wavetrainer 0.0.18__tar.gz → 0.0.20__tar.gz

{wavetrainer-0.0.18/wavetrainer.egg-info → wavetrainer-0.0.20}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: wavetrainer
-Version: 0.0.18
+Version: 0.0.20
 Summary: A library for automatically finding the optimal model within feature and hyperparameter space.
 Home-page: https://github.com/8W9aG/wavetrainer
 Author: Will Sackfield

{wavetrainer-0.0.18 → wavetrainer-0.0.20}/setup.py

@@ -23,7 +23,7 @@ def install_requires() -> typing.List[str]:
 
 setup(
     name='wavetrainer',
-    version='0.0.18',
+    version='0.0.20',
     description='A library for automatically finding the optimal model within feature and hyperparameter space.',
     long_description=long_description,
     long_description_content_type='text/markdown',

{wavetrainer-0.0.18 → wavetrainer-0.0.20}/wavetrainer/__init__.py

@@ -2,5 +2,5 @@
 
 from .create import create
 
-__VERSION__ = "0.0.18"
+__VERSION__ = "0.0.20"
 __all__ = ("create",)

{wavetrainer-0.0.18 → wavetrainer-0.0.20}/wavetrainer/model/catboost_model.py

@@ -1,6 +1,7 @@
 """A model that wraps catboost."""
 
 import json
+import logging
 import os
 from typing import Any, Self
 
@@ -152,6 +153,8 @@ class CatboostModel(Model):
             metric_period=100,
             eval_set=eval_pool,
         )
+        importances = catboost.get_feature_importance(prettified=True)
+        logging.info("Importances:\n%s", importances)
         return self
 
     def transform(self, df: pd.DataFrame) -> pd.DataFrame:

{wavetrainer-0.0.18 → wavetrainer-0.0.20}/wavetrainer/selector/selector.py

@@ -66,14 +66,6 @@ class Selector(Params, Fit):
         )
         try:
             self._selector.fit(df, y=y, sample_weight=w, **model_kwargs)
-            importances = self._model.estimator.feature_importances_
-            try:
-                importances_len = len(importances)
-                columns = self._selector.get_feature_names_out()
-                for i in range(importances_len):
-                    logging.info("Feature %s: %f", columns[i], importances[i])
-            except TypeError:
-                pass
         except ValueError as exc:
             # Catch issues with 1 feature as a reduction target.
             logging.warning(str(exc))

{wavetrainer-0.0.18 → wavetrainer-0.0.20}/wavetrainer/trainer.py

@@ -284,7 +284,9 @@ class Trainer(Fit):
             start_test_index = (
                 test_dt_index.to_list()[-int(len(test_dt_index) * self._test_size)]
                 if isinstance(self._test_size, float)
-                else test_dt_index[test_dt_index >= self._test_size][0]
+                else test_dt_index[
+                    test_dt_index >= (start_validation_index - self._test_size)  # type: ignore
+                ].to_list()[0]
             )
 
             def test_objective(trial: optuna.Trial) -> float:

{wavetrainer-0.0.18 → wavetrainer-0.0.20/wavetrainer.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: wavetrainer
-Version: 0.0.18
+Version: 0.0.20
 Summary: A library for automatically finding the optimal model within feature and hyperparameter space.
 Home-page: https://github.com/8W9aG/wavetrainer
 Author: Will Sackfield