-
Notifications
You must be signed in to change notification settings - Fork 4
/
functions_and_classes.py
2305 lines (1583 loc) · 87.2 KB
/
functions_and_classes.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 copy
import datetime
import numpy as np
import os
import pandas as pd
import psycopg2
import statsmodels.api as sm
from dotenv import load_dotenv, find_dotenv
from scipy import stats
from sklearn.linear_model import LinearRegression
from sklearn.metrics import mean_squared_error
from statsmodels.tsa.arima_model import ARIMA
from statsmodels.tsa.stattools import adfuller
from sqlalchemy import create_engine
load_dotenv()
class Clean_and_classify_class:
def __init__(self):
self.description = ''' This class groups the main functions that are used to clean,
prepare the data, build the ALPS bands and also label the
product prices.'''
def set_columns(self,data):
''' Builds a dataframe with the raw data comming from the db. '''
data = pd.DataFrame(data)
data = data.rename(columns={0:'date_price',1:'unit_scale',2:'observed_price'})
return data
def basic_cleanning(self,df):
'''
Removes duplicates in dates column.
Verify unique unit scale.
Try to correct typos.
Returns the metric and the dataframe with the basic cleaned data.
'''
cfd = df.copy()
# Set dates into date format.
cfd['date_price'] = pd.to_datetime(cfd['date_price'])
# Remove duplicates in dates column.
cfd = cfd.sort_values(by=['date_price',cfd.columns[-1]])
drop_index = list(cfd[cfd.duplicated(['date_price'], keep='first')].index)
cfd = cfd.drop(labels=drop_index, axis=0).reset_index(drop=True)
# Drop values with prices zero.
drop_index = list(cfd[cfd.iloc[:,-1] == 0].index)
cfd = cfd.drop(labels=drop_index, axis=0).reset_index(drop=True)
if cfd.empty:
return None, cfd
else:
# Consider the mode of unit scale as the one.
metric = stats.mode(cfd.iloc[:,1])[0][0]
discording_scale = list(cfd[cfd['unit_scale'] != metric].index)
if discording_scale:
cfd = cfd.drop(labels=discording_scale, axis=0).reset_index(drop=True)
# Try to correct typos that seems to be missing a decimal point.
if 9 <= cfd.describe().T['max'].values[0] / cfd.describe().T['75%'].values[0] <= 11:
Q99 = cfd.quantile(0.99).values[0]
selected_indices = list(cfd[cfd.iloc[:,-1] > Q99].index)
for i in selected_indices:
cfd.iloc[i,-1] = cfd.iloc[i,-1] / 10
# Drop typos we can't solve.
drop_index = cfd[(cfd.iloc[:,-1].rolling(window=1, min_periods= 1).max() / cfd.iloc[:,-1].rolling(window=30, min_periods= 10).quantile(0.75) ) > 5].index
cfd = cfd.drop(labels=drop_index, axis=0).reset_index(drop=True)
# TODO the following not sure if they apport meaningful results, because the two lines before this are very aggresive.
Q01 = cfd.iloc[:,-1].quantile(.1)
if cfd.describe().T['min'].values[0] < Q01:
drop_index = list(cfd[cfd.iloc[:,-1] < Q01].index)
cfd = cfd.drop(labels=drop_index, axis=0).reset_index(drop=True)
# Drop outliers.
if (cfd.describe().T['max'].values[0]) > (cfd.describe().T['75%'].values[0] + cfd.describe().T['std'].values[0]):
# First round
z = np.abs(stats.zscore(cfd.iloc[:,-1], nan_policy='omit'))
drop_index = list(np.where(z>4)[0])
cfd = cfd.drop(labels=drop_index, axis=0).reset_index(drop=True)
if (cfd.describe().T['max'].values[0]) > (cfd.describe().T['75%'].values[0] + cfd.describe().T['std'].values[0]):
# Second round.
z = np.abs(stats.zscore(cfd.iloc[:,-1], nan_policy='omit'))
drop_index = list(np.where(z>5)[0])
cfd = cfd.drop(labels=drop_index, axis=0).reset_index(drop=True)
return metric, cfd
def limit_2019_and_later(self,df):
'''
Limit the info to the 2020 or later and assigns its month, so the price could be compared with the bands.
'''
df = df[df['date_price'] > pd.to_datetime('2018-12-31')]
df['date_price'] = df['date_price'].astype('datetime64')
df['month'] = [str(df.iloc[i,0])[:8] + '01' for i in range(len(df))]
df = df.reset_index(drop=True)
return df
def prepare_data_to_ALPS(self,df):
'''
Make a dataframe with the last Sunday before the dates of the input dataframe, and the saturday of the last week in within the dates.
Then Merge both dataframes to have one with all the possible weeks within the dates of the original dataframe.
Interpolate the missing values.
'''
cfd = df.copy()
# Turn the dataframe into a calendar.
if cfd['date_price'].min().day == 1:
start = cfd['date_price'].min()
else:
start = cfd['date_price'].min() - datetime.timedelta(days=cfd['date_price'].min().day + 1)
if cfd['date_price'].max().day >= 28:
end = cfd['date_price'].max()
else:
end = cfd['date_price'].max() - datetime.timedelta(days=cfd['date_price'].max().day +1)
dummy = pd.DataFrame()
dummy['date_price'] = pd.date_range(start=start, end=end)
dummy = dummy.set_index('date_price')
cfd = cfd.set_index('date_price')
cfd = dummy.merge(cfd,how='outer',left_index=True, right_index=True)
del dummy
cfd['max_price_30days'] = cfd.iloc[:,-1].rolling(window=30,min_periods=1).max()
cfd['max_price_30days'] = cfd['max_price_30days'].shift(-1)
cfd = cfd[cfd.index.day == 1]
cfd = cfd[['max_price_30days']].interpolate()
cfd = cfd.dropna()
return cfd
def inmediate_forecast_ALPS_based(self,df):
'''
Takes the prices and the prediction for the next month, for the last
two years, using a basic linear regression, taking for variables,
the month in which the price was taken.
'''
forecasted_prices = []
basesetyear = df.index.max().year - 2
stop_0 = datetime.date(year=basesetyear,month=12,day=31)
baseset = df.iloc[:len(df.loc[:stop_0]),:].copy()
# For all the past months:
for i in range(len(df)-len(baseset)):
workset = df.iloc[:len(df.loc[:stop_0]) + i,:].copy()
# What month are we?
workset['month'] = workset.index.month
# Build dummy variables for the months.
dummies_df = pd.get_dummies(workset['month'])
dummies_df = dummies_df.T.reindex(range(1,13)).T.fillna(0)
workset = workset.join(dummies_df)
workset = workset.drop(labels=['month'], axis=1)
features = workset.columns[1:]
target = workset.columns[0]
X = workset[features]
y = workset[target]
reg = LinearRegression()
reg = reg.fit(X,y)
next_month = df.iloc[len(df.loc[:stop_0]) + i,:].name
raw_next_month = [0 if j != next_month.month else 1 for j in range(1,13)]
next_month_array = np.array(raw_next_month).reshape(1,-1)
forecasted_prices.append(reg.predict(next_month_array)[0])
# For the current month.
raw_next_month = [0 if j != next_month.month + 1 else 1 for j in range(1,13)]
next_month_array = np.array(raw_next_month).reshape(1,-1)
forecasted_prices.append(reg.predict(next_month_array)[0])
return stop_0, forecasted_prices
### ARIMA bands Model ###
def prepare_data_to_ARIMA(self,df):
'''
Make a pandas series with the maximun observed prices within the month.
If the latest observed prices is before the 25th of the month, that month
is considered as incomplete: on course.
Interpolate the missing values.
'''
cfd = df.copy()
cfd = cfd.set_index(pd.to_datetime(cfd['date_price'])).groupby(pd.Grouper(freq='M')).max().interpolate('linear')
cfd = cfd.iloc[:,-1]
if df['date_price'].max().day < 25:
cfd = cfd[:-1]
result_adft = adfuller(cfd)
if result_adft[0] < result_adft[4]['10%']:
error = None
else:
error = 'Not Stationary'
return cfd, error
# Evaluate an ARIMA model for a given order (p,d,q)
def evaluate_arima_model(self,X, arima_order):
# Train/test split
train_size = int(len(X)*.65)
X_train, X_test = X[0:train_size], X[train_size:]
history = [x for x in X_train]
# forecast
forecast = []
for t in range(len(X_test)):
model = ARIMA(history, order=arima_order)
model_fit = model.fit()
y_pred = model_fit.predict()[0]
forecast.append(y_pred)
history.append(X_test[t])
# Calculate the mean squared error
error = mean_squared_error(X_test,forecast)
return error
def evaluate_models(self,timeseries, p_values, d_values, q_values):
timeseries = timeseries.astype('float32')
best_score, best_params = float('inf'), None
possible_params = [(p,d,q) for p in p_values for d in d_values for q in q_values]
for params in possible_params:
try:
mse = self.evaluate_arima_model(timeseries, params)
if mse < best_score:
best_score, best_params = mse, params
except:
continue
return best_score, best_params
def find_best_possible_ARIMA_params(self,df):
res = sm.tsa.arma_order_select_ic(df, ic='aic', trend='c')
prop_params = res.aic_min_order
# Parameters for evaluation.
p_values = range(0,prop_params[0]+1)
d_values = range(0,3)
q_values = range(0,prop_params[1] + 1)
best_score, best_params = self.evaluate_models(df, p_values, d_values, q_values)
return best_score, best_params
def inmediate_forecast_ARIMA_based(self,df, params):
forecasted_prices = []
mqerrors = []
stop_0 = datetime.date(2018,12,31)
baseset = df[:stop_0]
future_length = len(df)-len(baseset)
history = [x for x in baseset]
# we are predicting the next month.
for i in range(future_length + 1):
model = ARIMA(history, params)
model_fit = model.fit()#start_ar_lags=None)
y_pred = model_fit.predict()[0]
forecasted_prices.append(y_pred)
try:
history.append(df[len(df[:stop_0])+i])
mqerrors.append(mean_squared_error([history[-1]],[y_pred]))
except:
mqerrors.append(None)
return stop_0,forecasted_prices, mqerrors
def build_bands_wfp_forecast(self,df, stop_0, forecasted_prices):
'''
Takes the forecasted prices and build a dataframe with the ALPS bands,
and calculates the stressness of them.
'''
if isinstance(df, pd.Series):
df = pd.DataFrame(df)
errorstable = pd.DataFrame(index=pd.date_range(df.loc[stop_0:].index[0],datetime.date(df.index[-1].year,df.index[-1].month + 1, 1), freq='MS'),
columns=['observed_price','forecast'])
errorstable.iloc[:,0] = None
errorstable.iloc[:-1,0] = [x[0] for x in df.iloc[len(df.loc[:stop_0]):,:].values.tolist()]
errorstable.iloc[:,1] = forecasted_prices
errorstable['residuals'] = errorstable.iloc[:,0] - errorstable['forecast']
errorstable['cum_residual_std'] = [np.std(errorstable.iloc[:i,2]) for i in range(1,len(errorstable)+1)]
errorstable['ALPS'] = [None] + list(errorstable.iloc[1:,2] / errorstable.iloc[1:,3])
errorstable['Price Status'] = None
errorstable['Stressness'] = None
errorstable['normal_limit'] = errorstable['forecast'] + 0.25 * errorstable['cum_residual_std']
errorstable['stress_limit'] = errorstable['forecast'] + errorstable['cum_residual_std']
errorstable['alert_limit'] = errorstable['forecast'] + 2 * errorstable['cum_residual_std']
for date in range(len(errorstable)-1):
if errorstable.iloc[date,4] < 0.25:
errorstable.iloc[date,5] = 'Normal'
errorstable.iloc[date,6] = errorstable.iloc[date,0] / errorstable.iloc[date,7]
elif errorstable.iloc[date,4] < 1:
errorstable.iloc[date,5] = 'Stress'
errorstable.iloc[date,6] = errorstable.iloc[date,0] / errorstable.iloc[date,8]
elif errorstable.iloc[date,4] < 2:
errorstable.iloc[date,5] = 'Alert'
errorstable.iloc[date,6] = errorstable.iloc[date,0] / errorstable.iloc[date,9]
else:
errorstable.iloc[date,5] = 'Crisis'
errorstable.iloc[date,6] = errorstable.iloc[date,0] / errorstable.iloc[date,9]
mae = mean_squared_error(errorstable.iloc[:-1,0],errorstable.iloc[:-1,1])
return errorstable, mae
def set_columns_bands_df(self,bands):
'''
Builds a dataframe from the raw data for the bands, from the db.
'''
bands= pd.DataFrame(bands)
bands = bands.rename(columns={0:'date_price',1:'normal_band_limit',2:'stress_band_limit',3:'alert_band_limit', 4:'class_method'})
return bands
def assign_classification(self,data,bands):
'''
Combine the data from the prices and the bands to classify the price in its status.
'''
results = data.copy()
results['Observed_class'] = None
results['Stressness'] = None
results['class_method'] = bands.iloc[0,-1]
for i in range(len(results)):
bands_limits = bands[bands['date_price'] == datetime.date.fromisoformat(data.iloc[i,3])]
if results.iloc[i,2] < bands_limits.iloc[0,1]:
results.iloc[i,4] = 'Normal'
results.iloc[i,5] = results.iloc[i,2] / bands_limits.iloc[0,1]
elif results.iloc[i,2] < bands_limits.iloc[0,2]:
results.iloc[i,4] = 'Stress'
results.iloc[i,5] = results.iloc[i,2] / bands_limits.iloc[0,2]
elif results.iloc[i,2] < bands_limits.iloc[0,3]:
results.iloc[i,4] = 'Alert'
results.iloc[i,5] = results.iloc[i,2] / bands_limits.iloc[0,3]
else:
results.iloc[i,4] = 'Crisis'
results.iloc[i,5] = results.iloc[i,2] / bands_limits.iloc[0,3]
results = results.drop(labels=['month'], axis=1)
return results
def run_build_ALPS_bands(self,data):
'''
A method that runs in a line the methods required for the (original) ALPS bands.
'''
metric, cleaned = self.basic_cleanning(data)
try:
stop_0, forecasted_prices = self.inmediate_forecast_ALPS_based(self.prepare_data_to_ALPS(cleaned))
result, mae = self.build_bands_wfp_forecast(self.prepare_data_to_ALPS(cleaned),stop_0,forecasted_prices)
return metric, stop_0, result, mae
except:
return None, None, None, None
def run_build_arima_ALPS_bands(self,data):
'''
A method that runs in a line the methods required for the ARIMA based ALPS bands.
'''
metric, cleaned = self.basic_cleanning(data)
cfd, error = self.prepare_data_to_ARIMA(cleaned)
if not error:
try:
_, best_params = self.find_best_possible_ARIMA_params(cfd)
stop_0, forecasted_prices, _ = self.inmediate_forecast_ARIMA_based(cfd, best_params)
result, mse = self.build_bands_wfp_forecast(cfd,stop_0,forecasted_prices)
return metric, stop_0, result, mse
except:
return None, None, None, None
else:
return None, None, None, None
def possible_product_market_pairs_for_alps():
'''
Pulls the data from the table raw_table and stablishes a set of product/market pair
might be worth to work with.
It makes a dictionary with dataframes for all combination possibles and also returns a list
of the 'worth ones' to build the stress bands.
'''
try:
# Stablishes connection with our db.
connection = psycopg2.connect(user=os.environ.get('aws_db_user'),
password=os.environ.get('aws_db_password'),
host=os.environ.get('aws_db_host'),
port=os.environ.get('aws_db_port'),
database=os.environ.get('aws_db_name'))
query_retail = '''
SELECT *
FROM retail_stats
WHERE data_length > 730
AND max_date >= '2019-01-01'::date
AND std > 0
AND "ADF Statistic" < "Critical value 0.1";
'''
query_wholesale = '''
SELECT *
FROM wholesale_stats
WHERE data_length > 730
AND max_date >= '2019-01-01'::date
AND std > 0
AND "ADF Statistic" < "Critical value 0.1";
'''
retail_df = pd.read_sql(query_retail, con=connection)
retail_df = retail_df.drop(labels='index', axis=1)
wholesale_df = pd.read_sql(query_wholesale, con=connection)
wholesale_df = wholesale_df.drop(labels='index', axis=1)
retail_df['stationary'] = None
for i in range(len(retail_df)):
if retail_df.iloc[i,-13] > retail_df.iloc[i,-17]:
retail_df.iloc[i,-1] = True
else:
retail_df.iloc[i,-1] = False
strong_candidates_retail = retail_df[(retail_df['data_length'] > 365*3) & (retail_df['stationary'] == True)]
retail_df = retail_df.drop(labels=list(strong_candidates_retail.index), axis=0)
weak_candidates_retail = retail_df.copy()
wholesale_df['stationary'] = None
for i in range(len(wholesale_df)):
if wholesale_df.iloc[i,-13] > wholesale_df.iloc[i,-17]:
wholesale_df.iloc[i,-1] = True
else:
wholesale_df.iloc[i,-1] = False
strong_candidates_wholesale = wholesale_df[(wholesale_df['data_length'] > 365*3) & (wholesale_df['stationary'] == True)]
wholesale_df = wholesale_df.drop(labels=list(strong_candidates_wholesale.index), axis=0)
weak_candidates_wholesale = wholesale_df.copy()
strong_candidates_retail = strong_candidates_retail[['product_name','market_id','source_id','currency_code']].values.tolist()
weak_candidates_retail = weak_candidates_retail[['product_name','market_id','source_id','currency_code']].values.tolist()
strong_candidates_wholesale = strong_candidates_wholesale[['product_name','market_id','source_id','currency_code']].values.tolist()
weak_candidates_wholesale = weak_candidates_wholesale[['product_name','market_id','source_id','currency_code']].values.tolist()
return strong_candidates_retail, weak_candidates_retail, strong_candidates_wholesale, weak_candidates_wholesale
except (Exception, psycopg2.Error) as error:
print('Error pulling the data or forming the dictionary.')
finally:
if (connection):
connection.close()
def possible_product_market_pairs_for_arima_alps():
'''
Pulls the data from the table raw_table and stablishes a set of product/market pair
might be worth to work with.
'''
try:
# Stablishes connection with our db.
connection = psycopg2.connect(user=os.environ.get('aws_db_user'),
password=os.environ.get('aws_db_password'),
host=os.environ.get('aws_db_host'),
port=os.environ.get('aws_db_port'),
database=os.environ.get('aws_db_name'))
query_retail = '''
SELECT *
FROM retail_stats
WHERE data_length > 179
AND max_date >= '2019-01-01'::date
AND std > 0
AND "ADF Statistic" < "Critical value 0.1";
'''
query_wholesale = '''
SELECT *
FROM wholesale_stats
WHERE data_length > 179
AND max_date >= '2019-01-01'::date
AND std > 0
AND "ADF Statistic" < "Critical value 0.1";
'''
retail_df = pd.read_sql(query_retail, con=connection)
retail_df = retail_df.drop(labels='index', axis=1)
wholesale_df = pd.read_sql(query_wholesale, con=connection)
wholesale_df = wholesale_df.drop(labels='index', axis=1)
candidates_retail = retail_df[['product_name','market_id','source_id','currency_code']].values.tolist()
candidates_wholesale = wholesale_df[['product_name','market_id','source_id','currency_code']].values.tolist()
return candidates_retail, candidates_wholesale
except (Exception, psycopg2.Error) as error:
print('Error pulling the data or forming the dictionary.')
finally:
if (connection):
connection.close()
def product_ws_hist_ALPS_bands(product_name, market_id, source_id, currency_code,model_name):
'''
Builds the wholesale historic ALPS bands.
'''
data = None
market_with_problems = []
try:
# Stablishes connection with our db.
connection = psycopg2.connect(user=os.environ.get('aws_db_user'),
password=os.environ.get('aws_db_password'),
host=os.environ.get('aws_db_host'),
port=os.environ.get('aws_db_port'),
database=os.environ.get('aws_db_name'))
# Create the cursor.
cursor = connection.cursor()
cursor.execute('''
SELECT date_price, unit_scale, wholesale_observed_price
FROM raw_table
WHERE product_name = %s
AND market_id = %s
AND source_id = %s
AND currency_code = %s
''', (product_name, market_id, source_id, currency_code))
data = cursor.fetchall()
except (Exception, psycopg2.Error) as error:
print('Error pulling the data.')
finally:
if (connection):
cursor.close()
connection.close()
if data:
data = pd.DataFrame(data, columns=['date_price', 'unit_scale', 'observed_price'])
# Clean, prepare the data, build the ALPS bands.
clean_class = Clean_and_classify_class()
metric, _, wfp_forecast, _ = clean_class.run_build_ALPS_bands(data)
if metric:
# If the bands were built, this code will be run to drop the info in the db.
wfp_forecast = wfp_forecast.reset_index()
# try:
# Stablishes connection with our db.
connection = psycopg2.connect(user=os.environ.get('aws_db_user'),
password=os.environ.get('aws_db_password'),
host=os.environ.get('aws_db_host'),
port=os.environ.get('aws_db_port'),
database=os.environ.get('aws_db_name'))
# Create the cursor.
cursor = connection.cursor()
for row in wfp_forecast.values.tolist():
date_price = str(row[0].strftime("%Y-%m-%d"))
date_run_model = str(datetime.date(datetime.datetime.today().year, datetime.datetime.today().month, datetime.datetime.today().day).strftime("%Y-%m-%d"))
observed_price = row[1]
if observed_price:
observed_price = round(row[1],4)
observed_class = row[6]
class_method = model_name
normal_band_limit = round(row[8],4)
stress_band_limit = round(row[9],4)
alert_band_limit = round(row[10],4)
vector = (product_name,market_id,source_id,currency_code,metric,date_price,
observed_price,observed_class,class_method,date_run_model,
normal_band_limit,stress_band_limit,alert_band_limit)
# try:
cursor.execute('''
DELETE FROM wholesale_bands
WHERE product_name = %s
AND market_id = %s
AND unit_scale = %s
AND source_id = %s
AND currency_code = %s
AND date_price = %s
AND observed_price IS NULL
AND class_method = %s
''', (product_name, market_id,metric,source_id,currency_code,date_price, model_name))
# except:
# pass
cursor.execute('''
SELECT id
FROM wholesale_bands
WHERE product_name = %s
AND market_id = %s
AND source_id = %s
AND currency_code = %s
AND unit_scale = %s
AND date_price = %s
AND observed_class = %s
AND class_method = %s
''', (product_name,market_id,source_id,currency_code,metric,date_price,
observed_class,class_method))
result = cursor.fetchall()
if not result:
query_insert_results ='''
INSERT INTO wholesale_bands (
product_name,
market_id,
source_id,
currency_code,
unit_scale,
date_price,
observed_price,
observed_class,
class_method,
date_run_model,
normal_band_limit,
stress_band_limit,
alert_band_limit
)
VALUES (
%s,
%s,
%s,
%s,
%s,
%s,
%s,
%s,
%s,
%s,
%s,
%s,
%s
);
'''
cursor.execute(query_insert_results, vector)
connection.commit()
connection.close()
else:
print('The combination:',product_name, market_id, source_id, currency_code, 'has problems.')
market_with_problems.append((product_name, market_id, source_id, currency_code))
return market_with_problems
def product_rt_hist_ALPS_bands(product_name, market_id, source_id, currency_code,model_name):
'''
Builds the retail historic ALPS bands.
'''
data = None
market_with_problems = []
try:
# Stablishes connection with our db.
connection = psycopg2.connect(user=os.environ.get('aws_db_user'),
password=os.environ.get('aws_db_password'),
host=os.environ.get('aws_db_host'),
port=os.environ.get('aws_db_port'),
database=os.environ.get('aws_db_name'))
# Create the cursor.
cursor = connection.cursor()
cursor.execute('''
SELECT date_price, unit_scale, wholesale_observed_price
FROM raw_table
WHERE product_name = %s
AND market_id = %s
AND source_id = %s
AND currency_code = %s
''', (product_name, market_id, source_id, currency_code))
data = cursor.fetchall()
except (Exception, psycopg2.Error) as error:
print('Error pulling the data.')
finally:
if (connection):
cursor.close()
connection.close()
if data:
# Clean, prepare the data, build the ALPS bands.
data = pd.DataFrame(data, columns=['date_price', 'unit_scale', 'observed_price'])
clean_class = Clean_and_classify_class()
metric, _, wfp_forecast, _ = clean_class.run_build_ALPS_bands(data)
if metric:
# If the bands were built, this code will be run to drop the info in the db.
wfp_forecast = wfp_forecast.reset_index()
# try:
# Stablishes connection with our db.
connection = psycopg2.connect(user=os.environ.get('aws_db_user'),
password=os.environ.get('aws_db_password'),
host=os.environ.get('aws_db_host'),
port=os.environ.get('aws_db_port'),
database=os.environ.get('aws_db_name'))
# Create the cursor.
cursor = connection.cursor()
for row in wfp_forecast.values.tolist():
date_price = str(row[0].strftime("%Y-%m-%d"))
date_run_model = str(datetime.date(datetime.datetime.today().year, datetime.datetime.today().month, datetime.datetime.today().day).strftime("%Y-%m-%d"))
observed_price = row[1]
observed_class = row[6]
class_method = model_name
normal_band_limit = round(row[8],4)
stress_band_limit = round(row[9],4)
alert_band_limit = round(row[10],4)
vector = (product_name,market_id,source_id,currency_code,metric,date_price,
observed_price,observed_class,class_method,date_run_model,
normal_band_limit,stress_band_limit,alert_band_limit)
# try:
cursor.execute('''
DELETE FROM retail_bands
WHERE product_name = %s
AND market_id = %s
AND unit_scale = %s
AND source_id = %s
AND currency_code = %s
AND date_price = %s
AND observed_price IS NULL
AND class_method = %s
''', (product_name, market_id,metric,source_id,currency_code,date_price, model_name))
# except:
# pass
cursor.execute('''
SELECT id
FROM retail_bands
WHERE product_name = %s
AND market_id = %s
AND source_id = %s
AND currency_code = %s
AND unit_scale = %s
AND date_price = %s
AND observed_class = %s
AND class_method = %s
''', (product_name,market_id,source_id,currency_code,metric,date_price,
observed_class,class_method))
result = cursor.fetchall()
if not result:
query_insert_results ='''
INSERT INTO retail_bands (
product_name,
market_id,
source_id,
currency_code,
unit_scale,
date_price,
observed_price,
observed_class,
class_method,
date_run_model,
normal_band_limit,
stress_band_limit,
alert_band_limit
)
VALUES (
%s,
%s,
%s,
%s,
%s,
%s,
%s,
%s,
%s,
%s,
%s,
%s,
%s
);
'''
cursor.execute(query_insert_results, vector)
connection.commit()
connection.close()
else: