wavetrainer 0.0.31__tar.gz → 0.0.32__tar.gz

{wavetrainer-0.0.31/wavetrainer.egg-info → wavetrainer-0.0.32}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: wavetrainer
-Version: 0.0.31
+Version: 0.0.32
 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-0.0.31 → wavetrainer-0.0.32}/setup.py

@@ -23,7 +23,7 @@ def install_requires() -> typing.List[str]:
 
 setup(
     name='wavetrainer',
-    version='0.0.31',
+    version='0.0.32',
     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-0.0.31 → wavetrainer-0.0.32}/wavetrainer/__init__.py

@@ -2,5 +2,5 @@
 
 from .create import create
 
-__VERSION__ = "0.0.31"
+__VERSION__ = "0.0.32"
 __all__ = ("create",)

{wavetrainer-0.0.31 → wavetrainer-0.0.32}/wavetrainer/reducer/combined_reducer.py

@@ -18,6 +18,7 @@ from .unseen_reducer import UnseenReducer
 
 _COMBINED_REDUCER_FILE = "combined_reducer.json"
 _REDUCERS_KEY = "reducers"
+_REMOVED_COLUMNS_FILE = "combined_reducer_removed_columns.json"
 
 
 class CombinedReducer(Reducer):
@@ -35,6 +36,7 @@ class CombinedReducer(Reducer):
             CorrelationReducer(),
             SmartCorrelationReducer(),
         ]
+        self._folder = None
 
     @classmethod
     def name(cls) -> str:
@@ -67,6 +69,7 @@ class CombinedReducer(Reducer):
                     self._reducers.append(SmartCorrelationReducer())
         for reducer in self._reducers:
             reducer.load(folder)
+        self._folder = folder
 
     def save(self, folder: str, trial: optuna.Trial | optuna.trial.FrozenTrial) -> None:
         with open(
@@ -89,17 +92,19 @@ class CombinedReducer(Reducer):
         eval_x: pd.DataFrame | None = None,
         eval_y: pd.Series | pd.DataFrame | None = None,
     ) -> Self:
+        removed_columns_dict = {}
         for reducer in self._reducers:
             before_columns = set(df.columns.values)
             df = reducer.fit_transform(df)
             after_columns = set(df.columns.values)
             removed_columns = before_columns.difference(after_columns)
             if removed_columns:
-                logging.info(
-                    "Removed columns %s using %s",
-                    ",".join(removed_columns),
-                    reducer.name(),
-                )
+                removed_columns_dict[reducer.name()] = list(removed_columns)
+        if self._folder is not None:
+            with open(
+                os.path.join(self._folder, _REMOVED_COLUMNS_FILE), encoding="utf8"
+            ) as handle:
+                json.dump(removed_columns_dict, handle)
         return self
 
     def transform(self, df: pd.DataFrame) -> pd.DataFrame:

{wavetrainer-0.0.31 → wavetrainer-0.0.32}/wavetrainer/trainer.py

@@ -37,7 +37,7 @@ _TEST_SIZE_KEY = "test_size"
 _VALIDATION_SIZE_KEY = "validation_size"
 _IDX_USR_ATTR_KEY = "idx"
 _DT_COLUMN_KEY = "dt_column"
-_BAD_OUTPUT = -1.0
+_BAD_OUTPUT = -1000.0
 
 
 def _assign_bin(timestamp, bins: list[datetime.datetime]) -> int:
@@ -345,7 +345,10 @@ class Trainer(Fit):
                     if self._max_train_timeout is None
                     else self._max_train_timeout.total_seconds(),
                 )
-            while study.best_trial.value is None or study.best_trial.value != _BAD_OUTPUT:
+            while (
+                study.best_trial.value is None or study.best_trial.value == _BAD_OUTPUT
+            ):
+                logging.info("Performing extra train")
                 study.optimize(
                     test_objective,
                     n_trials=1,
@@ -387,7 +390,7 @@ class Trainer(Fit):
 
                 test_df = df.iloc[: train_len + count + test_len]
                 test_series = y_series.iloc[: train_len + count + test_len]
-                if len(test_df) <= 2:
+                if len(test_df) <= 3:
                     continue
 
                 if test_idx < start_validation_index:

{wavetrainer-0.0.31 → wavetrainer-0.0.32/wavetrainer.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: wavetrainer
-Version: 0.0.31
+Version: 0.0.32
 Summary: A library for automatically finding the optimal model within feature and hyperparameter space.
 Home-page: https://github.com/8W9aG/wavetrainer
 Author: Will Sackfield