-
Notifications
You must be signed in to change notification settings - Fork 4
/
FNDDS_base.py
3012 lines (2315 loc) · 125 KB
/
FNDDS_base.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
# coding=utf-8
import os
import pyodbc
import pandas as pd
import joblib
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
# import matplotlib._color_data as mcd
import matplotlib.patches as mpatch
import plotly.express as px
import plotly
import plotly.figure_factory as ff
import plotly.graph_objs as go
import plotly.io as pio
from sklearn.decomposition import PCA
from sklearn import preprocessing
from sklearn.manifold import TSNE
import umap
import json
import scipy.spatial as sp
from collections import OrderedDict
# import networkx as nx
import re
"""
Static variables and functions
"""
def hex_to_rgb(value):
value = value.lstrip('#')
lv = len(value)
return tuple(int(value[i:i + lv // 3], 16) for i in range(0, lv, lv // 3))
def rgb_to_hex(rgb):
rgb = (int(rgb[0]), int(rgb[1]), int(rgb[2]))
result = '#%02x%02x%02x' % tuple(rgb)
# print(result)
return result
NOVA_predictions_colors_dict = {
'NOVA 1': rgb_to_hex(np.array([0.4, 0.7607843137254902, 0.6470588235294118]) * 255),
'NOVA 2': rgb_to_hex(np.array([0.9882352941176471, 0.5529411764705883, 0.3843137254901961]) * 255),
'NOVA 3': rgb_to_hex(np.array([0.5529411764705883, 0.6274509803921569, 0.796078431372549]) * 255),
'NOVA 4': rgb_to_hex(np.array([0.9058823529411765, 0.5411764705882353, 0.7647058823529411]) * 255)
}
# NOVA_class_predictions_colors_dict = {
# 0: NOVA_predictions_colors_dict["NOVA 1"],
# 1: NOVA_predictions_colors_dict["NOVA 2"],
# 2: NOVA_predictions_colors_dict["NOVA 3"],
# 3: NOVA_predictions_colors_dict["NOVA 4"]
# }
NOVA_truth_colors_dict = {
"NOVA-Truth 0": rgb_to_hex(np.array([102, 102, 102])),
"NOVA-Truth 12": rgb_to_hex(np.array([204, 204, 204])),
"NOVA-Truth 1": rgb_to_hex(np.array([0.4, 0.7607843137254902, 0.6470588235294118]) * 255),
"NOVA-Truth 2": rgb_to_hex(
np.array([0.9882352941176471, 0.5529411764705883, 0.3843137254901961]) * 255),
"NOVA-Truth 3": rgb_to_hex(
np.array([0.5529411764705883, 0.6274509803921569, 0.796078431372549]) * 255),
"NOVA-Truth 4": rgb_to_hex(
np.array([0.9058823529411765, 0.5411764705882353, 0.7647058823529411]) * 255)
}
def load_tables_from_access_database(database_path, table_names):
if len([x for x in pyodbc.drivers() if x.startswith('Microsoft Access Driver')]) == 0:
raise Exception(
'You need to install "Access Data Engine" depending your office X32 or X64 it might become challenging to install it.')
path, file_name = os.path.split(database_path)
connecion_string = 'DRIVER={{Microsoft Access Driver (*.mdb, *.accdb)}};DBQ={0};'.format(database_path)
cnxn = pyodbc.connect(connecion_string)
df_tables = {}
for table_name in table_names:
path_pickle = "{}/{}.pkl".format(path, table_name)
if os.path.exists(path_pickle):
print("Loaded from pickle: {}".format(table_name))
df_tables[table_name] = pd.read_pickle(path_pickle)
pass
else:
query = "SELECT * FROM {}".format(table_name)
df_tables[table_name] = pd.read_sql(query, cnxn)
df_tables[table_name].to_pickle(path_pickle)
pass
print("Table: {} | Number of rows: {} | Source: {}".format(
table_name, len(df_tables[table_name]), database_path))
pass
cnxn.close()
return df_tables
base_path = "D:/Dropbox (CCNR)/Foodome Team Folder/Ravandi, Babak FDB/FoodomeDev/"
# It is better to maintain all classifiers in a single location
Classifiers_path = 'D:/Dropbox (CCNR)/Foodome Team Folder/Menichetti, Giulia/FoodProcessing/Datasets/Classifiers/'
'''I just dont want to reinstall cutreeHybrid!!'''
# from dynamicTreeCut import cutreeHybrid
from scipy.spatial.distance import pdist
# from sklearn.cluster import AgglomerativeClustering
import scipy.cluster.hierarchy as sch
def create_ingredient_description_for_FNDDS_2009_10(line_break_char="<br />"):
ingreds = pd.read_csv("data/NHANES_2009-10_Giulia_1.csv").rename(columns={"dr12ifdcd": "Food_code"})
# ingreds = ingreds[:200]
food_with_ingreds = ingreds.groupby("Food_code").agg({"SR_description_t": np.size}).reset_index().rename(
columns={"SR_description_t": "Count_ingreds"})
def make_ingred_desciption(row):
food_code = row["Food_code"]
ingred_desc = "Count: {}".format(row["Count_ingreds"])
for index, row in ingreds[ingreds["Food_code"] == food_code][["SR_description_t", "SR_nova_group"]].iterrows():
ingred_desc += "{} ☼ (NOVA{}) {}".format(line_break_char, row["SR_nova_group"], row["SR_description_t"])
return ingred_desc
pass
food_with_ingreds["ingreds"] = food_with_ingreds.apply(make_ingred_desciption, axis=1)
return food_with_ingreds
# region CLUSTERS REGION | PLOT CLUSTER NUTRIENTS
def training_corrections(correction_groups=None,
path_corrections="data/FNDDS/Correction Training Data FNDDS 2009_10.xls"):
#
training_corrections_df = OrderedDict()
# ORDER MATTERS THEY OVERRIDE EACH OTHER. For example, "Raw Foods" overrides "Giulia Cleaning"
training_corrections_df["Giulia Cleaning"] = pd.read_excel(path_corrections, sheet_name="Recom Giulia Cleaning")
training_corrections_df["Giulia Cleaning"] = training_corrections_df["Giulia Cleaning"][
training_corrections_df["Giulia Cleaning"]["NOVA Recommendation"].isin([1, 2, 3, 4])
]
training_corrections_df["Bean Soups"] = pd.read_excel(path_corrections, sheet_name="Recom Beans")
training_corrections_df["Home Recipe"] = pd.read_excel(path_corrections, sheet_name="Recome Home Recipe ")
training_corrections_df["Raw Foods"] = pd.read_excel(path_corrections, sheet_name="Recom Raws")
if correction_groups is not None:
for correction_group in list(training_corrections_df.keys()):
if correction_group not in correction_groups:
del training_corrections_df[correction_group]
pass
pass
pass
correction_groups = list(training_corrections_df.keys())
correction_groups.reverse()
for group_high_priority in correction_groups:
start_checking = False
for groups_to_be_overridden in correction_groups:
if start_checking:
# print("key_up: {} key_fix: {}".format(group_high_priority, groups_to_be_overridden))
training_corrections_df[groups_to_be_overridden] = training_corrections_df[groups_to_be_overridden][
~training_corrections_df[groups_to_be_overridden]["Food_code"].isin(
training_corrections_df[group_high_priority]["Food_code"]
)
]
pass
if groups_to_be_overridden == group_high_priority:
start_checking = True
pass
pass
pass
return training_corrections_df
plotly_trace_symbols = [
'square', 'diamond', 'cross', 'x', 'star', 'hexagram', 'bowtie', 'hourglass', 'triangle-left',
'triangle-right', 'hexagon', 'triangle-up', 'diamond-tall', 'octagon', 'pentagon', 'star-open', 'x-open',
'circle-open', 'circle-dot', 'circle-open-dot', 'square-open', 'square-dot',
'square-open-dot', 'diamond-open', 'diamond-dot', 'diamond-open-dot', 'cross-open', 'cross-dot',
'cross-open-dot', 'x-dot', 'x-open-dot', 'triangle-up-open', 'triangle-up-dot',
'triangle-up-open-dot', 'triangle-down', 'triangle-down-open', 'triangle-down-dot',
'triangle-down-open-dot', 'triangle-left-open', 'triangle-left-dot', 'triangle-left-open-dot',
'triangle-right-open', 'triangle-right-dot', 'triangle-right-open-dot', 'triangle-ne',
'triangle-ne-open', 'triangle-ne-dot', 'triangle-ne-open-dot', 'triangle-se', 'triangle-se-open',
'triangle-se-dot', 'triangle-se-open-dot', 'triangle-sw', 'triangle-sw-open', 'triangle-sw-dot',
'triangle-sw-open-dot', 'triangle-nw', 'triangle-nw-open', 'triangle-nw-dot', 'triangle-nw-open-dot',
'pentagon-open', 'pentagon-dot', 'pentagon-open-dot', 'hexagon-open', 'hexagon-dot', 'hexagon-open-dot',
'hexagon2-open', 'hexagon2-dot', 'hexagon2-open-dot', 'octagon-open', 'octagon-dot', 'octagon-open-dot',
'star-dot', 'star-open-dot', 'hexagram-open', 'hexagram-dot', 'hexagram-open-dot',
'star-triangle-up', 'star-triangle-up-open', 'star-triangle-up-dot', 'star-triangle-up-open-dot',
'star-triangle-down', 'star-triangle-down-open', 'star-triangle-down-dot', 'star-triangle-down-open-dot',
'star-square', 'star-square-open', 'star-square-dot', 'star-square-open-dot', 'star-diamond',
'star-diamond-open', 'star-diamond-dot', 'star-diamond-open-dot', 'diamond-tall-open',
'diamond-tall-dot', 'diamond-tall-open-dot', 'diamond-wide', 'diamond-wide-open', 'diamond-wide-dot',
'diamond-wide-open-dot', 'hourglass-open', 'bowtie-open', 'circle-cross', 'circle-cross-open',
'circle-x', 'circle-x-open', 'square-cross', 'square-cross-open', 'square-x', 'square-x-open',
'diamond-cross', 'diamond-cross-open', 'diamond-x', 'diamond-x-open', 'cross-thin', 'cross-thin-open',
'x-thin', 'x-thin-open', 'asterisk-open', 'hash-open', 'hash-dot', 'hash-open-dot', 'y-up', 'y-up-open',
'y-down', 'y-down-open', 'y-left', 'y-left-open', 'y-right', 'y-right-open', 'line-ew', 'line-ew-open',
'line-ns', 'line-ns-open', 'line-ne', 'line-ne-open', 'line-nw', 'line-nw-open', 'asterisk', 'hash',
]
def plot_predictions_PCA(predictions_df, column_legend, column_food_code, column_hover_description,
color_lut_legend, width=700, height=500, trace_color_column=None,
trace_food=None, trace_marker_opacity=0.7, marker_opacity=0.8,
title="", trace_marker_default_size=10, axis_range=None,
hover_label_args=None):
if trace_food is None:
trace_food = {}
def cluster_label(cluster_number):
return "{} ({})".format(predictions_df["cluster_labels"][cluster_number], cluster_number)
pass
# if limit_clusters_df_size is not None:
# predictions_df = predictions_df[0:limit_clusters_df_size]
predictions_df["color"] = predictions_df[column_legend].map(color_lut_legend)
# predictions_df["Cluster description"] = predictions_df.apply(cluster_description, axis=1)
lb = {"p1": r'$p_1$', "p4": r'$p_4$'}
Xpca = predictions_df.loc[:, 'p1': 'p4']
# Apply PCA
pca = PCA()
# pca.fit(Xpca_std)
pc = pca.fit_transform(Xpca)
predictions_df['PC 1'] = pc[:, 0]
predictions_df['PC 2'] = pc[:, 1]
fig = go.Figure()
traced_items_all_df = pd.DataFrame([], columns=[column_food_code])
if len(trace_food) > 0:
traced_items_all_df = []
for trace_key, trace_dict in trace_food.items():
traced_items_all_df.append(trace_dict["df"])
pass
traced_items_all_df = pd.concat(traced_items_all_df)
pass
for legend_value, color in color_lut_legend.items():
predictions_selected_df = predictions_df[
(predictions_df[column_legend] == legend_value) &
(~predictions_df[column_food_code].isin(traced_items_all_df[column_food_code]))
]
# cluster_df["marker size"] = 5
# cluster_df["opacity"] = 0.5
# cluster_df.loc[cluster_df.index.isin(onions.index), "marker size"] = 20
# cluster_df.loc[cluster_df.index.isin(onions.index), "opacity"] = 0.85
fig.add_trace(go.Scattergl(
x=predictions_selected_df['PC 1'], y=predictions_selected_df['PC 2'],
text=predictions_selected_df[column_hover_description],
mode='markers', marker=dict(color=color,
# size=cluster_df["marker size"],
# opacity=cluster_df["opacity"]
opacity=marker_opacity
),
name="{} [{}]".format(legend_value, len(predictions_selected_df)),
# legendgroup="all",
visible='legendonly'
))
pass
unused_trace_symbol = 0
# if True and trace_food_df is not None:
for trace_key, trace_dict in trace_food.items():
# trace_food_df = None, trace_food_mark_food_codes = None
traced_items_df = predictions_df[predictions_df[column_food_code].isin(trace_dict["df"][column_food_code])]
if "symbol" in trace_dict:
traced_items_df["Marker symbol"] = trace_dict["symbol"]
else:
traced_items_df["Marker symbol"] = plotly_trace_symbols[unused_trace_symbol]
unused_trace_symbol += 1
pass
if "size" in trace_dict:
traced_items_df["Marker size"] = trace_dict["size"]
else:
traced_items_df["Marker size"] = trace_marker_default_size
pass
# # cluster_df[cluster_df.index.isin(onions.index)]["marker size"] = 10
# # cluster_df.loc[cluster_df.index.isin(onions.index)] = 10
# # cluster_df.loc[cluster_df.index.isin(onions.index), "marker size"] = 10
# if trace_food_mark_food_codes is not None:
# for food_code, symbol in trace_food_mark_food_codes.items():
# food_code_index = traced_items_df[traced_items_df["Food_code"] == food_code].index[0]
#
# traced_items_df.loc[food_code_index, "Marker symbol"] = symbol
# traced_items_df.loc[food_code_index, "Marker size"] = trace_food_marker["given_food_codes_marker_size"]
# pass
# pass
if "custom_mark_food_codes" in trace_dict:
for food_code, symbol in trace_dict["custom_mark_food_codes"].items():
food_code_index = traced_items_df[traced_items_df["Food_code"] == food_code].index[0]
traced_items_df.loc[food_code_index, "Marker symbol"] = symbol
traced_items_df.loc[food_code_index, "Marker size"] = trace_food_marker["given_food_codes_marker_size"]
pass
pass
if trace_color_column is None:
trace_color = traced_items_df[column_legend].map(color_lut_legend)
pass
else:
trace_color = traced_items_df[trace_color_column]
fig.add_trace(go.Scattergl(
x=traced_items_df['PC 1'], y=traced_items_df['PC 2'],
# text=traced_items_df["Cluster description"],
text=traced_items_df[column_hover_description],
mode='markers',
marker=dict(
color=trace_color,
size=traced_items_df["Marker size"],
opacity=trace_marker_opacity,
symbol=traced_items_df["Marker symbol"],
line=dict(color='black', width=1.2)
),
name="{} [{}]".format(trace_dict["name"], len(traced_items_df)),
legendgroup=trace_key,
visible='legendonly'
))
if False:
fig.add_trace(go.Scattergl(
x=traced_items_df['PC 1'], y=traced_items_df['PC 2'], opacity=0.4,
mode='lines',
name="Trajectory", line=dict(color="Black"),
visible='legendonly'
# legendgroup=trace_food_name
))
pass
pass
if axis_range is None:
axis_range = {}
axis_range["y"] = dict(range=[predictions_df["PC 1"].min() - 0.1, predictions_df["PC 1"].max() + 0.1]),
axis_range["x"] = dict(range=[predictions_df["PC 2"].min() - 0.1, predictions_df["PC 2"].max() + 0.1])
print("Axis is automatic!")
pass
fig.update_layout(
hoverlabel=hover_label_args,
xaxis_autorange=False,
yaxis_autorange=False,
yaxis=axis_range["x"],
xaxis=axis_range["y"],
autosize=False,
title=title,
width=width,
height=height,
margin=dict(l=0, r=0, b=0, t=30, pad=4),
paper_bgcolor="#fff",
plot_bgcolor="#fff",
xaxis_title="PC 1",
yaxis_title="PC 2",
)
fig.update_xaxes(showline=True, linewidth=2, linecolor='black')
fig.update_yaxes(showline=True, linewidth=2, linecolor='black')
# fig.show()
return fig
# endregion
def load_NOVA_classified_food_for_FNDDS_2009_2010_with_prediction():
return pd.read_csv("data/RFFNDDSpredS_cleaned.csv")
def plot_number_ingredients_recipes(df_all, name, path_save):
df = df_all.groupby(["Is ingredient", "NOVA class"]).count()
vals_dict = df.to_dict("index")
ingred_recipe_vals = np.zeros((2, 4), dtype=int)
for is_ingredient in range(0, 2):
for NOVA_class in range(1, 5):
if (is_ingredient, NOVA_class) in vals_dict:
ingred_recipe_vals[is_ingredient, NOVA_class - 1] = vals_dict[(is_ingredient, NOVA_class)]["Code"]
pass
NOVA_groups = ['NOVA 1', 'NOVA 2', 'NOVA 3', 'NOVA 4']
fig = go.Figure(data=[
go.Bar(name='Ingredients', x=NOVA_groups, y=ingred_recipe_vals[1]),
go.Bar(name='Recipes', x=NOVA_groups, y=ingred_recipe_vals[0])
])
fig.update_layout({"title": name})
fig.write_image("{}/food_ingredients_NOVA_class_count_{}.png".format(path_save, name))
# fig.show()
return fig
def load_USDA_Nutrient_Retention_Factors():
print("Loading USDA_Nutrient_Retention_Factors Release 6 2007")
ret = pd.read_csv("data/USDA_Nutrient_Retention_Factors_Release_6_2007/NutrientRetention.csv")
ret["Retn_Factor"] = (
ret["Retn_Factor"].str.replace("Sep-75", "100")
.astype(float)
/ 100
)
# ret["Retn_Factor"] = ret["Retn_Factor"] / 100
return ret
def plot_NOVA_classes_dist(df, name, path_save):
df["class"] += 1
group_data = [df['p1'], df['p2'], df['p3'], df['p4']]
group_labels = ['p1', 'p2', 'p3', 'p4']
fig = ff.create_distplot(group_data, group_labels, bin_size=.05)
fig.update_layout({"title": "NOVA classes distribution for: " + name})
plotly.offline.plot(fig, filename="{}/NOVA_classes_distribution{}.html".format(path_save, name),
auto_open=True)
return fig
def plot_NOVA_class_histogram(df, NOVA_df, name, path_save):
presentation_columns = ["Code", "Is ingredient", "Label_full", "WWEIA Category description"]
df = (pd.merge(df[presentation_columns],
NOVA_df[["Code", "class", "p1", "p2", "p3", "p4"]],
left_on="Code", right_on="Code")
.rename(columns={"class": "NOVA class"})
)
df["NOVA class"] += 1
group_data = [df['p1'], df['p4']]
group_labels = ['p1', 'p4']
# colors = ['rgb(255, 255, 229)', 'rgb(102, 37, 6)'] # creamy-white
colors = ['rgb(68, 114, 196)', 'rgb(102, 37, 6)'] # blue
# colors = ['rgb(112, 173, 71)', 'rgb(102, 37, 6)'] # green
# Create distplot with custom bin_size
fig = ff.create_distplot(group_data, group_labels, bin_size=0.075, colors=colors)
fig.update_layout({"title": name})
fig.write_image("{}/food_ingredients_p1_p4_dist_{}.png".format(path_save, name))
# df.groupby("")
return df, fig
pass
def presentation_ingred(ingred_df, NOVA_df):
presentation_columns = ["Seq num", "Ingredient code", "Ingredient description", "Retention code",
"Ingredient weight", "Food ingredient weight normalized"]
ingred_df = (pd.merge(ingred_df[presentation_columns],
NOVA_df[["Code", "class", "p1", "p2", "p3", "p4"]],
left_on="Ingredient code", right_on="Code")
.drop(columns="Code")
.rename(columns={"class": "NOVA class"})
)
ingred_df["NOVA class"] += 1
return ingred_df
def hover_data_for_charts(df):
cols = []
for column in df.columns:
if column in ["Main food description", "pred class", "novaclass", "Food code", "category_groupped",
"WWEIA Category description", "Count Ingredients", "Ingredient description", "p2", "p3"]:
cols.append(column)
pass
pass
return cols
from pandas import ExcelWriter
def save_xls(dfs_dict, xls_path, open=False):
with ExcelWriter(xls_path) as writer:
for df_name, df in dfs_dict.items():
df.to_excel(writer, df_name)
writer.save()
if open is True:
os.system('start EXCEL.EXE "{}"'.format(os.path.abspath(xls_path)))
def save_excel(df, relative_path, open, index):
df.to_excel(relative_path, index=index)
if open is True:
os.system('start EXCEL.EXE "{}"'.format(os.path.abspath(relative_path)))
pass
# region GIULIA CLASSIFIER COLUMN NAMES and ORDERS
FNDDS_table_salt = {"Ingredient code": 2047, "Description": "Salt, table"}
NOVA_color_discrete_map = {'Unprocessed': 'green', 'processed culinary': 'blue', 'processed': 'yellow',
'ultra processed': 'red'}
giulia_classifier_12_panel_cols = clf_cols_order = [
'Protein', 'Total Fat', 'Carbohydrate', 'Sugars, total', 'Fiber, total dietary', 'Calcium', 'Iron',
'Sodium', 'Vitamin C', 'Cholesterol', 'Fatty acids, total saturated', 'Total Vitamin A']
giulia_classifier_99_panel_cols = [ # Used to be giulia_classifier_order_cols
"Protein", "Total Fat", "Carbohydrate", "Alcohol", "Water", "Caffeine", "Theobromine",
"Sugars, total", "Fiber, total dietary", "Calcium", "Iron", "Magnesium", "Phosphorus", "Potassium",
"Sodium", "Zinc", "Copper", "Selenium",
"Retinol", "Carotene, beta", "Carotene, alpha", "Vitamin E (alpha-tocopherol)",
"Vitamin D (D2 + D3)", # not in 58
"Cryptoxanthin, beta", "Lycopene", "Lutein + zeaxanthin", "Vitamin C", "Thiamin", "Riboflavin",
"Niacin", "Vitamin B-6", "Folate, total", "Vitamin B-12",
"Choline, total", # not in 58
"Vitamin K (phylloquinone)", "Folic acid", "Folate, food",
"Vitamin E, added", # not in 58
"Vitamin B-12, added", # not in 58
"Cholesterol", "Fatty acids, total saturated", "4:0", "6:0", "8:0", "10:0", "12:0",
"14:0", "16:0", "18:0", "18:1", "18:2", "18:3", "20:4", "22:6 n-3", "16:1", "18:4", "20:1", "20:5 n-3",
"22:1", "22:5 n-3", "Fatty acids, total monounsaturated", "Fatty acids, total polyunsaturated",
# Flavonoids nutrients only available in special database
"Daidzein", "Genistein", "Glycitein", "Cyanidin", "Petunidin", "Delphinidin", "Malvidin",
"Pelargonidin", "Peonidin", "(+)-Catechin", "(-)-Epigallocatechin", "(-)-Epicatechin", "(-)-Epicatechin 3-gallate",
"(-)-Epigallocatechin 3-gallate", "Theaflavin", "Thearubigins", "Eriodictyol",
"Hesperetin", "Naringenin", "Apigenin", "Luteolin", "Isorhamnetin", "Kaempferol", "Myricetin", "Quercetin",
"Theaflavin-3,3\'-digallate", "Theaflavin-3\'-gallate", "Theaflavin-3-gallate",
"(+)-Gallocatechin", "Total flavonoids", "Total anthocyanidins", "Total catechins (monomeric flavan-3-ols only)",
"Total flavan-3-ols", "Total flavanones", "Total flavones", "Total flavonols", "Total isoflavones"]
giulia_classifier_62_panel_cols = giulia_classifier_99_panel_cols[:62]
# giulia_classifier_order_cols_for_FNDDS_2015_2016 = giulia_classifier_order_cols + ['Folate, DFE', 'Vitamin A, RAE']
giulia_classifier_cols_nutrient_facts = [
'Protein', 'Total Fat', 'Carbohydrate', 'Sugars, total', 'Fiber, total dietary', 'Calcium', 'Iron',
'Sodium', 'Vitamin C', 'Cholesterol', 'Fatty acids, total saturated', 'Total Vitamin A']
giulia_classifier_cols_removed_on_FNDDS_2015_2016 = [
'Daidzein', 'Genistein', 'Glycitein', 'Cyanidin', 'Petunidin', 'Delphinidin', 'Malvidin', 'Pelargonidin',
'Peonidin', '(+)-Catechin', '(-)-Epigallocatechin', '(-)-Epicatechin', '(-)-Epicatechin 3-gallate',
'(-)-Epigallocatechin 3-gallate', 'Theaflavin', 'Thearubigins', 'Eriodictyol', 'Hesperetin', 'Naringenin',
'Apigenin', 'Luteolin', 'Isorhamnetin', 'Kaempferol', 'Myricetin', 'Quercetin', "Theaflavin-3,3'-digallate",
"Theaflavin-3'-gallate", 'Theaflavin-3-gallate', '(+)-Gallocatechin', 'Total flavonoids', 'Total anthocyanidins',
'Total catechins (monomeric flavan-3-ols only)', 'Total flavan-3-ols', 'Total flavanones', 'Total flavones',
'Total flavonols', 'Total isoflavones']
giulia_classifier_order_cols_2015_2016 = [
'Protein', 'Total Fat', 'Carbohydrate', 'Alcohol', 'Water', 'Caffeine',
'Theobromine', 'Sugars, total', 'Fiber, total dietary', 'Calcium',
'Iron', 'Magnesium', 'Phosphorus', 'Potassium', 'Sodium', 'Zinc',
'Copper', 'Selenium', 'Retinol', 'Carotene, beta', 'Carotene, alpha',
'Vitamin E (alpha-tocopherol)', 'Vitamin D (D2 + D3)',
'Cryptoxanthin, beta', 'Lycopene', 'Lutein + zeaxanthin', 'Vitamin C',
'Thiamin', 'Riboflavin', 'Niacin', 'Vitamin B-6', 'Folate, total',
'Vitamin B-12', 'Choline, total', 'Vitamin K (phylloquinone)',
'Folic acid', 'Folate, food', 'Vitamin E, added', 'Vitamin B-12, added',
'Cholesterol', 'Fatty acids, total saturated', '4:0', '6:0', '8:0', '10:0', '12:0', '14:0', '16:0', '18:0', '18:1',
'18:2', '18:3',
'20:4', '22:6 n-3', '16:1', '18:4', '20:1', '20:5 n-3', '22:1', '22:5 n-3', 'Fatty acids, total monounsaturated',
'Fatty acids, total polyunsaturated'
]
giulia_calc_vitamin_A_by_sum_of_columns = [
(319, 'Retinol'), (321, 'Carotene, beta'), (322, 'Carotene, alpha'), (334, 'Cryptoxanthin, beta')
]
# giulia_classifier_99_nuts_codes = {
# 203: 'Protein', 204: 'Total Fat', 205: 'Carbohydrate', 221: 'Alcohol', 255: 'Water', 262: 'Caffeine',
# 263: 'Theobromine', 269: 'Sugars, total', 291: 'Fiber, total dietary', 301: 'Calcium', 303: 'Iron',
# 304: 'Magnesium', 305: 'Phosphorus', 306: 'Potassium', 307: 'Sodium', 309: 'Zinc', 312: 'Copper',
# 317: 'Selenium', 319: 'Retinol', 321: 'Carotene, beta', 322: 'Carotene, alpha', 323: 'Vitamin E (alpha-tocopherol)',
# 328: 'Vitamin D (D2 + D3)', 334: 'Cryptoxanthin, beta', 337: 'Lycopene', 338: 'Lutein + zeaxanthin',
# 401: 'Vitamin C', 404: 'Thiamin', 405: 'Riboflavin', 406: 'Niacin', 415: 'Vitamin B-6', 417: 'Folate, total',
# 418: 'Vitamin B-12', 421: 'Choline, total', 430: 'Vitamin K (phylloquinone)', 431: 'Folic acid',
# 432: 'Folate, food', 573: 'Vitamin E, added', 578: 'Vitamin B-12, added', 601: 'Cholesterol',
# 606: 'Fatty acids, total saturated', 607: '4:0', 608: '6:0', 609: '8:0', 610: '10:0', 611: '12:0',
# 612: '14:0', 613: '16:0', 614: '18:0', 617: '18:1', 618: '18:2', 619: '18:3', 620: '20:4',
# 621: '22:6 n-3', 626: '16:1', 627: '18:4', 628: '20:1', 629: '20:5 n-3', 630: '22:1', 631: '22:5 n-3',
# 645: 'Fatty acids, total monounsaturated', 646: 'Fatty acids, total polyunsaturated', 7000: 'Total flavonoids',
# 731: 'Cyanidin', 740: 'Petunidin', 741: 'Delphinidin', 742: 'Malvidin', 743: 'Pelargonidin', 745: 'Peonidin',
# 7100: 'Total anthocyanidins', 749: '(+)-Catechin', 750: '(-)-Epigallocatechin', 751: '(-)-Epicatechin',
# 752: '(-)-Epicatechin 3-gallate', 753: '(-)-Epigallocatechin 3-gallate', 755: 'Theaflavin', 756: 'Thearubigins',
# 791: "Theaflavin-3,3'-digallate", 792: "Theaflavin-3'-gallate", 793: 'Theaflavin-3-gallate',
# 794: '(+)-Gallocatechin', 7200: 'Total catechins (monomeric flavan-3-ols only)', 7300: 'Total flavan-3-ols',
# 758: 'Eriodictyol', 759: 'Hesperetin', 762: 'Naringenin', 7400: 'Total flavanones', 770: 'Apigenin',
# 773: 'Luteolin', 7500: 'Total flavones', 785: 'Isorhamnetin', 786: 'Kaempferol', 788: 'Myricetin',
# 789: 'Quercetin', 7600: 'Total flavonols', 710: 'Daidzein', 711: 'Genistein', 712: 'Glycitein',
# 7700: 'Total isoflavones'}
# giulia_classifier_99_nuts_codes_df = pd.DataFrame(
# zip(giulia_classifier_99_nuts_codes.keys(), giulia_classifier_99_nuts_codes.values()),
# columns=["Nutrient code", "Nutrient description"]
# )
giulia_classifier_99_nuts_codes_df = pd.DataFrame(
[
[203, 'Protein', 'g'], [204, 'Total Fat', 'g'], [205, 'Carbohydrate', 'g'], [221, 'Alcohol', 'g'],
[255, 'Water', 'g'], [262, 'Caffeine', 'mg'], [263, 'Theobromine', 'mg'], [269, 'Sugars, total', 'g'],
[291, 'Fiber, total dietary', 'g'], [301, 'Calcium', 'mg'], [303, 'Iron', 'mg'], [304, 'Magnesium', 'mg'],
[305, 'Phosphorus', 'mg'], [306, 'Potassium', 'mg'], [307, 'Sodium', 'mg'], [309, 'Zinc', 'mg'],
[312, 'Copper', 'mg'], [317, 'Selenium', 'mcg'], [319, 'Retinol', 'mcg'], [321, 'Carotene, beta', 'mcg'],
[322, 'Carotene, alpha', 'mcg'], [323, 'Vitamin E (alpha-tocopherol)', 'mg'],
[328, 'Vitamin D (D2 + D3)', 'mcg'],
[334, 'Cryptoxanthin, beta', 'mcg'], [337, 'Lycopene', 'mcg'], [338, 'Lutein + zeaxanthin', 'mcg'],
[401, 'Vitamin C', 'mg'], [404, 'Thiamin', 'mg'], [405, 'Riboflavin', 'mg'], [406, 'Niacin', 'mg'],
[415, 'Vitamin B-6', 'mg'], [417, 'Folate, total', 'mcg'], [418, 'Vitamin B-12', 'mcg'],
[421, 'Choline, total', 'mg'], [430, 'Vitamin K (phylloquinone)', 'mcg'], [431, 'Folic acid', 'mcg'],
[432, 'Folate, food', 'mcg'], [573, 'Vitamin E, added', 'mg'], [578, 'Vitamin B-12, added', 'mcg'],
[601, 'Cholesterol', 'mg'], [606, 'Fatty acids, total saturated', 'g'], [607, '4:0', 'g'], [608, '6:0', 'g'],
[609, '8:0', 'g'], [610, '10:0', 'g'], [611, '12:0', 'g'], [612, '14:0', 'g'], [613, '16:0', 'g'],
[614, '18:0', 'g'], [617, '18:1', 'g'], [618, '18:2', 'g'], [619, '18:3', 'g'], [620, '20:4', 'g'],
[621, '22:6 n-3', 'g'], [626, '16:1', 'g'], [627, '18:4', 'g'], [628, '20:1', 'g'], [629, '20:5 n-3', 'g'],
[630, '22:1', 'g'], [631, '22:5 n-3', 'g'], [645, 'Fatty acids, total monounsaturated', 'g'],
[646, 'Fatty acids, total polyunsaturated', 'g'],
# Dine 62 nuts
[7000, 'Total flavonoids', 'mg'], [731, 'Cyanidin', 'mg'],
[740, 'Petunidin', 'mg'], [741, 'Delphinidin', 'mg'], [742, 'Malvidin', 'mg'], [743, 'Pelargonidin', 'mg'],
[745, 'Peonidin', 'mg'], [7100, 'Total anthocyanidins', 'mg'], [749, '(+)-Catechin', 'mg'],
[750, '(-)-Epigallocatechin', 'mg'], [751, '(-)-Epicatechin', 'mg'], [752, '(-)-Epicatechin 3-gallate', 'mg'],
[753, '(-)-Epigallocatechin 3-gallate', 'mg'], [755, 'Theaflavin', 'mg'], [756, 'Thearubigins', 'mg'],
[791, "Theaflavin-3,3'-digallate", 'mg'], [792, "Theaflavin-3'-gallate", 'mg'],
[793, 'Theaflavin-3-gallate', 'mg'], [794, '(+)-Gallocatechin', 'mg'],
[7200, 'Total catechins (monomeric flavan-3-ols only)', 'mg'], [7300, 'Total flavan-3-ols', 'mg'],
[758, 'Eriodictyol', 'mg'], [759, 'Hesperetin', 'mg'], [762, 'Naringenin', 'mg'],
[7400, 'Total flavanones', 'mg'], [770, 'Apigenin', 'mg'], [773, 'Luteolin', 'mg'],
[7500, 'Total flavones', 'mg'], [785, 'Isorhamnetin', 'mg'], [786, 'Kaempferol', 'mg'],
[788, 'Myricetin', 'mg'], [789, 'Quercetin', 'mg'], [7600, 'Total flavonols', 'mg'], [710, 'Daidzein', 'mg'],
[711, 'Genistein', 'mg'], [712, 'Glycitein', 'mg'], [7700, 'Total isoflavones', 'mg']
],
columns=["Nutrient code", "Nutrient description", "unit"]
)
flavones_nutrient_desc = pd.DataFrame(
[
(7000, 'Total flavonoids', 'mg'), (731, 'Cyanidin', 'mg'), (740, 'Petunidin', 'mg'), (741, 'Delphinidin', 'mg'),
(742, 'Malvidin', 'mg'), (743, 'Pelargonidin', 'mg'), (745, 'Peonidin', 'mg'),
(7100, 'Total anthocyanidins', 'mg'), (749, '(+)-Catechin', 'mg'), (750, '(-)-Epigallocatechin', 'mg'),
(751, '(-)-Epicatechin', 'mg'), (752, '(-)-Epicatechin 3-gallate', 'mg'),
(753, '(-)-Epigallocatechin 3-gallate', 'mg'), (755, 'Theaflavin', 'mg'), (756, 'Thearubigins', 'mg'),
(791, "Theaflavin-3,3'-digallate", 'mg'), (792, "Theaflavin-3'-gallate", 'mg'),
(793, 'Theaflavin-3-gallate', 'mg'), (794, '(+)-Gallocatechin', 'mg'),
(7200, 'Total catechins (monomeric flavan-3-ols only)', 'mg'), (7300, 'Total flavan-3-ols', 'mg'),
(758, 'Eriodictyol', 'mg'), (759, 'Hesperetin', 'mg'), (762, 'Naringenin', 'mg'),
(7400, 'Total flavanones', 'mg'), (770, 'Apigenin', 'mg'), (773, 'Luteolin', 'mg'),
(7500, 'Total flavones', 'mg'), (785, 'Isorhamnetin', 'mg'), (786, 'Kaempferol', 'mg'),
(788, 'Myricetin', 'mg'), (789, 'Quercetin', 'mg'), (7600, 'Total flavonols', 'mg'),
(710, 'Daidzein', 'mg'), (711, 'Genistein', 'mg'), (712, 'Glycitein', 'mg'), (7700, 'Total isoflavones', 'mg')
],
columns=["Nutrient code", "Nutrient description", "unit"]
)
# endregion
"""
PCA visualization
"""
def NOVA_PCA_analysis(n_components, df, hover_data, method, col_pred_class="pred class",
features=["p1", "p2", "p3", "p4"], width=1000, height=650, standard_scaler_fit=True):
x = df.loc[:, features].values
if standard_scaler_fit is True:
x = preprocessing.StandardScaler().fit_transform(x)
pass
method = method.lower()
if method == "umap":
reducer = umap.UMAP(n_components=n_components)
principalComponents = reducer.fit_transform(x)
principalDf = pd.DataFrame(data=principalComponents
, columns=['PC{}'.format(i) for i in range(1, n_components + 1)])
pass
elif method == "pca":
pca = PCA(n_components=n_components)
principalComponents = pca.fit_transform(x)
principalDf = pd.DataFrame(data=principalComponents
, columns=['PC{}'.format(i) for i in range(1, n_components + 1)])
elif method == "tsne":
tsne = TSNE(n_components=n_components, verbose=0, perplexity=60, n_iter=800)
# tsne = TSNE(n_components=n_components)
principalComponents = tsne.fit_transform(x)
principalDf = pd.DataFrame(data=principalComponents
, columns=['PC{}'.format(i) for i in range(1, n_components + 1)])
else:
raise Exception("Method {} is not handeled.".format(method))
# hover_data = list(set(hover_data) - set(features))
finalDf = pd.concat([principalDf, df[list(set(hover_data) - set(features)) + features]], axis=1)
finalDf['cat'] = df[col_pred_class].map(
# {1: 'Unprocessed', 2: 'processed culinary', 3: 'processed', 4: 'ultra processed'}
{1: 'NOVA 1', 2: 'NOVA 2', 3: 'NOVA 3', 4: 'NOVA 4'}
)
# NOVA_color_discrete_map = {'Unprocessed': 'green', 'processed culinary': 'blue', 'processed': 'yellow',
# 'ultra processed': 'red'}
color_discrete_map = {'NOVA 1': 'green', 'NOVA 2': 'blue', 'NOVA 3': 'yellow', 'NOVA 4': 'red'}
if n_components == 1:
# hover_data=hover_data + features
# return finalDf
fig = px.scatter(finalDf, x="PC1", y="PC1", hover_data=hover_data, width=width, height=height,
color="cat", color_discrete_map=color_discrete_map)
elif n_components == 2:
# hover_data=hover_data + features
fig = px.scatter(finalDf, x="PC1", y="PC2", hover_data=hover_data, width=width, height=height,
color="cat", color_discrete_map=color_discrete_map)
elif n_components == 3:
# hover_data[:-2] + features
fig = px.scatter_3d(finalDf, x="PC1", y="PC2", z="PC3", hover_data=hover_data,
color="cat", color_discrete_map=color_discrete_map)
pass
else:
raise Exception("n_components can be only 1, 2 or 3")
fig.update_traces(mode='markers', marker_line_width=0.0, marker_size=4, marker_opacity=0.5)
# fig.update_layout(title="PCA | Num Food-Items: {}".format(len(df)))
# # fig.update_xaxes(range=[0.0, 1.0])
# # fig.update_yaxes(range=[0.0, 1.0])
# fig.show()
return fig
def convert_to_mg(value, unit):
if unit == "g":
r = value * 1000
elif unit == "mg":
r = value
elif unit == "mcg":
r = value * 0.001
pass
else:
return value
pass
return r
def convert_to_g_supported_units():
return ["iu", "fl oz", "ml", "liter", "g", "gram", "lb", "fluid ounce",
"lbs", "oz"]
pass
def scale_to_100_g():
raise Exception("NOT COMPLETE")
pass
def convert_to_g_accepted_units():
return ["g", "gram", "gm", "grams"] + ["mg"] + ['µg', "mcg", "mcg_rae", "mcg_dfe"] + \
["lb", "lbs"] + ["oz"] + ["fl oz", "fl", "fluid ounce"] + ["ml"] + ["liter"] # + #["iu"] + ["re"]
def convert_to_g(value, unit, nutrient):
# https: // dietarysupplementdatabase.usda.nih.gov / Conversions.php
if value is None:
raise Exception("X")
nutrient = nutrient.lower()
r = None
unit = unit.lower()
if unit in ["g", "gram", "grams", "gm"]:
r = value
elif unit == "mg":
r = value * 0.001
elif unit in ['µg', "mcg", "mcg_rae", "mcg_dfe", 'µg_dfe']:
r = value * 0.000001
elif unit in ["lb", "lbs"]:
r = value * 453.59237
elif unit in ["oz"]:
r = value * 28.3495
elif unit == "iu":
if nutrient in ['vitamin e', 'vitamine']:
# 1 IU = 0.67 mg for d-alpha-tocopherol (natural)
# 1 IU = 0.9 mg for dl-alpha-tocopherol (synthetic)
r = value * 0.67 * 0.001
pass
elif unit in ["re", 'µg_rae']:
# https://www.dietobio.com/vegetarisme/en/vit_a.html
"""WARNING!!! DOUBLE CHECK 'RE' UNIT CONVERSION TO GRAMS!!!!"""
if nutrient in ['carotene']:
r = value * 0.000001 * 6
pass
if nutrient in ['vitamin a', 'vitamina', 'vitamin a, rae']:
r = value * 0.000001
pass
elif any(x in unit for x in ["calories", "kcal", "cals", 'kj']):
r = value
elif any(x in unit for x in ["fl oz", "fl", "fluid ounce"]):
r = value * 29.57352956
elif unit == "ml":
r = value * 1
elif unit == "liter":
r = value * 1000
else:
raise Exception("Can not convert unit '{}' for {}".format(unit, nutrient))
return r
def convert_nutrient_unit_to_grams(df, nutrient_values_column):
"""
Convert all units to MG
"""
# if "Nutrient value grams" not in df.columns:
#
df.insert(
list(df.columns).index(nutrient_values_column),
nutrient_values_column + " grams",
df.apply(
lambda row: convert_to_g(row[nutrient_values_column], row["Unit"], row['Nutrient description']), axis=1),
True
)
pass
def draw_color_text_box(overlap, figsize, title, title_font_size=16, title_font_y=1.1, fontfamily=None):
"""
Source: https://matplotlib.org/tutorials/colors/colors.html
:param overlap:
:return:
"""
# import matplotlib._color_data as mcd
# import matplotlib.patches as mpatch
# overlap = {name for name in mcd.CSS4_COLORS
# if "xkcd:" + name in mcd.XKCD_COLORS}
# overlap = {'p1': 'g', 'p2': 'b', 'p3': 'y', 'p4': 'r'}
# fig = plt.figure(figsize=[4.8, 2])
# figsize = [2, 2]
fig = plt.figure(figsize=figsize)
fig.suptitle(title, fontsize=title_font_size, y=title_font_y, fontfamily=fontfamily)
ax = fig.add_axes([0, 0, 1, 1])
# for j, n in enumerate(sorted(overlap.keys(), reverse=True)):
for j, key in enumerate(reversed(list(overlap.keys()))):
weight = None
# cn = mcd.CSS4_COLORS[n]
cn = overlap[key]
# xkcd = mcd.XKCD_COLORS["xkcd:" + n].upper()
# xkcd = n
# if cn == xkcd:
# weight = 'bold'
r1 = mpatch.Rectangle((0, j), 1, 1, color=cn)
# r2 = mpatch.Rectangle((1, j), 1, 1, color=xkcd)
txt = ax.text(figsize[0] / 2, j + 0.5, ' ' + key, va='center', fontsize=10,
weight=weight, fontfamily=fontfamily)
ax.add_patch(r1)
# ax.add_patch(r2)
ax.axhline(j, color='k')
pass
# ax.text(.5, j + 1.5, 'Color', ha='center', va='center')
# ax.text(1.5, j + 1.5, 'xkcd', ha='center', va='center')
ax.set_xlim(0, figsize[0])
ax.set_ylim(0, j + (figsize[1] / 2) + 0.2)
ax.axis('off')
return fig
pass
def predict_NOVA_class(clf, clf_cols_order, nutrients_df, pos_insert_preds, convert_to_log, new_cols_name_prefix=""):
"""
Run classifier_DEL_ME on a given dataframe
:param clf:
:param clf_cols_order:
:param nutrients_df:
:param pos_insert_preds:
:param convert_to_log:
:return:
"""
if new_cols_name_prefix != "":
new_cols_name_prefix = new_cols_name_prefix.strip() + " "
nutrients_df_nuts = nutrients_df[clf_cols_order]
if convert_to_log is True:
nutrients_df_nuts = np.log(nutrients_df_nuts)
nutrients_df_nuts = nutrients_df_nuts.fillna(-20)
nutrients_df_nuts = nutrients_df_nuts.replace([np.inf, -np.inf], -20)
pass
predicted_class = clf.predict(nutrients_df_nuts)
predict_prob = clf.predict_proba(nutrients_df_nuts)
nutrients_df.insert(pos_insert_preds, new_cols_name_prefix + "class", predicted_class, True)
nutrients_df.insert(pos_insert_preds + 1, new_cols_name_prefix + "p4", predict_prob[:, 3], True)
nutrients_df.insert(pos_insert_preds + 1, new_cols_name_prefix + "p3", predict_prob[:, 2], True)
nutrients_df.insert(pos_insert_preds + 1, new_cols_name_prefix + "p2", predict_prob[:, 1], True)
nutrients_df.insert(pos_insert_preds + 1, new_cols_name_prefix + "p1", predict_prob[:, 0], True)
return nutrients_df
def get_giulia_classifier(name, only_clf_cols_order=False):
nut_panels = {
'58P': [
'Protein', 'Total Fat', 'Carbohydrate', 'Alcohol', 'Water',
'Caffeine', 'Theobromine', 'Sugars, total', 'Fiber, total dietary',
'Calcium', 'Iron', 'Magnesium', 'Phosphorus', 'Potassium',
'Sodium', 'Zinc', 'Copper', 'Selenium', 'Retinol',
'Carotene, beta', 'Carotene, alpha',
'Vitamin E (alpha-tocopherol)', 'Cryptoxanthin, beta', 'Lycopene',
'Lutein + zeaxanthin', 'Vitamin C', 'Thiamin', 'Riboflavin',
'Niacin', 'Vitamin B-6', 'Folate, total', 'Vitamin B-12',
'Vitamin K (phylloquinone)', 'Folic acid', 'Folate, food',
'Cholesterol', 'Fatty acids, total saturated', '4:0', '6:0', '8:0',
'10:0', '12:0', '14:0', '16:0', '18:0', '18:1', '18:2', '18:3',
'20:4', '22:6 n-3', '16:1', '18:4', '20:1', '20:5 n-3', '22:1',
'22:5 n-3', 'Fatty acids, total monounsaturated',
'Fatty acids, total polyunsaturated'
]
}
clf_cols_order = None
clf = None
if name == "nutritionfacts_with_salt":
if only_clf_cols_order is False: