Merge pull request #29 from EchoDel/feature/package_updates

Updated LightGBM to 4.1 and Optuna to 3.3, and the other packages

lightgbmlss/logger.py

@@ -0,0 +1,17 @@
+import logging
+
+
+class CustomLogger:
+    def __init__(self):
+        self.logger = logging.getLogger('lightgbm_custom')
+        self.logger.setLevel(logging.ERROR)
+
+    def info(self, message):
+        self.logger.info(message)
+
+    def warning(self, message):
+        # Suppress warnings by not doing anything
+        pass
+
+    def error(self, message):
+        self.logger.error(message)

lightgbmlss/model.py

@@ -9,6 +9,9 @@
 
 
 import lightgbm as lgb
+
+from lightgbmlss.distributions.distribution_utils import DistributionClass
+from lightgbmlss.logger import CustomLogger
 from lightgbmlss.utils import *
 import optuna
 from optuna.samplers import TPESampler
@@ -47,6 +50,7 @@
     Tuple[Dataset, Dataset, Dict[str, Any]]
 ]
 
+lgb.register_logger(CustomLogger())
 
 class LightGBMLSS:
     """
@@ -59,7 +63,7 @@ class LightGBMLSS:
      start_values : np.ndarray
         Starting values for each distributional parameter.
     """
-    def __init__(self, dist):
+    def __init__(self, dist: DistributionClass):
         self.dist = dist             # Distribution object
         self.start_values = None     # Starting values for distributional parameters
 
@@ -79,9 +83,9 @@ def set_params(self, params: Dict[str, Any]) -> Dict[str, Any]:
         """
         params_adj = {"num_class": self.dist.n_dist_param,
                       "metric": "None",
-                      "objective": "None",
                       "random_seed": 123,
-                      "verbose": -1}
+                      "verbose": -1,
+                      "objective": self.dist.objective_fn}
         params.update(params_adj)
 
         return params
@@ -171,7 +175,6 @@ def train(self,
         self.booster = lgb.train(params,
                                  train_set,
                                  num_boost_round=num_boost_round,
-                                 fobj=self.dist.objective_fn,
                                  feval=self.dist.metric_fn,
                                  valid_sets=valid_sets,
                                  valid_names=valid_names,
@@ -265,7 +268,6 @@ def cv(self,
 
         self.bstLSS_cv = lgb.cv(params,
                                 train_set,
-                                fobj=self.dist.objective_fn,
                                 feval=self.dist.metric_fn,
                                 num_boost_round=num_boost_round,
                                 folds=folds,
@@ -389,13 +391,13 @@ def objective(trial):
                                           callbacks=[pruning_callback, early_stopping_callback],
                                           seed=seed,
                                           )
-
+            print(lgblss_param_tuning)
             # Extract the optimal number of boosting rounds
-            opt_rounds = np.argmin(np.array(lgblss_param_tuning[f"{self.dist.loss_fn}-mean"])) + 1
+            opt_rounds = np.argmin(np.array(lgblss_param_tuning[f"valid {self.dist.loss_fn}-mean"])) + 1
             trial.set_user_attr("opt_round", int(opt_rounds))
 
             # Extract the best score
-            best_score = np.min(np.array(lgblss_param_tuning[f"{self.dist.loss_fn}-mean"]))
+            best_score = np.min(np.array(lgblss_param_tuning[f"valid {self.dist.loss_fn}-mean"]))
 
             return best_score
 

setup.py

@@ -16,20 +16,20 @@
     zip_safe=True,
     python_requires=">=3.9",
     install_requires=[
-        "lightgbm~=3.3.5",
-        "torch~=2.0.1",
-        "pyro-ppl~=1.8.5",
-        "optuna~=3.2.0",
+        "lightgbm~=4.1.0",
+        "torch~=2.1.1",
+        "pyro-ppl~=1.8.6",
+        "optuna~=3.4.0",
         "properscoring~=0.1",
-        "scikit-learn~=1.2.2",
-        "numpy~=1.24.3",
-        "pandas~=2.0.3",
-        "plotnine~=0.12.1",
-        "scipy~=1.11.1",
-        "seaborn~=0.12.2",
-        "tqdm~=4.65.0",
-        "matplotlib~=3.7.2",
-        "ipython~=8.14.0",
+        "scikit-learn~=1.3.2",
+        "numpy~=1.26.2",
+        "pandas~=2.1.3",
+        "plotnine~=0.12.4",
+        "scipy~=1.11.4",
+        "seaborn~=0.13.0",
+        "tqdm~=4.66.1",
+        "matplotlib~=3.8.2",
+        "ipython~=8.18.1",
     ],
     extras_require={
         "docs": ["mkdocs", "mkdocstrings[python]", "mkdocs-jupyter"]