Fraud-Detection-MLOps/dvc.lock at dev · Mark007-R/Fraud-Detection-MLOps · GitHub

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
schema: '2.0'
stages:
  load_fdb:
    cmd: .venv\Scripts\python.exe src/load_fdb.py
    deps:
    - path: src/load_fdb.py
      hash: md5
      md5: efb5600ad2bee3915e50d271b9c1a3c8
      size: 5192
    params:
      params.yaml:
        fdb:
          api_key: ''
          dataset_key: sparkov
          dataset_version: latest
          enable_cache: true
          cache_dir: data/raw/fdb_cache
    outs:
    - path: data/raw/sparkov.csv
      hash: md5
      md5: 3f43d5a6aab3b5d0b5bc2676e3626496
      size: 347981313
  combine:
    cmd: .venv\Scripts\python.exe src/combine_datasets.py
    deps:
    - path: data/raw/paysim.csv
      hash: md5
      md5: e92a5f7447f43712f1dca473d0b0fa85
      size: 493534783
    - path: data/raw/sparkov.csv
      hash: md5
      md5: 3f43d5a6aab3b5d0b5bc2676e3626496
      size: 347981313
    - path: src/combine_datasets.py
      hash: md5
      md5: d140106f41317873e30825e6dac7ec90
      size: 5770
    params:
      params.yaml:
        combine:
          paysim_path: data/raw/paysim.csv
          sparkov_path: data/raw/sparkov.csv
          output_path: data/processed/combined_transactions.csv
    outs:
    - path: data/processed/combined_transactions.csv
      hash: md5
      md5: df13f1afe13e868571868bfdc0163b5c
      size: 400995390
  preprocess:
    cmd: .venv\Scripts\python.exe src/preprocess.py
    deps:
    - path: data/processed/combined_transactions.csv
      hash: md5
      md5: df13f1afe13e868571868bfdc0163b5c
      size: 400995390
    - path: src/preprocess.py
      hash: md5
      md5: 92c323d41d4fbd0fb68053af778cbad6
      size: 6981
    params:
      params.yaml:
        preprocess:
          input: data/processed/combined_transactions.csv
          output: data/processed/features.csv
    outs:
    - path: data/processed/features.csv
      hash: md5
      md5: 10dc68d489cf1975392ec2272d11e941
      size: 1532739766
  train:
    cmd: .venv\Scripts\python.exe src/train.py
    deps:
    - path: data/processed/features.csv
      hash: md5
      md5: 10dc68d489cf1975392ec2272d11e941
      size: 1532739766
    - path: src/train.py
      hash: md5
      md5: 54316d8b2f2521510e2faa333bb42d93
      size: 3815
    params:
      params.yaml:
        data:
          raw_paysim_path: data/raw/paysim.csv
          raw_sparkov_path: data/raw/sparkov.csv
          combined_path: data/processed/combined_dataset.csv
          features_path: data/processed/features.csv
          train_path: data/processed/train.csv
          test_path: data/processed/test.csv
          x_test_path: data/processed/X_test.csv
          y_test_path: data/processed/y_test.csv
          fdb_sparkov_test_path: data/raw/sparkov_test_split.csv
          model_path: models/fraud_model.pkl
          metrics_path: metrics/scores.json
          confusion_matrix_path: reports/confusion_matrix.png
          benchmark_path: metrics/fdb_benchmark.json
        train:
          model_type: xgboost
          test_size: 0.2
          random_state: 42
          n_estimators: 200
          max_depth: 6
          learning_rate: 0.1
          scale_pos_weight: 50
          use_smote: false
    outs:
    - path: data/processed/X_test.csv
      hash: md5
      md5: 3145d4b82ad9d0afffcad4038526947d
      size: 301889620
    - path: data/processed/y_test.csv
      hash: md5
      md5: c2e17bc827fe3df84961f5696475e023
      size: 4595587
    - path: models/fraud_model.pkl
      hash: md5
      md5: 1346c7d7474cca4d1c8ba8eed5b674fa
      size: 675622
  evaluate:
    cmd: .venv\Scripts\python.exe src/evaluate.py
    deps:
    - path: data/processed/X_test.csv
      hash: md5
      md5: 3145d4b82ad9d0afffcad4038526947d
      size: 301889620
    - path: data/processed/features.csv
      hash: md5
      md5: 10dc68d489cf1975392ec2272d11e941
      size: 1532739766
    - path: data/processed/y_test.csv
      hash: md5
      md5: c2e17bc827fe3df84961f5696475e023
      size: 4595587
    - path: models/fraud_model.pkl
      hash: md5
      md5: 1346c7d7474cca4d1c8ba8eed5b674fa
      size: 675622
    - path: src/evaluate.py
      hash: md5
      md5: 369422d6dcc6edc74c755665affa3a80
      size: 8360
    params:
      params.yaml:
        data:
          raw_paysim_path: data/raw/paysim.csv
          raw_sparkov_path: data/raw/sparkov.csv
          combined_path: data/processed/combined_dataset.csv
          features_path: data/processed/features.csv
          train_path: data/processed/train.csv
          test_path: data/processed/test.csv
          x_test_path: data/processed/X_test.csv
          y_test_path: data/processed/y_test.csv
          fdb_sparkov_test_path: data/raw/sparkov_test_split.csv
          model_path: models/fraud_model.pkl
          metrics_path: metrics/scores.json
          confusion_matrix_path: reports/confusion_matrix.png
          benchmark_path: metrics/fdb_benchmark.json
        evaluate:
          model_path: models/fraud_model.pkl
          metrics_path: metrics/scores.json
          reports_dir: reports/
    outs:
    - path: metrics/scores.json
      hash: md5
      md5: 11405fe4787d5f3b38df843f90d4eeeb
      size: 221
    - path: reports/confusion_matrix.svg
      hash: md5
      md5: 8557f84c12b17934388b9c2a0546d214
      size: 1745
    - path: reports/roc_curve.svg
      hash: md5
      md5: 25719a0a97d8aae231723d26d8bf1a26
      size: 2421320
  benchmark_fdb:
    cmd: .venv\Scripts\python.exe -m src.benchmark_fdb
    deps:
    - path: models/fraud_model.pkl
      hash: md5
      md5: 1346c7d7474cca4d1c8ba8eed5b674fa
      size: 675622
    - path: src/benchmark_fdb.py
      hash: md5
      md5: e72d8cf3e3743466df2eb6537705e7ba
      size: 7398
    params:
      params.yaml:
        benchmark:
          model_path: models/fraud_model.pkl
          fdb_key: sparknov
          output_path: metrics/fdb_benchmark.json
    outs:
    - path: metrics/fdb_benchmark.json
      hash: md5
      md5: d4934d32df89a506592cda8980002015
      size: 573