Psychometric-AI-Core/run_config.json at main · wayne810901/Psychometric-AI-Core · 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
{
  "created_at": "2026-01-23T14:12:01.070247",
  "models": {
    "SVM_LinearSVC": "Pipeline(steps=[('scaler', StandardScaler()),\n                ('clf',\n                 LinearSVC(class_weight='balanced', max_iter=8000,\n                           random_state=42))])",
    "SVM_linear": "Pipeline(steps=[('scaler', StandardScaler()),\n                ('clf', SVC(C=2000, class_weight='balanced', kernel='linear'))])",
    "SVM_polynomial": "Pipeline(steps=[('scaler', StandardScaler()),\n                ('clf', SVC(class_weight='balanced', kernel='poly'))])",
    "SVM_RBF": "Pipeline(steps=[('scaler', StandardScaler()),\n                ('clf', SVC(C=1000, class_weight='balanced', gamma=0.1))])",
    "SVM_sigmoid": "Pipeline(steps=[('scaler', StandardScaler()),\n                ('clf', SVC(class_weight='balanced', kernel='sigmoid'))])",
    "NaiveBayes": "Pipeline(steps=[('scaler', StandardScaler()), ('clf', GaussianNB())])",
    "LogisticRegression": "Pipeline(steps=[('scaler', StandardScaler()),\n                ('clf',\n                 LogisticRegression(class_weight='balanced', max_iter=3000,\n                                    random_state=42))])",
    "DecisionTree": "Pipeline(steps=[('clf',\n                 DecisionTreeClassifier(class_weight='balanced',\n                                        random_state=42))])",
    "RandomForest": "Pipeline(steps=[('clf',\n                 RandomForestClassifier(class_weight='balanced_subsample',\n                                        n_estimators=600, n_jobs=-1,\n                                        random_state=42))])",
    "GradientBoosting": "Pipeline(steps=[('clf', GradientBoostingClassifier(random_state=42))])",
    "KNN": "Pipeline(steps=[('scaler', StandardScaler()),\n                ('clf',\n                 KNeighborsClassifier(n_neighbors=15, weights='distance'))])",
    "XGBoost": "Pipeline(steps=[('clf',\n                 XGBClassifier(base_score=None, booster=None, callbacks=None,\n                               colsample_bylevel=None, colsample_bynode=None,\n                               colsample_bytree=0.9, device=None,\n                               early_stopping_rounds=None,\n                               enable_categorical=False, eval_metric='mlogloss',\n                               feature_types=None, feature_weights=None,\n                               gamma=None, grow_policy=None,\n                               importance_type=None,\n                               interaction_constraints=None, learning_rate=0.05,\n                               max_bin=None, max_cat_threshold=None,\n                               max_cat_to_onehot=None, max_delta_step=None,\n                               max_depth=6, max_leaves=None,\n                               min_child_weight=None, missing=nan,\n                               monotone_constraints=None, multi_strategy=None,\n                               n_estimators=600, n_jobs=-1,\n                               num_parallel_tree=None, ...))])",
    "LightGBM": "Pipeline(steps=[('clf',\n                 LGBMClassifier(colsample_bytree=0.9, learning_rate=0.05,\n                                n_estimators=800, n_jobs=-1,\n                                objective='multiclass', random_state=42,\n                                subsample=0.9))])"
  },
  "best_model_name": "SVM_linear",
  "baseline_models": {
    "LPA_posterior_scoring_baseline": {
      "estimator": "LPAScoringClassifier()",
      "cv_result": {
        "acc_mean": 0.9354458607161105,
        "acc_sd": 0.0023005400972295194,
        "f1_macro_mean": 0.895830811500324,
        "f1_macro_sd": 0.004441067130885221,
        "prec_macro_mean": 0.8906752608084358,
        "prec_macro_sd": 0.003306660873889857,
        "rec_macro_mean": 0.9087907382023868,
        "rec_macro_sd": 0.006556489593492045
      },
      "description": "Posterior assignment using class priors and Gaussian class-conditional densities."
    }
  },
  "label_encoder": "LabelEncoder()",
  "feature_cols": [
    "R",
    "I",
    "A",
    "S",
    "E",
    "C"
  ],
  "cv_config": {
    "cv_type": "StratifiedKFold",
    "n_splits": 5,
    "shuffle": true,
    "random_state": 42,
    "n_jobs": -1
  },
  "metrics": {
    "scoring": [
      "accuracy",
      "f1_macro",
      "precision_macro",
      "recall_macro"
    ]
  },
  "data_config": {
    "id_col": "No",
    "target_col": "Group",
    "dropna": true,
    "n_rows": 32221,
    "n_features": 6,
    "n_classes": 72
  },
  "svm_fixed_params": {
    "SVM_linear": {
      "kernel": "linear",
      "C": 2000,
      "class_weight": "balanced"
    },
    "SVM_RBF": {
      "kernel": "rbf",
      "C": 1000,
      "gamma": 0.1,
      "class_weight": "balanced"
    }
  }
}