toolbar_location='above', 

sizing_mode='fixed', line_color=None, grid_line_color=None, axis_line_color=None, 

major_tick_line_color=None, major_label_text_font_size='9pt', 

major_label_standoff=0, title=True, bin_height=0.5, 

transpose_data=False, bar_color='#053967'): 

"""Plots an interactive barchart. 

 

Args: 

index Union(int, str): Index of document-topics matrix column or 

name of column. 

describer (str): Describer of what the plot shows, e.g. either document 

or topic. 

bar_color (str), optional: Color of bars. Defaults to ``#053967``. 

transpose_data (bool): If True. document-topics matrix will be transposed. 

Defaults to False. 

title (bool), optional: If True, figure will have a title in the format 

``describer: index``. 

tools (str), optional: Tools, which will be includeded. Defaults to ``hover, 

pan, reset, save, wheel_zoom, zoom_in, zoom_out``. 

width (int), optional: Width of the figure. Defaults to 1000. 

height (int), optional: Height of the figure. Defaults to 400. 

x_axis_location (str), optional: Location of the x-axis. Defaults to 

``below``. 

toolbar_location (str), optional: Location of the toolbar. Defaults to 

``above``. 

sizing_mode (str), optional: Size fixed or width oriented. Defaults to ``fixed``. 

line_color (str): Color for lines. Defaults to None. 

grid_line_color (str): Color for grid lines. Defaults to None. 

axis_line_color (str): Color for axis lines. Defaults to None. 

major_tick_line_color (str): Color for major tick lines. Defaults to None. 

major_label_text_font_size (str): Font size for major label text. Defaults 

to ``9pt``. 

major_label_standoff (int): Standoff for major labels. Defaults to 0. 

 

Returns: 

Figure object. 

""" 

if isinstance(index, int): 

if transpose_data: 

proportions = self.document_topics.T.iloc[index] 

else: 

proportions = self.document_topics.iloc[index] 

if title: 

plot_title = '{}: {}'.format(describer, proportions.name) 

elif isinstance(index, str): 

if transpose_data: 

proportions = self.document_topics.T.loc[index] 

else: 

proportions = self.document_topics.loc[index] 

if title: 

plot_title = '{}: {}'.format(describer, index) 

else: 

raise ValueError("{} must be int or str.".format(index)) 

 

x_axis = proportions 

y_range = list(proportions.index) 

 

source = ColumnDataSource(dict(Describer=y_range, Proportion=x_axis)) 

 

fig = figure(y_range=y_range, title=plot_title, plot_width=width, plot_height=height, 

tools=tools, toolbar_location=toolbar_location, 

sizing_mode=sizing_mode, logo=None) 

fig.hbar(y='Describer', right='Proportion', height=bin_height, source=source, 

line_color=line_color, color=bar_color) 

 

fig.xgrid.grid_line_color = None 

fig.x_range.start = 0 

fig.grid.grid_line_color = grid_line_color 

fig.axis.axis_line_color = axis_line_color 

fig.axis.major_tick_line_color = major_tick_line_color 

fig.axis.major_label_text_font_size = major_label_text_font_size 

fig.axis.major_label_standoff = major_label_standoff 

 

if 'hover' in tools: 

fig.select_one(HoverTool).tooltips = [('Proportion', '@Proportion')] 

return fig 

 

def interactive_barchart_per_topic(self, **kwargs): 

"""Plots an interactive barchart per topic. 

 

Args: 

index Union(int, str): Index of document-topics matrix column or 

name of column. 

describer (str): Describer of what the plot shows, e.g. either document 

or topic. 

bar_color (str), optional: Color of bars. Defaults to ``#053967``. 

transpose_data (bool): If True. document-topics matrix will be transposed. 

Defaults to False. 

title (bool), optional: If True, figure will have a title in the format 

``describer: index``. 

tools (str), optional: Tools, which will be includeded. Defaults to ``hover, 

pan, reset, save, wheel_zoom, zoom_in, zoom_out``. 

width (int), optional: Width of the figure. Defaults to 1000. 

height (int), optional: Height of the figure. Defaults to 400. 

x_axis_location (str), optional: Location of the x-axis. Defaults to 

``below``. 

toolbar_location (str), optional: Location of the toolbar. Defaults to 

``above``. 

sizing_mode (str), optional: Size fixed or width oriented. Defaults to ``fixed``. 

line_color (str): Color for lines. Defaults to None. 

grid_line_color (str): Color for grid lines. Defaults to None. 

axis_line_color (str): Color for axis lines. Defaults to None. 

major_tick_line_color (str): Color for major tick lines. Defaults to None. 

major_label_text_font_size (str): Font size for major label text. Defaults 

to ``9pt``. 

major_label_standoff (int): Standoff for major labels. Defaults to 0. 

 

Returns: 

Figure object. 

""" 

return self.__interactive_barchart(**kwargs) 

 

def interactive_barchart_per_document(self, **kwargs): 

"""Plots an interactive barchart per document. 

 

Args: 

index Union(int, str): Index of document-topics matrix column or 

name of column. 

describer (str): Describer of what the plot shows, e.g. either document 

or topic. 

bar_color (str), optional: Color of bars. Defaults to ``#053967``. 

transpose_data (bool): If True. document-topics matrix will be transposed. 

Defaults to False. 

title (bool), optional: If True, figure will have a title in the format 

``describer: index``. 

tools (str), optional: Tools, which will be includeded. Defaults to ``hover, 

pan, reset, save, wheel_zoom, zoom_in, zoom_out``. 

width (int), optional: Width of the figure. Defaults to 1000. 

height (int), optional: Height of the figure. Defaults to 400. 

x_axis_location (str), optional: Location of the x-axis. Defaults to 

``below``. 

toolbar_location (str), optional: Location of the toolbar. Defaults to 

``above``. 

sizing_mode (str), optional: Size fixed or width oriented. Defaults to ``fixed``. 

line_color (str): Color for lines. Defaults to None. 

grid_line_color (str): Color for grid lines. Defaults to None. 

axis_line_color (str): Color for axis lines. Defaults to None. 

major_tick_line_color (str): Color for major tick lines. Defaults to None. 

major_label_text_font_size (str): Font size for major label text. Defaults 

to ``9pt``. 

major_label_standoff (int): Standoff for major labels. Defaults to 0. 

 

Returns: 

Figure object. 

""" 

return self.__interactive_barchart(transpose_data=True, **kwargs) 

 

def topic_over_time(self, metadata_df, threshold=0.1, starttime=1841, endtime=1920): 

"""Creates a visualization that shows topics over time. 

 

Description: 

With this function you can plot topics over time using metadata stored in the documents name. 

Only works with mallet output. 

 

Args: 

metadata_df(pd.Dataframe()): metadata created by metadata_toolbox 

threshold(float): threshold set to define if a topic in a document is viable 

starttime(int): sets starting point for visualization 

endtime(int): sets ending point for visualization 

 

 

Returns:  

matplotlib plot 

 

Note: this function is created for a corpus with filenames that looks like: 

1866_ArticleName.txt 

 

ToDo: make it compatible with gensim output 

Doctest 

 

""" 

years = list(range(starttime, endtime)) 

 

for topiclabel in self.document_topics.index.values: 

topic_over_threshold_per_year = [] 

mask = self.document_topics.loc[topiclabel] > threshold 

df = self.document_topics.loc[topiclabel].loc[mask] 

cnt = Counter() 

for filtered_topiclabel in df.index.values: 

year = metadata_df.loc[filtered_topiclabel, 'year'] 

print(year) 

cnt[year] += 1 

for year in years: 

topic_over_threshold_per_year.append(cnt[str(year)]) 

plt.plot(years, topic_over_threshold_per_year, label=topiclabel) 

 

plt.xlabel('Year') 

plt.ylabel('count topics over threshold') 

plt.legend() 

# fig.set_size_inches(18.5, 10.5) 

# fig = plt.figure(figsize=(18, 16)) 

return plt.gcf().set_size_inches(18.5, 10.5) 

 

@staticmethod 

def to_file(fig, filename): 

"""Saves a figure object to file. 

 

Args: 

fig Union(bokeh.figure, matplotlib.figure): Figure produced by either 

bokeh or matplotlib. 

filename (str): Name of the file with an extension, e.g. ``plot.png``. 

 

Returns: 

None. 

""" 

import matplotlib 

import bokeh 

if isinstance(fig, bokeh.plotting.figure.Figure): 

ext = os.path.splitext(filename)[1] 

if ext == '.png': 

export_png(fig, filename) 

elif ext == '.svg': 

fig.output_backend = 'svg' 

export_svgs(fig, filename) 

elif ext == '.html': 

output_file(filename) 

elif isinstance(fig, matplotlib.figure.Figure): 

fig.savefig(filename) 

return None