change glmtree classifier default predict function from pred proba to decision function; version 0.2.7

Author: ZebinYang

examples/demo.ipynb

@@ -5,8 +5,8 @@
    "execution_count": null,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-12-27T11:25:01.467037Z",
-     "start_time": "2021-12-27T11:24:59.706592Z"
+     "end_time": "2021-12-27T11:41:27.070305Z",
+     "start_time": "2021-12-27T11:41:25.800526Z"
     }
    },
    "outputs": [],
@@ -32,8 +32,8 @@
    "execution_count": null,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-12-27T11:25:01.500112Z",
-     "start_time": "2021-12-27T11:25:01.469302Z"
+     "end_time": "2021-12-27T11:41:27.095134Z",
+     "start_time": "2021-12-27T11:41:27.071798Z"
     }
    },
    "outputs": [],
@@ -48,8 +48,8 @@
    "execution_count": null,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-12-27T11:25:50.669194Z",
-     "start_time": "2021-12-27T11:25:01.501343Z"
+     "end_time": "2021-12-27T11:43:04.465481Z",
+     "start_time": "2021-12-27T11:41:27.096046Z"
     },
     "scrolled": true
    },
@@ -71,8 +71,8 @@
    "execution_count": null,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-12-27T11:25:50.670794Z",
-     "start_time": "2021-12-27T11:25:50.670779Z"
+     "end_time": "2021-12-27T11:43:05.991216Z",
+     "start_time": "2021-12-27T11:43:04.467280Z"
     }
    },
    "outputs": [],
@@ -97,8 +97,8 @@
    "execution_count": null,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-12-27T11:25:50.671677Z",
-     "start_time": "2021-12-27T11:25:50.671662Z"
+     "end_time": "2021-12-27T11:43:05.992967Z",
+     "start_time": "2021-12-27T11:43:05.992948Z"
     }
    },
    "outputs": [],
@@ -113,8 +113,8 @@
    "execution_count": null,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-12-27T11:25:50.672528Z",
-     "start_time": "2021-12-27T11:25:50.672513Z"
+     "end_time": "2021-12-27T11:43:05.994028Z",
+     "start_time": "2021-12-27T11:43:05.994010Z"
     }
    },
    "outputs": [],
@@ -135,8 +135,8 @@
    "execution_count": null,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-12-27T11:25:50.673375Z",
-     "start_time": "2021-12-27T11:25:50.673360Z"
+     "end_time": "2021-12-27T11:43:05.995089Z",
+     "start_time": "2021-12-27T11:43:05.995071Z"
     }
    },
    "outputs": [],

simtree/glmtree.py

@@ -43,7 +43,7 @@ def build_leaf(self, sample_indice):
         nx = (self.x[sample_indice] - mx) / sx
 
         if len(self.reg_lambda) > 1:
-            best_estimator = LassoCV(alphas=self.reg_lambda, cv=5, n_jobs=self.n_jobs, precompute=False, random_state=self.random_state)
+            best_estimator = LassoCV(alphas=self.reg_lambda, cv=5, precompute=False, random_state=self.random_state)
             best_estimator.fit(nx, self.y[sample_indice], self.sample_weight[sample_indice])
         else:
             if self.reg_lambda[0] > 0:
@@ -93,7 +93,7 @@ def build_leaf(self, sample_indice):
         else:
             if len(self.reg_lambda) > 1:
                 best_estimator = LogisticRegressionCV(Cs=self.reg_lambda, penalty="l1", solver="liblinear", scoring="roc_auc",
-                                      cv=5, n_jobs=self.n_jobs, random_state=self.random_state)
+                                      cv=5, random_state=self.random_state)
             else:
                 best_estimator = LogisticRegression(alpha=self.reg_lambda[0], precompute=False, random_state=self.random_state)
 

simtree/mobtree.py

@@ -256,7 +256,7 @@ def fit(self, x, y):
         if self.split_features is None:
             self.split_features = np.arange(n_features).tolist()
 
-        if self.n_feature_search > len(self.split_features) and (self.max_depth >= 1):
+        if self.n_feature_search > len(self.split_features) and (self.max_depth >= 0):
             self.important_split_features = self.split_features
         else:
             self.important_split_features = self.screen_features(sample_indice)