mstz commited on
Commit
e89b705
·
1 Parent(s): 5c743f2

Upload 4 files

Browse files
Files changed (5) hide show
  1. .gitattributes +2 -0
  2. README.md +61 -0
  3. Test_Dataset.csv +3 -0
  4. Train_Dataset.csv +3 -0
  5. nbfi.py +235 -0
.gitattributes CHANGED
@@ -52,3 +52,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
52
  *.jpg filter=lfs diff=lfs merge=lfs -text
53
  *.jpeg filter=lfs diff=lfs merge=lfs -text
54
  *.webp filter=lfs diff=lfs merge=lfs -text
 
 
 
52
  *.jpg filter=lfs diff=lfs merge=lfs -text
53
  *.jpeg filter=lfs diff=lfs merge=lfs -text
54
  *.webp filter=lfs diff=lfs merge=lfs -text
55
+ Test_Dataset.csv filter=lfs diff=lfs merge=lfs -text
56
+ Train_Dataset.csv filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+ tags:
5
+ - nbfi
6
+ - tabular_classification
7
+ - binary_classification
8
+ pretty_name: Adult
9
+ size_categories:
10
+ - 10K<n<100K
11
+ task_categories: # Full list at https://github.com/huggingface/hub-docs/blob/main/js/src/lib/interfaces/Types.ts
12
+ - tabular-classification
13
+ configs:
14
+ - default
15
+ ---
16
+ # NBFI
17
+ The [NBFI dataset](https://www.kaggle.com/datasets/meastanmay/nbfi-vehicle-loan-repayment-dataset) from the [Kaggle](https://www.kaggle.com/datasets).
18
+ Client default prediction
19
+
20
+ # Configurations and tasks
21
+ - `default`, for binary classification of the individual's future default status.
22
+
23
+ # Features
24
+ |**Feature** |**Type** |
25
+ |-----------------------------------------------|---------------|
26
+ |`income` | `float32` |
27
+ |`owns_a_car` | `bool` |
28
+ |`owns_a_bike` | `bool` |
29
+ |`has_an_active_loan` | `bool` |
30
+ |`owns_a_house` | `bool` |
31
+ |`nr_children` | `int8` |
32
+ |`credit` | `float32` |
33
+ |`loan_annuity` | `float32` |
34
+ |`accompanied_by` | `string` |
35
+ |`income_type` | `string` |
36
+ |`education_level` | `float32` |
37
+ |`marital_status` | `float32` |
38
+ |`is_male` | `bool` |
39
+ |`type_of_contract` | `string` |
40
+ |`type_of_housing` | `string` |
41
+ |`residence_density` | `float32` |
42
+ |`age_in_days` | `int32` |
43
+ |`consecutive_days_of_employment` | `int16` |
44
+ |`nr_days_since_last_registration_change` | `int32` |
45
+ |`nr_days_since_last_document_change` | `int32` |
46
+ |`owned_a_house_for_nr_days` | `int32` |
47
+ |`has_provided_a_mobile_number` | `bool` |
48
+ |`has_provided_a_home_number` | `bool` |
49
+ |`was_reachable_at_work` | `bool` |
50
+ |`job` | `string` |
51
+ |`nr_family_members` | `int8` |
52
+ |`city_rating` | `int8` |
53
+ |`weekday_of_application` | `int8` |
54
+ |`hour_of_application` | `float32` |
55
+ |`same_residence_and_home` | `bool` |
56
+ |`same_work_and_home` | `bool` |
57
+ |`score_1` | `float32` |
58
+ |`score_2` | `float32` |
59
+ |`score_3` | `float32` |
60
+ |`nr_defaults_in_social_circle` | `int8` |
61
+ |`inquiries_in_last_year` | `float32` |
Test_Dataset.csv ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fce61c89339cc849186d4e9964c8d154c933359ea51e5c8d401c4312040aa137
3
+ size 15229326
Train_Dataset.csv ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bfaa92b0e2bf3cb523c2c287d00502328bdc19366cdb2a49486c58ce7cae5e33
3
+ size 23193688
nbfi.py ADDED
@@ -0,0 +1,235 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ """NBFI: A Census Dataset"""
2
+
3
+ from typing import List
4
+ from functools import partial
5
+
6
+ import datasets
7
+
8
+ import pandas
9
+
10
+
11
+ VERSION = datasets.Version("1.0.0")
12
+ _ORIGINAL_FEATURE_NAMES = [
13
+ "ID",
14
+ "Client_Income",
15
+ "Car_Owned",
16
+ "Bike_Owned",
17
+ "Active_Loan",
18
+ "House_Own",
19
+ "Child_Count",
20
+ "Credit_Amount",
21
+ "Loan_Annuity",
22
+ "Accompany_Client",
23
+ "Client_Income_Type",
24
+ "Client_Education",
25
+ "Client_Marital_Status",
26
+ "Client_Gender",
27
+ "Loan_Contract_Type",
28
+ "Client_Housing_Type",
29
+ "Population_Region_Relative",
30
+ "Age_Days",
31
+ "Employed_Days",
32
+ "Registration_Days",
33
+ "ID_Days",
34
+ "Own_House_Age",
35
+ "Mobile_Tag",
36
+ "Homephone_Tag",
37
+ "Workphone_Working",
38
+ "Client_Occupation",
39
+ "Client_Family_Members",
40
+ "Cleint_City_Rating",
41
+ "Application_Process_Day",
42
+ "Application_Process_Hour",
43
+ "Client_Permanent_Match_Tag",
44
+ "Client_Contact_Work_Tag",
45
+ "Type_Organization",
46
+ "Score_Source_1",
47
+ "Score_Source_2",
48
+ "Score_Source_3",
49
+ "Social_Circle_Default",
50
+ "Phone_Change",
51
+ "Credit_Bureau",
52
+ "Default"
53
+ ]
54
+ _BASE_FEATURE_NAMES = [
55
+ "income",
56
+ "owns_a_car",
57
+ "owns_a_bike",
58
+ "has_an_active_loan",
59
+ "owns_a_house",
60
+ "nr_children",
61
+ "credit",
62
+ "loan_annuity",
63
+ "accompanied_by",
64
+ "income_type",
65
+ "education_level",
66
+ "marital_status",
67
+ "is_male",
68
+ "type_of_contract",
69
+ "type_of_housing",
70
+ "residence_density",
71
+ "age_in_days",
72
+ "consecutive_days_of_employment",
73
+ "nr_days_since_last_registration_change",
74
+ "nr_days_since_last_document_change",
75
+ "owned_a_house_for_nr_days",
76
+ "has_provided_a_mobile_number",
77
+ "has_provided_a_home_number",
78
+ "was_reachable_at_work",
79
+ "job",
80
+ "nr_family_members",
81
+ "city_rating",
82
+ "weekday_of_application",
83
+ "hour_of_application",
84
+ "same_residence_and_home",
85
+ "same_work_and_home",
86
+ "score_1",
87
+ "score_2",
88
+ "score_3",
89
+ "nr_defaults_in_social_circle",
90
+ "inquiries_in_last_year",
91
+ "has_defaulted"
92
+ ]
93
+ features_types_per_config = {
94
+ "income": datasets.Value("float32"),
95
+ "owns_a_car": datasets.Value("bool"),
96
+ "owns_a_bike": datasets.Value("bool"),
97
+ "has_an_active_loan": datasets.Value("bool"),
98
+ "owns_a_house": datasets.Value("bool"),
99
+ "nr_children": datasets.Value("int8"),
100
+ "credit": datasets.Value("float32"),
101
+ "loan_annuity": datasets.Value("float32"),
102
+ "accompanied_by": datasets.Value("string"),
103
+ "income_type": datasets.Value("string"),
104
+ "education_level": datasets.Value("float32"),
105
+ "marital_status": datasets.Value("float32"),
106
+ "is_male": datasets.Value("bool"),
107
+ "type_of_contract": datasets.Value("string"),
108
+ "type_of_housing": datasets.Value("string"),
109
+ "residence_density": datasets.Value("float32"),
110
+ "age_in_days": datasets.Value("int32"),
111
+ "consecutive_days_of_employment": datasets.Value("int16"),
112
+ "nr_days_since_last_registration_change": datasets.Value("int32"),
113
+ "nr_days_since_last_document_change": datasets.Value("int32"),
114
+ "owned_a_house_for_nr_days": datasets.Value("int32"),
115
+ "has_provided_a_mobile_number": datasets.Value("bool"),
116
+ "has_provided_a_home_number": datasets.Value("bool"),
117
+ "was_reachable_at_work": datasets.Value("bool"),
118
+ "job": datasets.Value("string"),
119
+ "nr_family_members": datasets.Value("int8"),
120
+ "city_rating": datasets.Value("int8"),
121
+ "weekday_of_application": datasets.Value("int8"),
122
+ "hour_of_application": datasets.Value("float32"),
123
+ "same_residence_and_home": datasets.Value("bool"),
124
+ "same_work_and_home": datasets.Value("bool"),
125
+ "score_1": datasets.Value("float32"),
126
+ "score_2": datasets.Value("float32"),
127
+ "score_3": datasets.Value("float32"),
128
+ "nr_defaults_in_social_circle": datasets.Value("int8"),
129
+ "inquiries_in_last_year": datasets.Value("float32"),
130
+ "has_defaulted": datasets.ClassLabel(num_classes=2, names=("no", "yes"))
131
+ }
132
+ _ENCODING_DICS = {}
133
+ _EDUCATION_ENCODING = {
134
+ "Junior secondary": 0,
135
+ "Secondary": 1,
136
+ "Graduation dropout": 2,
137
+ "Graduation": 2,
138
+ "Post": 4
139
+ }
140
+
141
+ DESCRIPTION = "NBFI dataset from default prediction."
142
+ _HOMEPAGE = "https://www.kaggle.com/datasets/meastanmay/nbfi-vehicle-loan-repayment-dataset"
143
+ _URLS = ("https://www.kaggle.com/datasets/meastanmay/nbfi-vehicle-loan-repayment-dataset")
144
+ _CITATION = """"""
145
+
146
+ # Dataset info
147
+ urls_per_split = {
148
+ "train": "https://huggingface.co/datasets/mstz/nbfi/raw/main/Train_Dataset.csv",
149
+ "test": "https://huggingface.co/datasets/mstz/nbfi/raw/main/Test_Dataset.csv"
150
+ }
151
+
152
+ features_per_config = {k: datasets.Features(features_types_per_config[k]) for k in features_types_per_config}
153
+
154
+
155
+ class NBFIConfig(datasets.BuilderConfig):
156
+ def __init__(self, **kwargs):
157
+ super(NBFIConfig, self).__init__(version=VERSION, **kwargs)
158
+ self.features = features_per_config[kwargs["name"]]
159
+
160
+
161
+ class NBFI(datasets.GeneratorBasedBuilder):
162
+ # dataset versions
163
+ DEFAULT_CONFIG = "default"
164
+ BUILDER_CONFIGS = [
165
+ NBFIConfig(name="default",
166
+ description="NBFI for default binary classification.")
167
+ ]
168
+
169
+
170
+ def _info(self):
171
+ if self.config.name not in features_per_config:
172
+ raise ValueError(f"Unknown configuration: {self.config.name}")
173
+
174
+ info = datasets.DatasetInfo(description=DESCRIPTION, citation=_CITATION, homepage=_HOMEPAGE,
175
+ features=features_per_config[self.config.name])
176
+
177
+ return info
178
+
179
+ def _split_generators(self, dl_manager: datasets.DownloadManager) -> List[datasets.SplitGenerator]:
180
+ downloads = dl_manager.download_and_extract(urls_per_split)
181
+
182
+ return [
183
+ datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloads["train"]}),
184
+ datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={"filepath": downloads["test"]}),
185
+ ]
186
+
187
+ def _generate_examples(self, filepath: str):
188
+ if self.config.name == "default":
189
+ data = pandas.read_csv(filepath)
190
+ data = self.preprocess(data, config=self.config.name)
191
+
192
+ for row_id, row in data.iterrows():
193
+ data_row = dict(row)
194
+
195
+ yield row_id, data_row
196
+ else:
197
+ raise ValueError(f"Unknown config: {self.config.name}")
198
+
199
+
200
+
201
+ def preprocess(self, data: pandas.DataFrame, config: str = DEFAULT_CONFIG) -> pandas.DataFrame:
202
+ data.drop("ID", axis="columns", inplace=True)
203
+ data.drop("Own_House_Age", axis="columns", inplace=True)
204
+ data.drop("Type_Organization", axis="columns", inplace=True)
205
+ data.drop("Phone_Change", axis="columns", inplace=True)
206
+
207
+ data = data[~data.Client_Education.isna()]
208
+ data = data[~data.Client_Marital_Status.isna()]
209
+ data = data[~data.Client_Gender.isna()]
210
+ data = data[~data.Loan_Contract_Type.isna()]
211
+ data = data[~data.Client_Housing_Type.isna()]
212
+ data = data[~data.Age_Days.isna()]
213
+ data = data[~data.Employed_Days.isna()]
214
+ data = data[~data.Registration_Days.isna()]
215
+ data = data[~data.ID_Days.isna()]
216
+ data = data[~data.Cleint_City_Rating.isna()]
217
+ data = data[~data.Application_Process_Day.isna()]
218
+ data = data[~data.Application_Process_Hour.isna()]
219
+ data = data[~data.Client_Permanent_Match_Tag.isna()]
220
+ data = data[~data.Client_Contact_Work_Tag.isna()]
221
+ data.columns = _BASE_FEATURE_NAMES
222
+
223
+ data.iloc[:, "owns_a_car"] = data["owns_a_car"].apply(bool)
224
+ data.iloc[:, "owns_a_bike"] = data["owns_a_bike"].apply(bool)
225
+ data.iloc[:, "has_an_active_loan"] = data["has_an_active_loan"].apply(bool)
226
+ data.iloc[:, "owns_a_house"] = data["owns_a_house"].apply(bool)
227
+ data.iloc[:, "is_male"] = data["is_male"].apply(bool)
228
+ data.iloc[:, "has_provided_a_mobile_number"] = data["has_provided_a_mobile_number"].apply(bool)
229
+ data.iloc[:, "has_provided_a_home_number"] = data["has_provided_a_home_number"].apply(bool)
230
+ data.iloc[:, "was_reachable_at_work"] = data["was_reachable_at_work"].apply(bool)
231
+ data.iloc[:, "same_residence_and_home"] = data["same_residence_and_home"].apply(bool)
232
+ data.iloc[:, "same_work_and_home"] = data["same_work_and_home"].apply(bool)
233
+
234
+ return data[list(features_types_per_config[config].keys())]
235
+