def __init__(self,
obj,
X=None,
X_train=None,
y_pred=None,
y_true=None,
y_train=None,
y_true_train=None,
store_X=False,
prob_return=False,
digits=3):
self.model = obj
self.n_class = np.shape(obj.classes_)[0]
self.labels = obj.classes_
self.variables = obj.n_features_in_
self.priors_weight, self.prior_size = _prior(y_true, digits)
self.labels_pred, self.labels_true, self.y_pred_prob, self.class_weight, self.class_size, self.acc, \
self.prc, self.rcl, self.f1, self.conf, self.y_train, self.y_pred_prob_train, \
self.class_weight_train, self.class_size_train, self.acc_train, \
self.prc_train, self.rcl_train, self.f1_train, self.conf_train, self.ce, self.ce_train, self.y_true_train = _class_pred(
obj, X, X_train, y_pred, y_train, y_true, y_true_train, prob_return, digits)
self.X = _store_X(X, store_X)
self.X_train = _store_X(X_train, store_X)
self.n_estimators = obj.n_estimators
self.objective = obj.objective
self.learning_rate = obj.learning_rate
self.reg_alpha = obj.reg_alpha
self.reg_lambda = obj.reg_lambda
self.feature_importances = _features_important(
obj.feature_importances_, X)
self.min_child_weight = obj.min_child_weight
self.gamma = obj.gamma
self.booster = obj.booster
self.max_delta_step = obj.max_delta_step
def __init__(self,
obj,
X=None,
X_train=None,
y_pred=None,
y_true=None,
y_train=None,
y_true_train=None,
store_X=False,
prob_return=False,
digits=3):
self.model = obj
self.n_class = np.shape(obj.classes_)[0]
self.labels = obj.classes_
self.variables = obj.n_features_
self.priors_weight, self.prior_size = _prior(y_true, digits)
self.labels_pred, self.labels_true, self.y_pred_prob, self.class_weight, self.class_size, self.acc, \
self.prc, self.rcl, self.f1, self.conf, self.y_train, self.y_pred_prob_train, \
self.class_weight_train, self.class_size_train, self.acc_train, \
self.prc_train, self.rcl_train, self.f1_train, self.conf_train, self.ce, self.ce_train, self.y_true_train = _class_pred(
obj, X, X_train, y_pred, y_train, y_true, y_true_train, prob_return, digits)
self.X = _store_X(X, store_X)
self.X_train = _store_X(X_train, store_X)
self.n_estimators = obj.n_estimators
self.base_estimator = obj.base_estimator_
self.criterion = obj.base_estimator_.criterion
self.max_depth = obj.base_estimator_.max_depth
self.max_leaf_nodes = obj.base_estimator_.max_leaf_nodes
self.ccp_alpha = obj.base_estimator_.ccp_alpha
self.min_samples_leaf = obj.base_estimator_.min_samples_leaf
self.min_samples_split = obj.base_estimator_.min_samples_split
self.max_features = obj.max_features
self.bootstrap = obj.bootstrap
self.bootstrap_features = obj.bootstrap_features
def __init__(self,
obj,
X=None,
y_pred=None,
y_true=None,
store_X=False,
digits=3):
self.model = obj
self.n_class = np.shape(obj.classes_)[0]
self.labels = obj.classes_
self.variables = obj.n_features_in_
self.labels_pred = y_pred
self.labels_true = y_true
self.priors_weight, self.prior_size = _prior(y_true, digits)
self.y_pred, self.y_true, self.y_pred_prob, self.class_weight, self.class_size, self.acc, \
self.prc, self.rcl, self.f1, self.conf, self.ce = _class_pred(obj, X, y_pred, y_true, digits)
self.X = _store_X(X, store_X)
self.n_neighbors = obj.n_neighbors
self.radius = obj.radius
self.metric = self.effective_metric_
self.p = obj.p
self.fit_method = obj._fit_method
self.leaf_size = obj.leaf_size
self.p = obj.p
self.contamination = obj.offset_
self.novelty = obj.novelty
self.negative_outlier_factor_ = pd.DataFrame(
obj.negative_outlier_factor_)
def __init__(self,
obj,
X=None,
X_train=None,
y_pred=None,
y_true=None,
y_train=None,
y_true_train=None,
store_X=False,
prob_return=False,
digits=3):
self.model = obj
self.n_class = np.shape(obj.classes_)[0]
self.labels = obj.classes_
self.variables = obj.n_features_in_
self.priors_weight, self.prior_size = _prior(y_true, digits)
self.labels_pred, self.labels_true, self.y_pred_prob, self.class_weight, self.class_size, self.acc, \
self.prc, self.rcl, self.f1, self.conf, self.y_train, self.y_pred_prob_train, \
self.class_weight_train, self.class_size_train, self.acc_train, \
self.prc_train, self.rcl_train, self.f1_train, self.conf_train, self.ce, self.ce_train, self.y_true_train = _class_pred(
obj, X, X_train, y_pred, y_train, y_true, y_true_train, prob_return, digits)
self.X = _store_X(X, store_X)
self.X_train = _store_X(X_train, store_X)
self.n_estimators = obj.n_estimators
self.algorithm = obj.algorithm
self.base_estimator_ = obj.base_estimator_
self.learning_rate = obj.learning_rate
self.estimator_weights_ = obj.estimator_weights_
self.estimator_errors_ = obj.estimator_errors_
self.feature_importances = _features_important(
obj.feature_importances_, X)
def __init__(self,
obj,
X=None,
X_train=None,
y_pred=None,
y_true=None,
y_train=None,
y_true_train=None,
store_X=False,
prob_return=False,
digits=3):
self.model = obj
self.n_class = np.shape(obj.classes_)[0]
self.labels = obj.classes_
self.variables = np.shape(obj.means_)[1]
self.priors_weight, self.prior_size = _prior(y_true, digits)
self.class_prior = pd.Series(obj.priors_)
self.labels_pred, self.labels_true, self.y_pred_prob, self.class_weight, self.class_size, self.acc, \
self.prc, self.rcl, self.f1, self.conf, self.y_train, self.y_pred_prob_train, \
self.class_weight_train, self.class_size_train, self.acc_train, \
self.prc_train, self.rcl_train, self.f1_train, self.conf_train, self.ce, self.ce_train, self.y_true_train = _class_pred(obj, X, X_train, y_pred, y_train, y_true, y_true_train, prob_return, digits)
#self.means = _class_centers(obj.means_)
self.X = _store_X(X, store_X)
self.X_train = _store_X(X_train, store_X)
self.intercept = obj.intercept_
self.coef = obj.coef_
self.solver = obj.solver
self.shrinkage = obj.shrinkage
def __init__(self,
obj,
X=None,
X_train=None,
y_pred=None,
y_true=None,
y_train=None,
y_true_train=None,
store_X=False,
prob_return=False,
digits=3):
self.model = obj
self.n_class = np.shape(obj.classes_)[0]
self.labels = obj.classes_
self.variables = obj.n_features_in_
self.priors_weight, self.prior_size = _prior(y_true, digits)
self.labels_pred, self.labels_true, self.y_pred_prob, self.class_weight, self.class_size, self.acc, \
self.prc, self.rcl, self.f1, self.conf, self.y_train, self.y_pred_prob_train, \
self.class_weight_train, self.class_size_train, self.acc_train, \
self.prc_train, self.rcl_train, self.f1_train, self.conf_train, self.ce, self.ce_train, self.y_true_train = _class_pred(
obj, X, X_train, y_pred, y_train, y_true, y_true_train, prob_return, digits)
self.X = _store_X(X, store_X)
self.X_train = _store_X(X_train, store_X)
self.n_neighbors = obj.n_neighbors
self.radius = obj.radius
self.metric = obj.effective_metric_
self.p = obj.p
self.metric_params = obj.metric_params
self.weights = obj.weights
def __init__(self, obj, X=None, X_train = None, y_pred=None, y_true=None,
y_train = None, y_true_train = None, store_X=False, prob_return=False, digits = 3):
self.model = obj
self.n_class = np.shape(obj.classes_)[0]
self.labels = obj.classes_
self.variables = obj.n_features_in_
self.priors_weight, self.prior_size = _prior(y_true, digits)
self.labels_pred, self.labels_true, self.y_pred_prob, self.class_weight, self.class_size, self.acc, \
self.prc, self.rcl, self.f1, self.conf, self.y_train, self.y_pred_prob_train, \
self.class_weight_train, self.class_size_train, self.acc_train, \
self.prc_train, self.rcl_train, self.f1_train, self.conf_train, self.ce, self.ce_train, self.y_true_train = _class_pred(
obj, X, X_train, y_pred, y_train, y_true, y_true_train, prob_return, digits)
self.centroids = obj.centroids_
self.X = _store_X(X, store_X)
self.X_train = _store_X(X_train, store_X)
self.shrink_threshold = obj.shrink_threshold
self.metric = obj.metric
def __init__(self, obj, X=None, X_train = None, y_pred=None, y_true=None,
y_train = None, y_true_train = None, store_X=False, prob_return=False, digits = 3):
self.model = obj
self.n_class = obj.n_classes
#self.labels = obj.classes_
self.variables = obj._n_features
self.priors_weight, self.prior_size = _prior(y_true, digits)
self.labels_pred, self.labels_true, self.y_pred_prob, self.class_weight, self.class_size, self.acc, \
self.prc, self.rcl, self.f1, self.conf, self.y_train, self.y_pred_prob_train, \
self.class_weight_train, self.class_size_train, self.acc_train, \
self.prc_train, self.rcl_train, self.f1_train, self.conf_train, self.ce, self.ce_train, self.y_true_train = _class_pred(
obj, X, X_train, y_pred, y_train, y_true, y_true_train, prob_return, digits)
self.X = _store_X(X, store_X)
self.X_train = _store_X(X_train, store_X)
self.b = obj.b_
self.w = obj.w_
self.regularization = obj.l2
self.learning_rate = obj.eta
self.epochs = obj.epochs
self.cross_entropy = obj.cost_
self.minibatches = obj.minibatches
def __init__(self, obj, X=None, X_train=None, y_pred=None, y_true=None,
y_train=None, y_true_train=None, store_X=False, prob_return=False, digits=3):
self.model = obj
self.n_class = np.shape(obj.classes_)[0]
self.labels = obj.classes_
self.variables = obj.n_features_in_
self.priors_weight, self.prior_size = _prior(y_true, digits)
self.labels_pred, self.labels_true, self.y_pred_prob, self.class_weight, self.class_size, self.acc, \
self.prc, self.rcl, self.f1, self.conf, self.y_train, self.y_pred_prob_train, \
self.class_weight_train, self.class_size_train, self.acc_train, \
self.prc_train, self.rcl_train, self.f1_train, self.conf_train, self.ce, self.ce_train, self.y_true_train = _class_pred(
obj, X, X_train, y_pred, y_train, y_true, y_true_train, prob_return, digits)
self.X = _store_X(X, store_X)
self.X_train = _store_X(X_train, store_X)
self.C = obj.C
self.kernel = obj.kernel
self.degree = obj.degree
self.gamma = obj.gamma
self.shrinking = obj.shrinking
self.support_vectors_ = pd.DataFrame(obj.support_vectors_)
self.n_support = pd.DataFrame(obj.n_support_)
self.decision_function = obj.decision_function_shape