forked from pielstroem/Topics
-
Notifications
You must be signed in to change notification settings - Fork 13
/
visualization.py
executable file
·709 lines (637 loc) · 34.8 KB
/
visualization.py
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
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
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Visualizing the Output of LDA Models
************************************
Functions and classes of this module are for visualizing LDA models.
Contents
********
*
"""
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
)
import regex
from collections import defaultdict
from wordcloud import WordCloud
log = logging.getLogger(__name__)
log.addHandler(logging.NullHandler())
logging.basicConfig(level = logging.ERROR,
format = '%(levelname)s %(name)s: %(message)s')
def plot_wordcloud(weights, enable_notebook=True, **kwargs):
"""Plots a wordcloud based on tokens and frequencies.
Args:
weights (dict): A dictionary (or :module:``pandas`` Series) with tokens
as keys and frequencies as values.
enable_notebook (bool), optional: If True, enables :module:``matplotlib``
to show its figures within a Jupyter notebook.
font_path (str), optional: Font path to the font that will be used (OTF or TTF).
Defaults to DroidSansMono path on a Linux machine. If you are on
another OS or don't have this font, you need to adjust this path.
width (int), optional: Width of the canvas. Defaults to 400.
height (int), optional: Height of the canvas. Defaults to 200.
prefer_horizontal (float): The ratio of times to try horizontal fitting
as opposed to vertical. If ``prefer_horizontal < 1``, the algorithm
will try rotating the word if it doesn't fit. (There is currently no
built-in way to get only vertical words. Defaults to 0.90.
mask (nd-array), optional: If not None, gives a binary mask on where to draw words.
If mask is not None, width and height will be ignored and the shape
of mask will be used instead. All white (#FF or #FFFFFF) entries
will be considerd 'masked out' while other entries will be free to
draw on. Defaults to None.
scale (float), optional: Scaling between computation and drawing. For large word-cloud
images, using scale instead of larger canvas size is significantly
faster, but might lead to a coarser fit for the words. Defaults to 1.
min_font_size (int), optional: Smallest font size to use. Will stop when there is
no more room in this size. Defaults to 4.
font_step (int), optional: Step size for the font. ``font_step > 1`` might speed
up computation but give a worse fit. Defaults to 1.
max_words (int), optional: The maximum number of words. Defaults to 200.
stopwords (set), optional: The words that will be eliminated. If None, the build-in
stopwords list will be used.
background_color (str), optional: Background color for the word cloud image.
Defaults to ``black``.
max_font_size (int), optional: Maximum font size for the largest word. If None,
height of the image is used.
mode (str), optional: Transparent background will be generated when mode is ``RGBA``
and background_color is None. Defaults to ``RGB``.
relative_scaling (float), optional: Importance of relative word frequencies for
font-size. With ``relative_scaling=0``, only word-ranks are considered.
With ``relative_scaling=1``, a word that is twice as frequent will
have twice the size. If you want to consider the word frequencies and
not only their rank, ``relative_scaling`` around .5 often looks good.
Defaults to 0.5.
color_func (callable), optional: Callable with parameters ``word``, ``font_size``,
``position``, ``orientation``, ``font_path``, ``random_state`` that
returns a PIL color for each word. Overwrites ``colormap``. See ``colormap``
for specifying a :module:``matplotlib`` colormap instead.
collocations (bool), optional: Whether to include collocations (bigrams) of two words.
Defaults to True.
colormap (str), optional: :module:``matplotlib`` colormap to randomly draw colors
from for each word. Ignored if ``color_func`` is specified. Defaults to
``viridis``.
normalize_plurals (bool), optional: Whether to remove trailing 's' from words. If
True and a word appears with and without a trailing 's', the one with
trailing 's' is removed and its counts are added to the version without
trailing 's' -- unless the word ends with 'ss'. Defaults to True.
Returns:
WordCloud object.
Example:
>>> weights = {'an': 2, 'example': 1}
>>> plot_wordcloud(weights, enable_notebook=False) # doctest: +ELLIPSIS
<wordcloud.wordcloud.WordCloud object at ...>
"""
wordcloud = WordCloud(**kwargs).fit_words(weights)
if enable_notebook:
from IPython import get_ipython
get_ipython().run_line_magic('matplotlib', 'inline')
try:
fig, ax = plt.subplots(figsize=(kwargs['width'] / 96, kwargs['height'] / 96))
except KeyError:
fig, ax = plt.subplots(figsize=(400 / 96, 200 / 96))
ax.axis('off')
ax.imshow(wordcloud)
return wordcloud
def plot_key_frequencies(keys=None, overall_freqs=None, within_topic_freqs=None,
within_topic_color='#FF1727', document_term_matrix=None,
model=None, vocabulary=None, topic_no=None, overall_color='#053967',
figsize=(15, 7), dpi=None, overall_edgecolor=None,
overall_linewidth=None, overall_alpha=0.9, within_topic_edgecolor=None,
within_topic_linewidth=None, within_topic_alpha=0.9,
label_fontsize=15, move_yticks=0.4, num_keys=None,
tick_fontsize=14, legend_fontsize=15, legend=True,
enable_notebook=True):
"""Plots key frequencies overall and from within topic.
Args:
keys (list): A list of tokens. Defaults to None.
overall_freqs (list): A list of frequencies. Defaults to None.
within_topic_freqs (list): A list of frequencies. Defaults to None.
within_topic_color (str), optional: Color for topic frequencies bar. Defaults to
``#FF1727``.
document_term_matrix (pandas DataFrame), optional: A document-term matrix. Defaults
to None.
model, optional: A LDA model. Defaults to None.
vocabulary (list), optional: Vocabulary of the corpus. Defaults to None.
topic_no (int), optional: Number of topic. Defaults to None.
overall_color (str), optional: Color for overall frequencies bar. Defaults to ``#053967``.
figsize (tuple), optional: Size of the figure. Defaults to ``(15, 7)``.
dpi (int), optional: Dots per inch. Defaults to None.
overall_edgecolor (str), optional: Color for edgecolors of overall frequencies bar.
Defaults to None.
overall_linewidth (int), optional: Linewidth of overall frequencies bar. Defaults to
None.
overall_alpha (int), optional: Alpha for overall frequencies bar. Defaults to 0.9.
within_topic_edgecolor (str), optional: Color for edgecolors of overall frequencies bar.
Defaults to None.
within_topic_linewidth (int), optional: Linewidth of overall frequencies bar. Defaults to
None.
within_topic_alpha (int), optional: Alpha for overall frequencies bar. Defaults to 0.9.
label_fontsize (int), optional: Fontsize of x-axis and y-axis labels. Defaults to 15.
move_yticks (float), optional: Adjust this parameter to move ``yticks``. Defaults
to 0.4.
num_keys (int), optional: Number of tokens for y-axis. Defaults to None.
tick_fontsize (int), optional: Fontsize of x- and y-ticks. Defaults to 14.
legend_fontsize (int), optional: Fontsize of the legend. Defaults to 15.
legend (bool), optional: If True, legend will be displayed. Defaults to True.
enable_notebook (bool), optional: If True, enables :module:``matplotlib``
to show its figures within a Jupyter notebook.
Returns:
Figure object.
Example:
>>> keys = ['one', 'example']
>>> overall_freqs = [20, 10]
>>> within_topic_freqs = [10, 5]
>>> plot_key_frequencies(keys=keys,
... overall_freqs=overall_freqs,
... within_topic_freqs=within_topic_freqs,
... enable_notebook=False) # doctest: +ELLIPSIS
<matplotlib.figure.Figure object at ...>
"""
if enable_notebook:
from IPython import get_ipython
get_ipython().run_line_magic('matplotlib', 'inline')
if model:
within_topic_freqs = postprocessing.get_sorted_values_from_distribution(model.components_[topic_no],
model.components_[topic_no],
num_keys)
within_topic_freqs = [dist * len(vocabulary) for dist in within_topic_freqs]
total = [document_term_matrix[token].sum() for token in vocabulary]
overall_freqs = postprocessing.get_sorted_values_from_distribution(total,
model.components_[topic_no],
num_keys)
keys = postprocessing.get_sorted_values_from_distribution(vocabulary,
model.components_[topic_no],
num_keys)
fig, ax = plt.subplots(figsize=figsize, dpi=dpi)
y_axis = np.arange(len(keys))
overall = ax.barh(y_axis, overall_freqs, color=overall_color, edgecolor=overall_edgecolor,
linewidth=overall_linewidth, alpha=overall_alpha)
within = ax.barh(y_axis, within_topic_freqs, color=within_topic_color,
edgecolor=within_topic_edgecolor, linewidth=within_topic_linewidth,
alpha=within_topic_alpha)
ax.set_xlabel('Frequency', fontsize=label_fontsize)
ax.set_ylabel('Key', fontsize=label_fontsize)
ax.set_yticks(y_axis + move_yticks)
ax.set_yticklabels(keys, fontsize=tick_fontsize)
ax.set_ylim([0, len(keys)])
ax.tick_params(axis='x', labelsize=tick_fontsize)
if legend:
ax.legend(handles=[overall, within], labels=['Overall', 'Within Topic'], loc='best',
fontsize=legend_fontsize)
return fig
class PlotDocumentTopics:
"""
Class to visualize document-topic matrix.
"""
def __init__(self, document_topics, enable_notebook=True):
self.document_topics = document_topics
if enable_notebook:
self.enable_notebook = enable_notebook
self.show = self.notebook_handling()
@staticmethod
def notebook_handling():
"""Runs cell magic for Jupyter notebooks
"""
from IPython import get_ipython
get_ipython().run_line_magic('matplotlib', 'inline')
from bokeh.io import output_notebook, show
output_notebook()
return show
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=True):
"""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, move_yticks=0.4, 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:
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 + move_yticks)
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(**kwargs, transpose_data=True)
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',
responsive=True, 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``.
responsive (bool), optional: If True, ``sizing_mode`` is set to ``width``.
False sets ``sizing_mode`` to "fixed". Defaults to True.
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,
responsive=responsive,
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 = [('Document', '@Documents'),
('Topic', '@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')
if self.enable_notebook:
self.show(fig, notebook_handle=True)
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',
responsive=True, 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``.
responsive (bool), optional: If True, ``sizing_mode`` is set to ``width``.
False sets ``sizing_mode`` to "fixed". Defaults to True.
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,
responsive=responsive, 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')]
if self.enable_notebook:
self.show(fig, notebook_handle=True)
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``.
responsive (bool), optional: If True, ``sizing_mode`` is set to ``width``.
False sets ``sizing_mode`` to "fixed". Defaults to True.
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``.
responsive (bool), optional: If True, ``sizing_mode`` is set to ``width``.
False sets ``sizing_mode`` to "fixed". Defaults to True.
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, transpose_data=True)
def topic_over_time(self, pattern = r"\d{4}", 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:
labels(list): first three keys in a topic to select
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))
#doc_topicT = doc_topics.T
topiclabels = []
for topiclabel in doc_topics.index.values:
for topiclabel in topiclabels:
topic_over_threshold_per_year = []
mask = doc_topics.loc[topiclabel] > threshold
df = doc_topics.loc[topiclabel].loc[mask]
#df = doc_topics.loc[doc_topics.loc[topiclabel] > threshold]
#print (df)
d = defaultdict(int)
for item in df.index.values:
reg = regex.compile(pattern)
year = reg.findall(item)
d[year[0]]+=1
for year in years:
topic_over_threshold_per_year.append(d[str(year)])
plt.plot(years, topic_over_threshold_per_year, label=topiclabel)
plt.xlabel('Year')
plt.ylabel('count topics over threshold')
plt.legend()
fig = plt.gcf()
fig.set_size_inches(18.5, 10.5)
return fig
@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