-
Notifications
You must be signed in to change notification settings - Fork 1
/
pcwg03_plot.py
1693 lines (1156 loc) · 66.4 KB
/
pcwg03_plot.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
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
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
import os
import numpy as np
import pandas as pd
import itertools
import matplotlib.pyplot as plt
import matplotlib.gridspec as gridspec
import seaborn as sns
from matplotlib.colors import LinearSegmentedColormap
import geopandas as gpd
from descartes import PolygonPatch
import pcwg03_initialize as p_init
import pcwg03_config as pc
import pcwg03_convert_df as pcd
import pcwg03_energy_fraction as pef
import pcwg03_slice_df as psd
meta_df = p_init.meta_df
save_fig = pc.save_fig
dpi_choice = 600 # output plot resolution
fs, f14, f15, f16 = 12, 14, 15, 16
plt.rcParams.update({'font.size': fs})
xp_f1, yp_f1 = 0.04, 0.93 # x, y positions
fmt_code = '.1f' # for bootstrap t-test mean heatmap
def save_plot(sub_dir, var, plot_type, pdf=True):
"""Export figure to either pdf or png file."""
if not os.path.exists(p_init.out_plot_path+'/'+sub_dir):
os.makedirs(p_init.out_plot_path+'/'+sub_dir)
if pdf is True:
plt.savefig(p_init.out_plot_path+sub_dir+'/'+var+'_'+plot_type+'.pdf',
bbox_inches='tight', dpi=dpi_choice)
else:
plt.savefig(p_init.out_plot_path+'/'+sub_dir+'/'+var+'_'+plot_type+'.png',
bbox_inches='tight', dpi=dpi_choice)
def finish_plot(sub_dir, var, plot_type, tight_layout=True, save_fig=False, pdf=True):
"""Terminating procedures for plotting."""
if tight_layout is True:
plt.tight_layout()
if save_fig is True:
save_plot(sub_dir, var, plot_type, pdf)
plt.show()
def plot_pdm_example():
"""Plot example power deviation matrix.
Using input from Excel file.
"""
file = p_init.py_file_path+'/pdm_example.xls'
df = pd.read_excel(file)
df.set_index('TI', inplace=True)
ax = sns.heatmap(df, cmap='RdYlBu', robust=True, center=0, cbar_kws={'label': 'Power deviation(%)'})
# ax.set_xlabel(r'Wind speed (m s$^{-1}$)')
ax.set_xlabel('Normalized wind speed')
ax.set_ylabel('TI (%)')
finish_plot('meta', 'pdm_example', 'heatmap')
def plot_wsti_energy_fraction_box():
"""Plot 4 box plots for WS-TI, ITI-OS, and Inner-Outer Ranges.
A pair of box plots on energy and data fractions, and a pair box plots on NMEs.
Similar to `plot_outws_energy_fraction_box`.
"""
ef_filter_df1, ef_filter_df2, wsti_nme_df1, wsti_nme_df2 = pef.get_wsti_ef_nme()
fig, ((ax1, ax2), (ax3, ax4)) = plt.subplots(2, 2, figsize=(8, 6), gridspec_kw={'width_ratios': [3, 1]})
ax1 = sns.boxplot(x='bin_name', y='value', data=ef_filter_df1, hue='error_name',
palette='colorblind', ax=ax1)
ax1 = sns.swarmplot(x='bin_name', y='value', data=ef_filter_df1, hue='error_name',
alpha=0.7, dodge=True, ax=ax1)
ax1.set_ylabel('Fraction (%)')
ax1.set_xlabel('')
ax1.set_ylim([-5, 95])
handles, labels = ax1.get_legend_handles_labels()
ax1.legend(handles[:2], ['Data fraction', 'Energy fraction'], loc='upper left')
ax2 = sns.boxplot(x='bin_name', y='value', data=ef_filter_df2, hue='error_name',
palette='colorblind', ax=ax2)
ax2 = sns.swarmplot(x='bin_name', y='value', data=ef_filter_df2, hue='error_name',
alpha=0.7, dodge=True, ax=ax2)
ax2.legend_.remove()
ax2.set_xlabel('')
ax2.set_ylabel('')
ax2.set_ylim([-5, 95])
ax2.set_yticklabels([])
ax3 = sns.boxplot(x='bin_name', y='error_value', data=wsti_nme_df1, hue='error_name',
palette='BuGn_r', ax=ax3)
ax3.axhline(0, ls='--', color='grey')
ax3.set_ylabel('NME (%)')
ax3.set_xlabel('WS-TI bins')
ax3.legend_.remove()
ax3.set_ylim([-2.5, 2.5])
ax4 = sns.boxplot(x='bin_name', y='error_value', data=wsti_nme_df2, hue='error_name',
palette='BuGn_r', ax=ax4)
ax4.axhline(0, ls='--', color='grey')
ax4.set_ylabel('')
ax4.set_xlabel('Inner-Outer Range')
ax4.legend_.remove()
ax4.set_ylim([-2.5, 2.5])
ax4.set_yticklabels([])
ax1.text(0.94, 0.9, '(a)', color='k', fontsize=12, transform=ax1.transAxes)
ax2.text(0.82, 0.9, '(b)', color='k', fontsize=12, transform=ax2.transAxes)
ax3.text(0.94, 0.9, '(c)', color='k', fontsize=12, transform=ax3.transAxes)
ax4.text(0.82, 0.9, '(d)', color='k', fontsize=12, transform=ax4.transAxes)
finish_plot('meta', 'wsti_energyfraction', 'box')
def plot_outws_energy_fraction_box():
"""Plot 2 box plots for Outer Range WS.
A box plot on energy and data fractions, and a box plot on NMEs.
Similar to `plot_wsti_energy_fraction_box`.
"""
dc_ef_all_df_s, outws_nme_df_s = pef.get_outws_ef_nme()
fig, (ax1, ax2) = plt.subplots(2, 1, figsize=(12, 8))
ax1 = sns.boxplot(x='bin_name', y='value', data=dc_ef_all_df_s, hue='error_name',
palette='colorblind', ax=ax1)
ax1.set_ylabel('Fraction (%)', fontsize=f15)
ax1.set_xlabel('')
ax1.tick_params(labelsize=f15)
ax1.xaxis.set_tick_params(rotation=45)
handles, labels = ax1.get_legend_handles_labels()
ax1.legend(handles[:2], ['Data fraction', 'Energy fraction'], fontsize=f15)
ax2 = sns.boxplot(x='bin_name', y='error_value', data=outws_nme_df_s, hue='error_name',
palette='BuGn_r', ax=ax2)
ax2.axhline(0, ls='--', color='grey')
ax2.set_ylabel('NME (%)', fontsize=f15)
ax2.set_xlabel('Normalized wind speed category', fontsize=f15)
ax2.set_ylim([-1.2, 1.2])
ax2.legend_.remove()
ax2.tick_params(labelsize=f15)
ax2.xaxis.set_tick_params(rotation=45)
ax1.text(0.94, 0.91, '(a)', color='k', fontsize=f15, transform=ax1.transAxes)
ax2.text(0.94, 0.91, '(b)', color='k', fontsize=f15, transform=ax2.transAxes)
finish_plot('meta', 'outws_energyfraction', 'box')
def plot_hist_series(df, var, name):
"""Plot histogram for series of meta data."""
p_series = pd.Series(df[var])
p_series.value_counts(dropna=False).plot(kind='bar', rot=45)
plt.ylabel('Count')
plt.title(name)
finish_plot('meta', 'meta_' + var, 'hist')
def loop_meta_hist():
"""Generate histograms from available meta data."""
for var, name in zip(pc.meta_var_names_turb, pc.meta_xls_names_turb):
plot_hist_series(p_init.meta_df, var, name)
def plot_group_meta_hist():
"""Plot 4 histograms using grouped bins on x-axis."""
hist_df1 = pd.DataFrame({'turbi_dia_grouped': meta_df['turbi_dia_grouped'].value_counts(dropna=False)})
hist_df1.reset_index(inplace=True)
hist_df1.replace('143 - 154', '120+', inplace=True)
x_sorted1, x_nozero1 = psd.remove_0_in_label(hist_df1)
hist_df2 = pd.DataFrame({'turbi_hh_grouped': meta_df['turbi_hh_grouped'].value_counts(dropna=False)})
hist_df2.reset_index(inplace=True)
hist_df2.replace('132 - 143', '110+', inplace=True)
hist_df2.replace('110 - 121', '110+', inplace=True)
hist_df22 = hist_df2.groupby(['index'], as_index=False).agg('sum')
x_sorted2, x_nozero2 = psd.remove_0_in_label(hist_df22)
hist_df3 = pd.DataFrame({'turbi_spower_grouped': meta_df['turbi_spower_grouped'].value_counts(dropna=False)})
hist_df3.reset_index(inplace=True)
hist_df3.replace('489 - 536', '441+', inplace=True)
hist_df3.replace('536 - 583', '441+', inplace=True)
hist_df33 = hist_df3.groupby(['index'], as_index=False).agg('sum')
x_sorted3, x_nozero3 = psd.remove_0_in_label(hist_df33)
meta_var_names_array = np.array(pc.meta_var_names)
year_measure_idx = np.where(meta_var_names_array == 'year_measuremt')
year_str = pc.meta_var_names[int(year_measure_idx[0])]
hist_df4 = pd.DataFrame({year_str: meta_df[year_str].value_counts(dropna=False)})
hist_df4.reset_index(inplace=True)
x_sorted4, x_nozero4 = psd.remove_0_in_label(hist_df4)
fig, ((ax1, ax2), (ax3, ax4)) = plt.subplots(2, 2, figsize=(10, 10))
sns.barplot(x='index', y='turbi_dia_grouped', data=hist_df1, ax=ax1, order=x_sorted1, color='c')
ax1.set_xticklabels(labels=x_nozero1['index'], rotation=45)
ax1.set_xlabel('Turbine rotor diameter (m)', fontsize=fs+1)
ax1.set_ylabel('Count', fontsize=fs+1)
sns.barplot(x='index', y='turbi_hh_grouped', data=hist_df22, ax=ax2, order=x_sorted2, color='c')
ax2.set_xticklabels(labels=x_nozero2['index'], rotation=45)
ax2.set_xlabel('Turbine hub height (m)', labelpad=10, fontsize=fs+1)
ax2.set_ylabel('')
sns.barplot(x='index', y='turbi_spower_grouped', data=hist_df33, ax=ax3, order=x_sorted3, color='c')
ax3.set_xticklabels(labels=x_nozero3['index'], rotation=45)
ax3.set_xlabel(r'Turbine specific power (W m$^{-2}$)', fontsize=fs+1)
ax3.set_ylabel('Count', fontsize=fs+1)
sns.barplot(x='index', y=year_str, data=hist_df4, ax=ax4, order=x_sorted4, color='c')
ax4.set_xticklabels(labels=x_nozero4['index'], rotation=45)
ax4.set_xlabel('Year of measurement', labelpad=25, fontsize=fs+1)
ax4.set_ylabel('')
ax1.text(xp_f1, yp_f1, '(a)', color='k', fontsize=fs, transform=ax1.transAxes)
ax2.text(xp_f1, yp_f1, '(b)', color='k', fontsize=fs, transform=ax2.transAxes)
ax3.text(xp_f1, yp_f1, '(c)', color='k', fontsize=fs, transform=ax3.transAxes)
ax4.text(xp_f1, yp_f1, '(d)', color='k', fontsize=fs, transform=ax4.transAxes)
finish_plot('meta', 'meta_', 'hist_ranked')
def plot_map():
"""Map submission origins, if available."""
country_series = meta_df['geog_country'].value_counts(dropna=False)
country_series = country_series.rename_axis('country').reset_index()
country_na = country_series.loc[country_series['country'].isnull()].index
country_series_plot = country_series.drop(country_na[0]) # drop NaN
# count of NaN
nan_country = str(country_series.loc[country_series['country'].isnull()]['geog_country'][0])
total_country = str(len(meta_df['geog_country']))
colmap = plt.cm.get_cmap('viridis')
colmap(1)
country_num_max = max(country_series_plot['geog_country'])
country_num = np.linspace(0, 1, country_num_max + 1)
country_num_on_map = country_num[country_series_plot['geog_country']]
world = gpd.read_file(gpd.datasets.get_path('naturalearth_lowres'))
def plot_country_patch(axes, country_name, fcolor):
# plot a country on the provided axes
nami = world[world.name == country_name]
namigm = nami.__geo_interface__['features'] # geopandas's geo_interface
namig0 = {'type': namigm[0]['geometry']['type'], 'coordinates': namigm[0]['geometry']['coordinates']}
axes.add_patch(PolygonPatch(namig0, fc=fcolor, ec='black', alpha=0.85, zorder=2))
colmap = plt.cm.get_cmap('viridis')
ax = world.plot(figsize=(8, 4), edgecolor=u'gray', color='w')
for x, y in zip(country_series_plot['country'].values, country_num_on_map):
plot_country_patch(ax, x, colmap(y))
plt.ylabel('Latitude')
plt.xlabel('Longitude')
plt.title(nan_country + ' of ' + total_country + ' submissions have unknown countries')
fig = ax.get_figure()
cax = fig.add_axes([0.95, 0.1, 0.03, 0.8])
sm = plt.cm.ScalarMappable(cmap=colmap)
sm._A = []
cbr = fig.colorbar(sm, cax=cax)
cbr.set_ticks(np.linspace(0, 1, country_num_max))
cbr.ax.set_yticklabels(np.linspace(1, country_num_max, country_num_max))
finish_plot('meta', 'meta', 'map', tight_layout=False)
def plot_nme_hist():
"""Plot NME histograms: pre-NME-filtering and post-NME-filtering."""
plt.figure(1, figsize=(8, 6))
plt.rcParams.update({'font.size': 14})
gridspec.GridSpec(2, 2)
# need to get pre-nme-filter error data frame
df1 = pcd.get_error_df_dict(p_init.data_file)['base_total_e']
df1p = df1.loc[(df1['error_cat'] == 'by_range') & (df1['error_name'] == 'nme')]
inner_nme = df1p['error_value'].loc[df1p['bin_name'] == 'Inner'] * 100
outer_nme = df1p['error_value'].loc[df1p['bin_name'] == 'Outer'] * 100
ax1 = plt.subplot2grid((2, 2), (0, 0), colspan=2, rowspan=1)
a_value = 0.7
sns.distplot(list(outer_nme.values), color='#73c0c4', label='Outer Range', bins=10, kde=False, ax=ax1,
hist_kws={'alpha': a_value})
sns.distplot(list(inner_nme.values), color='#3c758b', label='Inner Range', bins=5, kde=False, ax=ax1,
hist_kws={'alpha': a_value})
plt.ylabel('Count')
plt.xlabel('NME (%)')
plt.legend()
sheet_bt_choice = 'base_total_e'
df2 = p_init.error_df
def choose_in_out_def(in_or_out, in_out_def):
selection = ((df2[sheet_bt_choice]['error_cat'] == 'by_range')
& (df2[sheet_bt_choice]['error_name'] == 'nme')
& (df2[sheet_bt_choice]['bin_name'] == in_or_out)
& (df2[sheet_bt_choice]['file_name']
.isin(meta_df.loc[meta_df['inner_def'] == in_out_def]['file_name'])))
return selection
inner_a = df2[sheet_bt_choice].loc[choose_in_out_def('Inner', 'A')]['error_value'] * 100
inner_b = df2[sheet_bt_choice].loc[choose_in_out_def('Inner', 'B')]['error_value'] * 100
inner_c = df2[sheet_bt_choice].loc[choose_in_out_def('Inner', 'C')]['error_value'] * 100
outer_a = df2[sheet_bt_choice].loc[choose_in_out_def('Outer', 'A')]['error_value'] * 100
outer_b = df2[sheet_bt_choice].loc[choose_in_out_def('Outer', 'B')]['error_value'] * 100
outer_c = df2[sheet_bt_choice].loc[choose_in_out_def('Outer', 'C')]['error_value'] * 100
p23_c = ['seagreen', 'limegreen', 'lawngreen']
# p23_c = ['red', 'darkorange', 'gold']
ax2 = plt.subplot2grid((2, 2), (1, 0))
ax2.hist([inner_a, inner_b, inner_c], stacked=True, color=p23_c)
ax2.set_ylabel('Count')
ax2.set_xlabel('Filtered Inner Range NME (%)')
ax3 = plt.subplot2grid((2, 2), (1, 1))
ax3.hist([outer_a, outer_b, outer_c], stacked=True, color=p23_c)
ax3.set_ylabel('Count')
ax3.set_xlabel('Filtered Outer Range NME (%)')
labels = ['A', 'B', 'C']
plt.legend(labels, title='Definition', loc='upper left')
ax1.text(0.03, 0.89, '(a)', color='k', fontsize=12, transform=ax1.transAxes)
ax2.text(0.05, 0.88, '(b)', color='k', fontsize=12, transform=ax2.transAxes)
ax3.text(0.89, 0.88, '(c)', color='k', fontsize=12, transform=ax3.transAxes)
finish_plot('error_hist', 'nme', '3def_hist')
plt.rcParams.update({'font.size': fs})
def plot_wsti_nme_box():
"""Plot 4 panel box plots for WS-TI NME."""
box_plot_y_scale = 0.5 # zoom in
# box_plot_y_scale = 1
def ws_ti_df_by_sheet(sheet_name_short, df, bt_choice, error_name, file_num):
sheet_bt_choice = sheet_name_short + bt_choice
ws_ti_df = df[sheet_bt_choice].loc[(df[sheet_bt_choice]['error_cat'] == 'by_ws_ti')
& (df[sheet_bt_choice]['error_name'] == error_name)]
if file_num is True:
sheet_name_end = ': ' + str(round(len(ws_ti_df) / 4))
else:
sheet_name_end = ''
ws_ti_df.insert(0, 'sheet_name', str(sheet_name_short)[:-1] + sheet_name_end)
return ws_ti_df
def loop_box_plot(bt_choice, error_name, error_df, file_num_choice=False, extra_error_df=None):
for i, i_short in zip(pc.matrix_sheet_name, pc.matrix_sheet_name_short):
dum_df = ws_ti_df_by_sheet(i_short, error_df, bt_choice, error_name, file_num=file_num_choice)
if pc.matrix_sheet_name.index(i) == 0:
ws_ti_df = dum_df
else:
ws_ti_df = ws_ti_df.append(dum_df)
if extra_error_df is not None:
for i, i_short in zip(pc.correction_list, pc.extra_matrix_sheet_name_short):
dum_df = ws_ti_df_by_sheet(i_short, extra_error_df, bt_choice, error_name,
file_num=file_num_choice)
if pc.correction_list.index(i) == 0:
ws_ti_df_extra = dum_df
else:
ws_ti_df_extra = ws_ti_df_extra.append(dum_df)
ws_ti_df = ws_ti_df.append(ws_ti_df_extra)
ws_ti_df['error_value'] = ws_ti_df['error_value'].astype(float) * 100
ws_ti_error_min, ws_ti_error_max = ws_ti_df['error_value'].min(), ws_ti_df['error_value'].max()
ws_ti_error_abs_max = np.max([abs(ws_ti_error_min), abs(ws_ti_error_max)])
lws_lti_df = ws_ti_df.loc[ws_ti_df['bin_name'] == 'LWS-LTI']
lws_hti_df = ws_ti_df.loc[ws_ti_df['bin_name'] == 'LWS-HTI']
hws_lti_df = ws_ti_df.loc[ws_ti_df['bin_name'] == 'HWS-LTI']
hws_hti_df = ws_ti_df.loc[ws_ti_df['bin_name'] == 'HWS-HTI']
fig, ((ax1, ax2), (ax3, ax4)) = plt.subplots(2, 2, figsize=(14, 8))
def plot_box_by_sheet(df, ax, sub_t):
# add grey to colorblind... manually
# list(sns.color_palette(['grey']))
# sns.color_palette('colorblind')
new_p = [(0.5019607843137255, 0.5019607843137255, 0.5019607843137255),
(0.00392156862745098, 0.45098039215686275, 0.6980392156862745),
(0.8705882352941177, 0.5607843137254902, 0.0196078431372549),
(0.00784313725490196, 0.6196078431372549, 0.45098039215686275),
(0.8352941176470589, 0.3686274509803922, 0.0),
(0.8, 0.47058823529411764, 0.7372549019607844),
(0.792156862745098, 0.5686274509803921, 0.3803921568627451),
(0.984313725490196, 0.6862745098039216, 0.8941176470588236),
(0.5803921568627451, 0.5803921568627451, 0.5803921568627451),
(0.9254901960784314, 0.8823529411764706, 0.2),
(0.33725490196078434, 0.7058823529411765, 0.9137254901960784)]
# ax = sns.boxplot(x='sheet_name', y='error_value', data=df, ax=ax, palette=new_p)
ax = sns.boxplot(x='sheet_name', y='error_value', data=df, ax=ax, palette='colorblind')
# ax = sns.swarmplot(x='sheet_name', y='error_value', data=df, ax=ax, palette='colorblind')
if error_name == 'nme':
ax.axhline(0, ls='--', color='grey')
ax.set_xticklabels(ax.get_xticklabels(), rotation=30)
ax.set_title(df['bin_name'].iloc[0])
ax.set_ylabel(bt_choice + 'nergy ' + error_name + ' (%)')
# ax.set_ylim([ws_ti_error_min*box_plot_y_scale, ws_ti_error_max*box_plot_y_scale])
ax.set_ylim([-ws_ti_error_abs_max * box_plot_y_scale, ws_ti_error_abs_max * box_plot_y_scale])
ax.text(0.95, 0.92, sub_t, color='k', fontsize=12, transform=ax.transAxes)
return ax
plot_box_by_sheet(lws_lti_df, ax1, '(a)')
plot_box_by_sheet(lws_hti_df, ax2, '(b)')
plot_box_by_sheet(hws_lti_df, ax3, '(c)')
plot_box_by_sheet(hws_hti_df, ax4, '(d)')
if extra_error_df is not None:
var = 'wsti_nme_boxplot_extra'
else:
var = 'wsti_nme_boxplot'
finish_plot('results', var, bt_choice + '_' + error_name)
loop_box_plot('total_e', 'nme', p_init.error_df, extra_error_df=p_init.extra_error_df)
def plot_nme_avg_spread_heatmap(ee_df=None, rr_choice=None):
"""Mass generate heatmaps of NME average and NME spread."""
def loop_nme_avg_spread_heatmap(by_bin, bt_choice, error_name, e_df=p_init.error_df, ee_df=None,
rr_choice=pc.robust_resistant_choice):
for idx, i_short in enumerate(pc.matrix_sheet_name_short):
df = psd.get_error_in_bin(e_df, i_short+bt_choice, by_bin, error_name)
u_bin, average, spread = psd.find_unique_bin_create_dum(df['bin_name'])
if idx == 0:
average_df = pd.DataFrame(index=u_bin, columns=[pc.matrix_sheet_name_short])
spread_df = pd.DataFrame(index=u_bin, columns=[pc.matrix_sheet_name_short])
psd.cal_average_spread(df, u_bin, average_df, spread_df, i_short, rr_choice)
if ee_df is not None:
for idx, i_short in enumerate(pc.extra_matrix_sheet_name_short):
df = psd.get_error_in_bin(p_init.extra_error_df, i_short+bt_choice, by_bin, error_name)
u_bin, average, spread = psd.find_unique_bin_create_dum(df['bin_name'])
psd.cal_average_spread(df, u_bin, average_df, spread_df, i_short, rr_choice)
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14, 6))
if error_name == 'nme':
sns.heatmap(average_df, linewidths=.5, cmap='RdBu_r', center=0, ax=ax1)
elif error_name == 'nmae':
sns.heatmap(average_df, linewidths=.5, ax=ax1)
if rr_choice is None:
avg_title, spd_title, out_name = 'mean', 'standard deviation', 'mean_sd'
else:
avg_title, spd_title, out_name = 'median', 'interquartile range', 'median_iqr'
ax1.set_xlabel('correction methods')
ax1.set_ylabel(by_bin)
ax1.yaxis.set_tick_params(rotation=0)
ax1.set_title(bt_choice + 'nergy ' + avg_title + ' ' + error_name + ' (%)')
sns.heatmap(spread_df, linewidths=.5, cmap='viridis', ax=ax2)
ax2.set_xlabel('correction methods')
ax2.set_ylabel(by_bin)
ax2.yaxis.set_tick_params(rotation=0)
ax2.set_title(bt_choice + 'nergy ' + spd_title + ' of ' + error_name + ' (%)')
if ee_df is not None:
var = 'nme_avg_spread_boxplot_extra'
else:
var = 'nme_avg_spread_boxplot'
finish_plot('results', var, bt_choice + '_heatmap')
for idx, (cat_i, bt_j, error_k) in enumerate(itertools.product(pc.error_cat_short[1:],
pc.bt_choice, pc.error_name[1:])):
loop_nme_avg_spread_heatmap(cat_i, bt_j, error_k,
e_df=p_init.error_df, ee_df=ee_df,
rr_choice=rr_choice)
def plot_inner_outer_data_count_box_hist():
"""Plot Inner Range and Outer Range data count"""
box_io_df = p_init.error_df['base_total_e'].loc[(p_init.error_df['base_total_e']['error_cat'] == 'by_range')
& (p_init.error_df['base_total_e']['error_name'] == 'data_count')]
u_bin = box_io_df['bin_name'].unique()
for idx, val in enumerate(u_bin):
data_count = box_io_df.loc[box_io_df['bin_name'] == val]['error_value'].values
if idx == 0:
box_dict = {val: data_count}
else:
box_dict[val] = data_count
box_df = pd.DataFrame(box_dict)
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(10, 5))
sns.boxplot(data=box_df, ax=ax1, palette='GnBu_d')
ax1.set_xticklabels(labels=['Inner Range', 'Outer Range'])
ax1.set_ylabel('10-minute data count')
base_inner_count = p_init.error_df['base_bin_e'].loc[(p_init.error_df['base_bin_e']['error_cat'] == 'by_range')
& (p_init.error_df['base_bin_e']['bin_name'] == 'Inner')
& (p_init.error_df['base_bin_e']['error_name']
== 'data_count')]
base_outer_count = p_init.error_df['base_bin_e'].loc[(p_init.error_df['base_bin_e']['error_cat'] == 'by_range')
& (p_init.error_df['base_bin_e']['bin_name'] == 'Outer')
& (p_init.error_df['base_bin_e']['error_name']
== 'data_count')]
ratio_inout_count = pd.Series([base_outer_count['error_value'].values / base_inner_count['error_value'].values])
ax2.hist(ratio_inout_count.values, color='grey')
ax2.set_ylabel('File count')
ax2.set_xlabel(r'$\mathrm{\mathsf{\frac{Outer\/\/Range\/\/data\/\/count}{Inner\/\/Range\/\/data\/\/count}}}$',
fontsize=fs + 5)
ax1.text(xp_f1, yp_f1, '(a)', color='k', fontsize=fs, transform=ax1.transAxes)
ax2.text(xp_f1, yp_f1, '(b)', color='k', fontsize=fs, transform=ax2.transAxes)
finish_plot('meta', 'data_count', 'box_hist')
def plot_file_data_count_hist_box():
"""Mass generate histogram of file count and box plot of data count for each category.
Check if every sheet/method has the same file count for each inflow category.
"""
def loop_count_histbox(by_bin, e_df, sheet_name, no_hist=None):
for idx, i_short in enumerate(sheet_name):
do_plot = False
bt_choice = 'bin_e' # same data count for total_e
df = e_df[i_short + bt_choice].loc[(e_df[i_short + bt_choice]['error_cat'] == by_bin)
& (e_df[i_short + bt_choice]['error_name'] == 'data_count')]
u_bin = df['bin_name'].unique()
nan_bin_count = np.zeros(len(u_bin))
for i in range(len(df)):
for index, val in enumerate(u_bin):
if df.iloc[i]['bin_name'] == val:
if pd.isnull(df.iloc[i]['error_value']):
nan_bin_count[index] += 1
file_num = len(df) / len(u_bin)
bin_count = file_num - nan_bin_count
try:
lump_bin_count # see if data duplicates already
except NameError:
lump_bin_count = bin_count
lump_bin_count = np.expand_dims(lump_bin_count, axis=0)
do_plot = True
if lump_bin_count.shape[0] > 1:
# for i in range(lump_bin_count.shape[0]):
for i in lump_bin_count:
if np.array_equal(bin_count, i):
do_plot = False
else:
do_plot = True
# if data count not duplicate
if do_plot:
print(i_short + ' contains unique file count distribution for ' + by_bin)
file_data = {'d_bin': u_bin, 'd_count': bin_count}
hist_df = pd.DataFrame(file_data)
for index, val in enumerate(u_bin):
data_count = df.loc[df['bin_name'] == val]['error_value'].values
if index == 0:
box_dict = {val: data_count}
else:
box_dict[val] = data_count
box_df = pd.DataFrame(box_dict)
if no_hist is None:
no_hist = ''
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 5))
sns.barplot(x='d_bin', y='d_count', data=hist_df, ax=ax1, color='orchid')
ax1.set_xticklabels(labels=hist_df['d_bin'], rotation=45)
ax1.set_title(str(i_short)[:-1] + ', max: ' + str(int(np.max(bin_count))) + ' files')
ax1.set_xlabel(by_bin)
ax1.set_ylabel('file count')
sns.boxplot(data=box_df, ax=ax2, palette='colorblind')
ax2.set_xticklabels(labels=hist_df['d_bin'], rotation=45)
ax2.set_title(str(i_short)[:-1])
ax2.set_xlabel(by_bin)
ax2.set_ylabel('data count')
if len(hist_df['d_bin']) > 15: # every other xtick label
for label_ax1, label_ax2 in zip(ax1.xaxis.get_ticklabels()[::2],
ax2.xaxis.get_ticklabels()[::2]):
label_ax1.set_visible(False)
label_ax2.set_visible(False)
elif no_hist == 'box':
no_hist = no_hist + '_'
fig, ax2 = plt.subplots(figsize=(6, 5))
sns.boxplot(data=box_df, ax=ax2, palette='colorblind')
ax2.set_xticklabels(labels=hist_df['d_bin'], rotation=45)
ax2.set_title(str(i_short)[:-1])
ax2.set_xlabel(by_bin)
ax2.set_ylabel('data count')
if len(hist_df['d_bin']) > 15: # every other xtick label
for label_ax2 in ax2.xaxis.get_ticklabels()[::2]:
label_ax2.set_visible(False)
finish_plot('meta', no_hist + by_bin + '_' + str(i_short)[:-1], 'count_hist_box')
# if i > 1:
lump_bin_count = np.concatenate((lump_bin_count, np.expand_dims(bin_count, axis=0)), axis=0)
else:
print(i_short + ' contains duplicating file count distribution for ' + by_bin)
del lump_bin_count
for i in pc.error_cat_short[1:]:
loop_count_histbox(i, p_init.error_df, pc.matrix_sheet_name_short)
loop_count_histbox(i, p_init.extra_error_df, pc.extra_matrix_sheet_name_short)
def loop_outer_diff_scatter(meta_var, one_plot=None, diff_choice=None):
"""Plot scatter plot between meta data and error, or between meta data and error difference.
Pair of plots: NME and NMAE
"""
if diff_choice is None:
y_var = 'error_value'
y_title_end = ''
else:
y_var = 'diff'
y_title_end = ' - Baseline'
def add_corr_text(ax, corr):
add_x = 0.13
if corr.size != 0:
ax.text(0.7, 0.94, 'correlation', color='k',
weight='semibold', transform=ax.transAxes) # ratio of axes
for idx, val in enumerate(corr):
ax.text(0.75, 1 - add_x, val, color=sheet_color[idx],
weight='semibold', transform=ax.transAxes)
add_x += 0.07
for bt_c in pc.bt_choice:
########################
##### SKIPPING ONE #####
########################
if meta_var != 'year_operatn': # HARD CODED -- no useful data there
nme_df, nme_corr = psd.get_outer_meta('nme', meta_var, bt_c, y_var)
nmae_df, nmae_corr = psd.get_outer_meta('nmae', meta_var, bt_c, y_var)
if diff_choice is not None:
nme_df = nme_df.loc[~(nme_df['sheet'] == 'base')]
nmae_df = nmae_df.loc[~(nmae_df['sheet'] == 'base')]
nme_file_num = str(round(nme_df[meta_var].count()/len(nme_df['sheet'].unique())))[:-2]
nmae_file_num = str(round(nmae_df[meta_var].count()/len(nmae_df['sheet'].unique())))[:-2]
if meta_var == 'turbi_spower': # change units for specific power
nme_df[meta_var] = nme_df[meta_var] * 1e3
nmae_df[meta_var] = nmae_df[meta_var] * 1e3
if one_plot is None:
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 4))
else:
fig, ax1 = plt.subplots(figsize=(6, 4))
axx = sns.color_palette(palette='colorblind')
sheet_color = axx.as_hex()
if one_plot is None:
sns.scatterplot(x=meta_var, y=y_var, hue='sheet', data=nme_df, alpha=0.5,
palette='colorblind', ax=ax1, legend=False)
sns.scatterplot(x=meta_var, y=y_var, hue='sheet', data=nmae_df, alpha=0.5,
palette='colorblind', ax=ax2)
ax2.set_xlabel(meta_var+': '+nmae_file_num+' files')
ax2.set_ylabel(bt_c+' nmae'+y_title_end)
ax2.axhline(y=0, linestyle='--', color='grey')
ax2.legend(bbox_to_anchor=(1.05, 1), loc=2, borderaxespad=0.)
add_corr_text(ax2, nmae_corr)
op_text = ''
# plot NMEs only
else:
sns.scatterplot(x=meta_var, y=y_var, hue='sheet', data=nme_df, alpha=0.5,
palette='colorblind', ax=ax1)
ax1.legend(bbox_to_anchor=(1.05, 1), loc=2, borderaxespad=0.)
op_text = '_one'
ax1.set_xlabel(meta_var+': '+nme_file_num+' files')
ax1.set_ylabel(bt_c+' nme'+y_title_end)
ax1.axhline(y=0, linestyle='--', color='grey')
add_corr_text(ax1, nme_corr)
finish_plot('results', 'scatter_diff_outer', meta_var+op_text+'_'+bt_c)
def plot_outer_diff_scatter(one_plot=None, diff_choice=None):
"""Mass generate scatter plot of error or error difference."""
for meta_var in pc.meta_var_names_turb:
loop_outer_diff_scatter(meta_var, one_plot=one_plot, diff_choice=diff_choice)
def plot_outer_nme_inner_dc_scatter():
"""Generate scatter plot between Outer Range NME and Inner Range data count."""
for sheet in pc.matrix_sheet_name_short:
target_df = p_init.error_df[sheet+'total_e']
outer_nme = target_df.loc[(target_df['error_cat'] == 'by_range') & (target_df['error_name'] == 'nme')
& (target_df['bin_name'] == 'Outer')]
inner_dc = target_df.loc[(target_df['error_cat'] == 'by_range') & (target_df['error_name'] == 'data_count')
& (target_df['bin_name'] == 'Inner')]
two_df_out_nme_in_dc = pd.merge(outer_nme, inner_dc, on='file_name')
ax = sns.scatterplot(x='error_value_y', y='error_value_x', data=two_df_out_nme_in_dc)
ax.set_title(sheet)
ax.set_ylabel('outer range nme')
ax.set_xlabel('inner range data count')
finish_plot('results', 'scatter_outer_nme', 'inner_dc')
def plot_nme_diff_box_range_scatter_hist():
"""Produce a panel of box plot, scatter, and histogram.
Plot absolute NME difference box plots for each submission.
Plot statistical range of absolute NME differences for each submission.
Plot histogram for the statistical ranges.
"""
nme_diff_df, nme_range_p_df = psd.get_nme_diff_range()
gs = gridspec.GridSpec(105, 100)
plt.subplots(figsize=[16, 6])
top_start, top_end = 0, 50
center_left = 75
ax1 = plt.subplot(gs[top_start:top_end, 0:center_left])
ax21 = plt.subplot(gs[top_end + 5::, 0:center_left])
ax22 = plt.subplot(gs[top_end + 5:, center_left + 4:])
sns.boxplot(data=nme_diff_df, color='orange', ax=ax1)
# fy_cp = ['orange'] * nme_diff_df.shape[1]
# sns.swarmplot(data=nme_diff_df, palette=fy_cp, ax=ax1)
ax1.set(xticklabels=[])
ax1.tick_params(bottom=False)
ax1.set_ylabel('|NME| difference (%)')
ax1.axhline(0, ls='--', color='grey')
pfy_c = ['dodgerblue', 'grey', 'red']
pf_m = ['^', 'o', 'v']
pf_ms = []
for item in nme_range_p_df['all']:
if item == 'Improved' or item == 'Worse':
pf_ms.append(110)
else:
pf_ms.append(80)
sns.scatterplot(x='index', y='nme', data=nme_range_p_df, hue='all', palette=pfy_c, style='all',
markers=pf_m, s=pf_ms, ax=ax21)
handles, labels = ax21.get_legend_handles_labels()
ax21.legend(handles[1:], labels[1:], ncol=3)
ax21.set_xlim([-0.6, 51.6])
ax21.set(xticklabels=[])
# ax2.set_markersizes = pf_ms
ax21.tick_params(bottom=False)
ax21.set_ylabel('Range of \n|NME| differences (%)')
ax21.set_xlabel('Data set submission')
ax21.xaxis.labelpad = 10
nme_range_p_df['nme'].hist(orientation='horizontal', color='k', ax=ax22)
# ax22 = plt.hist(nme_range_p_df['nme'].values, orientation='horizontal', color='k')
ax22.set_xlabel('Count')
ax22.grid(False)
xp_fx, yp_fx = 0.97, 0.88
ax1.text(xp_fx, yp_fx, '(a)', color='k', fontsize=fs, transform=ax1.transAxes)
ax21.text(xp_fx, yp_fx, '(b)', color='k', fontsize=fs, transform=ax21.transAxes)
plt.text(xp_fx - 0.07, yp_fx, '(c)', color='k', fontsize=fs, transform=ax22.transAxes)
finish_plot('results', 'nme_diff_box', 'range_scatter_hist', tight_layout=False)
def plot_wsti_nme_avg_spread_heatmap():
"""Plot 1 average and spread NME heatmap for WS-TI, ITI-OS, and Inner-Outer Ranges."""
average_df, spread_df, dum = psd.get_wsti_nme_stat()
average_df = psd.sort_plot_wsti_df_index(average_df)
spread_df = psd.sort_plot_wsti_df_index(spread_df)
average_df.rename(columns=pc.method_dict, inplace=True)
average_df1 = average_df.iloc[0:5]
average_df3 = average_df.iloc[5:]
ax13_max = average_df.values.max()
ax13_min = average_df.values.min()
ax13_mm = np.max([abs(ax13_max), abs(ax13_min)])
spread_df.rename(columns=pc.method_dict, inplace=True)
spread_df2 = spread_df.iloc[0:5]
spread_df4 = spread_df.iloc[5:]
ax24_max = spread_df.values.max()
ax24_min = spread_df.values.min()
plt.rcParams.update({'font.size': f16})
fig, ((ax1, ax2), (ax3, ax4)) = plt.subplots(2, 2, figsize=(14, 9), gridspec_kw={'height_ratios': [1, 1.2]})
sns.heatmap(average_df1, linewidths=.5, cmap='RdBu_r', center=0, ax=ax1, annot=True, vmin=-ax13_mm, vmax=ax13_mm,
cbar=False)
ax1.set_xlabel('')
ax1.set(xticklabels=[])
ax1.set_ylabel('WS-TI bin')
ax1.yaxis.set_tick_params(rotation=0)
ax1.tick_params(bottom=False)
sns.heatmap(spread_df2, linewidths=.5, cmap='viridis', ax=ax2, annot=True, vmin=ax24_min, vmax=ax24_max,
cbar=False)
ax2.set_xlabel('')
ax2.set(xticklabels=[])
ax2.set_ylabel('')
ax2.yaxis.set_tick_params(rotation=0)
ax2.tick_params(bottom=False)
sns.heatmap(average_df3, linewidths=.5, cmap='RdBu_r', center=0, ax=ax3, annot=True, vmin=-ax13_mm, vmax=ax13_mm,
cbar=False)
ax3.set_xlabel('Correction method')
ax3.yaxis.set_tick_params(rotation=0)
ax3.xaxis.set_tick_params(rotation=30)
ax3.set_ylabel('Inner-Outer Range')