-
Notifications
You must be signed in to change notification settings - Fork 1
/
main_DRIM.py
168 lines (115 loc) · 6.3 KB
/
main_DRIM.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
# -*- coding: utf-8 -*-
"""
Created on Mon Oct 26 21:02:50 2020
@author: mehdi
"""
import pandas as pd
import numpy as np
from pandas.tseries.offsets import MonthEnd
from datetime import datetime
from functools import reduce
import os
import matplotlib.dates as mdates
from scipy import stats
from sklearn.cluster import DBSCAN
from sklearn.preprocessing import *
from typing import Union
from statsmodels.tsa.ar_model import AutoReg
from tqdm import trange
# Ecrivez ICI le chemin ou vous avez mis le dossier DRIM après l'avoir décompréssé
chemin_DRIM = r'C:\Users\mehdi\Desktop\M2 MOSEF\Scoring\DRIM'
os.chdir(r'{}\Codes'.format(chemin_DRIM))
from cleaning import add_stat_features , parse_date_inter ,throw_very_censored,\
select_timeframe_inter , import_celan_macro_tables ,\
add_macro_features, split_maturity , export_bases_df_maturity,\
label_encode , outliers_processing, project_censored_contracts,\
tweak_maturity_lost_contracts , croiseur, correct_projection ,\
add_crise, add_temporality, valeur_ref , anomalie, add_fusion , \
add_fusion_import, concat_dicos
from plot_stab_temporelle import eval_risk, eval_amount , get_evol,\
plot_stability ,preprocess_for_stability_plot
from discretization import preprocess_for_discretization , calc_distance ,\
select_best_k , discretization_machine , discretization_k22,\
discretization_k_all , discretize
# Importer la base
data = pd.read_csv(r'{}\Base\Table_LGD.txt'.format(chemin_DRIM) , sep='\s')
df = data.dropna()
# Selectionner les maturités qui nous intéressent
df = df[df['_maturity_']<= 24]
# Supprimer les contrats très censurés
df = throw_very_censored(df)
# Définir si on veut exporter les bases nettoyées ou non
export = True
# Rajouter des features STATS
df = add_stat_features(df)
# Importer des features MACRO
chom_spain, chom_euro_area,\
cli_spain, cli_euro_area,\
tic_spain, tic_euro_area,\
cpi_spain , cpi_euro_area = import_celan_macro_tables(chemin_DRIM)
# Rajouter les features MACRO
dfs = [chom_spain, chom_euro_area, cli_spain, cli_euro_area, tic_spain, tic_euro_area, cpi_spain, cpi_euro_area]
df = add_macro_features(df,dfs)
# Diviser par le seuil
df = valeur_ref(df)
# Rajout des contrats fusionnés
#df = add_fusion(df)
# Split en maturité et
df = tweak_maturity_lost_contracts (df)
mat6 , mat9 , mat12 , mat18 , mat24 = split_maturity(df)
# Rajout des contrats fusionnés
fusion = pd.read_csv(r'{}\Output\fusion.csv'.format(chemin_DRIM) , sep=';')
fusion = fusion.drop_duplicates()
mat6 , mat9 , mat12 , mat18 , mat24 = add_fusion_import(mat6,fusion),add_fusion_import(mat9,fusion),\
add_fusion_import(mat12,fusion) , add_fusion_import(mat18,fusion), add_fusion_import(mat24,fusion)
# Traitement des valeurs aberrantes
type_locate = "dbscan" # dbscan , zscore
treat_or_delete = "treat" # treat , delete
columns = ['ead','MT_INI_FIN_','mt_appo_'] #liste de colonnes à checker pour valeurs aberrantes
mat6 , mat9 , mat12 , mat18 , mat24 = outliers_processing( mat6 , type_locate , treat_or_delete , columns),\
outliers_processing( mat9 , type_locate , treat_or_delete , columns),\
outliers_processing( mat12 , type_locate , treat_or_delete , columns),\
outliers_processing( mat18 , type_locate , treat_or_delete , columns),\
outliers_processing( mat24 , type_locate , treat_or_delete , columns)
# Rajout de crise
mat6 , mat9 , mat12 , mat18 , mat24 = add_crise(mat6,chemin_DRIM) , add_crise(mat9,chemin_DRIM),\
add_crise(mat12,chemin_DRIM) , add_crise(mat18,chemin_DRIM) , add_crise(mat24,chemin_DRIM)
# Projection des contrats censurés
# Paramètre optionnel 'ar' qui correspond au p de AR(p) par defaut p=2
# concat les 5 maturités
mat24 = project_censored_contracts(df , mat24 )
mat24 = correct_projection(mat6 , mat9 , mat12 , mat18 , mat24)
# Rajouter la temporalité
#mat9 , mat12 , mat18 , mat21 = add_temporality(mat6 , mat9 , mat12 , mat18 , mat21)
# Label encoding puis croiser les variables
for col in ['cat_seg','CD_CAT_EXPO_4','qual_veh','nat_veh']:
mat6, mat9 , mat12, mat18, mat24 = label_encode(mat6,col) , label_encode(mat9,col) , label_encode(mat12,col) ,\
label_encode(mat18,col) , label_encode(mat24,col)
list_cat = ['cat_seg','CD_CAT_EXPO_4','qual_veh']
list_cont = ['MT_INI_FIN_','mt_appo_','DUR_PREV_FIN','ead']
mat6, mat9 , mat12, mat18, mat24 = croiseur(mat6, list_cat , list_cont) , croiseur(mat9, list_cat , list_cont),\
croiseur(mat12, list_cat , list_cont) , croiseur(mat18, list_cat , list_cont),\
croiseur(mat24, list_cat , list_cont)
# Choisir si on discrétise
# disc = False
# VarList=['ead','DUR_PREV_FIN','MT_INI_FIN_','dur_b_defm_', 'dur_b_endm_', 'mt_appo_','pct_appo_','ratio_b_endm_']
# drop_existing=False
# if disc:
# # Choisir le type de discretisation ici
# type_disc = 'k_all' # k_all / k22 / gmm /
# mat6 , mat9 , mat12 , mat18 , mat21 , concat = discretize(mat6,type_disc,VarList,drop_existing),discretize(mat9,type_disc,VarList,drop_existing),\
# discretize(mat12,type_disc,VarList,drop_existing),discretize(mat18,type_disc,VarList,drop_existing),\
# discretize(mat21,type_disc,VarList,drop_existing)
# Exportation des bases
if export:
concat = pd.concat([mat6,mat9,mat12,mat18,mat24])
export_bases_df_maturity(chemin_DRIM, df, concat, mat6 , mat9 , mat12 , mat18 , mat24)
# Plot de stabilité Temporelle :
stability_plot = False
if stability_plot:
mat6_plot = preprocess_for_stability_plot(mat6)
mat21_plot = preprocess_for_stability_plot(mat21)
#Créer la liste des variables que l'on veut plus
Liste=['CD_CAT_EXPO_4', 'cat_seg', 'nat_veh', 'qual_veh']
plot_stability(mat6_plot, Liste, "dtf_per_trt", "Tx_rec_marg_Bin", 1)
plot_stability(mat21_plot, Liste, "dtf_per_trt", "Tx_rec_marg_Bin", 1)