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
|
- models:
- Logistic_Regression: LogisticRegression
- SGD_Classifier: SGDClassifier
- Random Forest: RandomForestClassifier
- Ada_Boost: AdaBoostClassifier
- Grad_Boost: GradientBoostingClassifier
- Light_GBM: LGBMClassifier
- Bagging_Classifier: BaggingClassifier
- ExtraTreesClassifier: ExtraTreesClassifier
- Hist_Grad_Boost_Classifier: HistGradientBoostingClassifier
- Decision_Tree_Classifier: DecisionTreeClassifier
- XGB_Classifier: XGBClassifier
- KNN_Classifier: KNeighborsClassifier
- optuna:
- Logistic_Regression:
- penalty: trial.suggest_categorical('penalty', ['l2', None])
- SGD_Classifier:
- loss: trial.suggest_categorical('loss', ['squared_epsilon_insensitive', 'epsilon_insensitive', 'huber', 'squared_error', 'perceptron', 'squared_hinge', 'hinge', 'log_loss', 'modified_huber'])
- Light_GBM:
- boosting_type: trial.suggest_categorical('boosting_type', ['gbdt','dart'])
- learning_rate: trial.suggest_float('learning_rate', .00001, 1.0)
- n_estimators: trial.suggest_int('n_estimators', 100, 150)
- class_weight: trial.suggest_categorical('class_weight', ['balanced'])
- n_jobs: trial.suggest_categorical('n_jobs', [-1])
- Random Forest:
- n_estimators: trial.suggest_int('n_estimators', 100, 150)
- criterion: trial.suggest_categorical('criterion', ['log_loss', 'entropy', 'gini'])
- max_features: trial.suggest_categorical('max_features', ['sqrt', 'log2', None])
- class_weight: trial.suggest_categorical('class_weight', ['balanced', 'balanced_subsample'])
- Ada_Boost:
- n_estimators: trial.suggest_int('n_estimators', 100, 150)
- algorithm: trial.suggest_categorical('algorithm', ['SAMME', 'SAMME.R'])
- Grad_Boost:
- loss: trial.suggest_categorical('loss', ['log_loss', 'exponential'])
- n_estimators: trial.suggest_int('n_estimators', 100, 150)
- criterion: trial.suggest_categorical('criterion', ['friedman_mse', 'squared_error'])
- max_features: trial.suggest_categorical('max_features', ['sqrt', 'log2', None])
- Bagging_Classifier:
- n_estimators: trial.suggest_int('n_estimators', 50, 100)
- n_jobs: trial.suggest_categorical('n_jobs', [-1])
- ExtraTreesClassifier:
- n_estimators: trial.suggest_int('n_estimators', 100, 500)
- criterion: trial.suggest_categorical('criterion', ['log_loss', 'entropy', 'gini'])
- max_features: trial.suggest_categorical('max_features', ['sqrt', 'log2', None])
- class_weight: trial.suggest_categorical('class_weight', ['balanced', 'balanced_subsample'])
- Hist_Grad_Boost_Classifier:
- max_iter: trial.suggest_int('max_iter', 100, 800)
- Decision_Tree_Classifier:
- criterion: trial.suggest_categorical('criterion', ['log_loss', 'entropy', 'gini'])
- splitter: trial.suggest_categorical('splitter', ['best', 'random'])
- max_features: trial.suggest_categorical('max_features', ['sqrt', 'log2', None])
- XGB_Classifier:
- n_estimators: trial.suggest_int('n_estimators', 100, 500)
- learning_rate: trial.suggest_float('learning_rate', .00001, 1.0)
- booster: trial.suggest_categorical('booster', ['gbtree', 'gblinear', 'dart'])
- tree_method: trial.suggest_categorical('tree_method', ['exact', 'approx', 'hist'])
- KNN_Classifier:
- n_neighbors: trial.suggest_int('n_neighbors', 3, 11, step=2)
- weights: trial.suggest_categorical('weights', ['uniform', 'distance'])
- algorithm: trial.suggest_categorical('algorithm', ['auto', 'ball_tree', 'kd_tree', 'brute'])
- # MLP_Classifier:
- # hidden_layer_sizes: trial.suggest_categorical('hidden_layer_sizes', [(500,), (500, 300, 200, 150,), (700, 500, 300, 100, ), (1500, 800, 400, 200, )])
- # activation: trial.suggest_categorical('activation', ['identity', 'logistic', 'tanh' , 'relu'])
- # learning_rate: trial.suggest_categorical('learning_rate', ['constant', 'invscaling', 'adaptive'])
- # max_iter: trial.suggest_int('max_iter', 100, 800)
- Stacked_Classifier:
- stack_method: trial.suggest_categorical('stack_method', ['auto', 'predict'])
- passthrough: trial.suggest_categorical('passthrough', [True, False])
- # hyperopt:
- # Logistic_Regression:
- # penalty: hp.choice('penalty', ['l2', None])
- # SGD_Classifier:
- # loss: hp.choice('loss', ['squared_epsilon_insensitive', 'epsilon_insensitive', 'huber', 'squared_error', 'perceptron', 'squared_hinge', 'hinge', 'log_loss', 'modified_huber'])
- # Random Forest:
- # n_estimators: scope.int(hp.quniform('n_estimators', 100, 150, 1))
- # criterion: hp.choice('criterion', ['log_loss', 'entropy', 'gini'])
- # max_features: hp.choice('max_features', ['sqrt', 'log2', None])
- # class_weight: hp.choice('class_weight', ['balanced', 'balanced_subsample'])
- # Ada_Boost:
- # n_estimators: scope.int(hp.quniform('n_estimators', 100, 150, 1))
- # algorithm: hp.choice('algorithm', ['SAMME', 'SAMME.R'])
- # Grad_Boost:
- # loss: hp.choice('loss', ['log_loss', 'exponential'])
- # n_estimators: scope.int(hp.quniform('n_estimators', 100, 150, 1))
- # criterion: hp.choice('criterion', ['friedman_mse', 'squared_error'])
- # max_features: hp.choice('max_features', ['sqrt', 'log2', None])
- # Bagging_Classifier:
- # n_estimators: scope.int(hp.quniform('n_estimators', 50, 100, 1))
- # ExtraTreesClassifier:
- # n_estimators: scope.int(hp.quniform('n_estimators', 100, 1000, 1))
- # criterion: hp.choice('criterion', ['log_loss', 'entropy', 'gini'])
- # max_features: hp.choice('max_features', ['sqrt', 'log2', None])
- # class_weight: hp.choice('class_weight', ['balanced', 'balanced_subsample'])
- # Hist_Grad_Boost_Classifier:
- # max_iter: scope.int(hp.quniform('max_iter', 100, 800, 1))
- # Decision_Tree_Classifier:
- # criterion: hp.choice('criterion', ['log_loss', 'entropy', 'gini'])
- # splitter: hp.choice('splitter', ['best', 'random'])
- # max_features: hp.choice('max_features', ['sqrt', 'log2', None])
- # XGB_Classifier:
- # n_estimators: scope.int(hp.quniform('n_estimators', 100, 200, 1))
- # learning_rate: hp.uniform('learning_rate', .00001, 1.0)
- # booster: hp.choice('booster', ['gbtree', 'gblinear', 'dart'])
- # tree_method: hp.choice('tree_method', ['exact', 'approx', 'hist'])
- # KNN_Classifier:
- # n_neighbors: scope.int(hp.quniform('n_neighbors', 3, 11, 2))
- # weights: hp.choice('weights', ['uniform', 'distance'])
- # algorithm: hp.choice('algorithm', ['auto', 'ball_tree', 'kd_tree', 'brute'])
- # MLP_Classifier:
- # hidden_layer_sizes: hp.choice('hidden_layer_sizes', [(500,), (500, 300, 200, 150,), (700, 500, 300, 100, ), (1500, 800, 400, 200, )])
- # activation: hp.choice('activation', ['identity', 'logistic', 'tanh' , 'relu'])
- # learning_rate: hp.choice('learning_rate', ['constant', 'invscaling', 'adaptive'])
- # max_iter: scope.int(hp.quniform('max_iter', 100, 800, 1))
- # Stacked_Classifier:
- # stack_method: hp.choice('stack_method', ['auto', 'predict'])
- # passthrough: hp.choice('passthrough', [True, False])
- # Challenger_Stacked_Classifier: mlflow-artifacts:/aacd27cc7a9d429ab2439e37a4e8cdfb/d3eb9ead8a5c49f9b8299efea62d58a2/artifacts/candidate_Stacked_Classifier/model.pkl
- # Challenger_Voting_Classifier: mlflow-artifacts:/cd6eb5acbadd414fb5d32e02d983d99b/b2aea20ef6094e1a94b1567aea90753f/artifacts/candidate_Voting_Classifier/model.pkl
- # Final_Estimator: mlflow-artifacts:/f2a1d0cb47994e43887a7bf8ff11fbc0/4f79ae8649044b24a593ec1c748e012b/artifacts/challenger_LGBMClassifier/model.pkl
- # Champion_Estimator: mlflow-artifacts:/cd6eb5acbadd414fb5d32e02d983d99b/b2aea20ef6094e1a94b1567aea90753f/artifacts/candidate_Voting_Classifier/model.pkl
|