wavetrainer 0.0.10__tar.gz → 0.0.12__tar.gz

{wavetrainer-0.0.10/wavetrainer.egg-info → wavetrainer-0.0.12}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: wavetrainer
-Version: 0.0.10
+Version: 0.0.12
 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-0.0.10 → wavetrainer-0.0.12}/setup.py

@@ -23,7 +23,7 @@ def install_requires() -> typing.List[str]:
 
 setup(
     name='wavetrainer',
-    version='0.0.10',
+    version='0.0.12',
     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-0.0.10 → wavetrainer-0.0.12}/wavetrainer/__init__.py

@@ -2,5 +2,5 @@
 
 from .create import create
 
-__VERSION__ = "0.0.10"
+__VERSION__ = "0.0.12"
 __all__ = ("create",)

{wavetrainer-0.0.10 → wavetrainer-0.0.12}/wavetrainer/create.py

@@ -1,5 +1,7 @@
 """A function for creating a new trainer."""
 
+# pylint: disable=too-many-arguments,too-many-positional-arguments
+
 import datetime
 
 from .trainer import Trainer
@@ -11,6 +13,8 @@ def create(
     test_size: float | datetime.timedelta | None = None,
     validation_size: float | datetime.timedelta | None = None,
     dt_column: str | None = None,
+    max_train_timeout: datetime.timedelta | None = None,
+    cutoff_dt: datetime.datetime | None = None,
 ) -> Trainer:
     """Create a trainer."""
     return Trainer(
@@ -19,4 +23,6 @@ def create(
         test_size=test_size,
         validation_size=validation_size,
         dt_column=dt_column,
+        max_train_timeout=max_train_timeout,
+        cutoff_dt=cutoff_dt,
     )

{wavetrainer-0.0.10 → wavetrainer-0.0.12}/wavetrainer/trainer.py

@@ -41,7 +41,7 @@ _DT_COLUMN_KEY = "dt_column"
 class Trainer(Fit):
     """A class for training and predicting from an array of data."""
 
-    # pylint: disable=too-many-arguments,too-many-positional-arguments,too-many-statements,too-many-locals,too-many-branches
+    # pylint: disable=too-many-arguments,too-many-positional-arguments,too-many-statements,too-many-locals,too-many-branches,too-many-instance-attributes
 
     def __init__(
         self,
@@ -51,6 +51,8 @@ class Trainer(Fit):
         test_size: float | datetime.timedelta | None = None,
         validation_size: float | datetime.timedelta | None = None,
         dt_column: str | None = None,
+        max_train_timeout: datetime.timedelta | None = None,
+        cutoff_dt: datetime.datetime | None = None,
     ):
         tqdm.tqdm.pandas()
 
@@ -64,6 +66,8 @@ class Trainer(Fit):
             test_size = 0.15
         if validation_size is None:
             validation_size = 0.15
+        if cutoff_dt is None:
+            cutoff_dt = datetime.datetime.now()
 
         params_file = os.path.join(self._folder, _PARAMS_FILENAME)
         if walkforward_timedelta is None:
@@ -137,6 +141,8 @@ class Trainer(Fit):
         self._test_size = test_size
         self._validation_size = validation_size
         self._dt_column = dt_column
+        self._max_train_timeout = max_train_timeout
+        self._cutoff_dt = cutoff_dt
 
     def _provide_study(self, column: str) -> optuna.Study:
         storage_name = f"sqlite:///{self._folder}/{column}/{_STUDYDB_FILENAME}"
@@ -170,6 +176,8 @@ class Trainer(Fit):
             if self._dt_column is None
             else pd.DatetimeIndex(pd.to_datetime(df[self._dt_column]))
         )
+        df = df[dt_index < self._cutoff_dt]  # type: ignore
+        y = y.iloc[: len(df)]
 
         def _fit_column(y_series: pd.Series):
             column_dir = os.path.join(self._folder, str(y_series.name))
@@ -185,6 +193,7 @@ class Trainer(Fit):
                 save: bool,
                 split_idx: datetime.datetime,
             ) -> float:
+                print(f"Beginning trial for: {split_idx.isoformat()}")
                 trial.set_user_attr(_IDX_USR_ATTR_KEY, split_idx.isoformat())
 
                 train_dt_index = dt_index[: len(x)]
@@ -289,7 +298,12 @@ class Trainer(Fit):
                 initial_trials = max(self._trials - len(study.trials), 0)
             if initial_trials > 0:
                 study.optimize(
-                    test_objective, n_trials=initial_trials, show_progress_bar=True
+                    test_objective,
+                    n_trials=initial_trials,
+                    show_progress_bar=True,
+                    timeout=None
+                    if self._max_train_timeout is None
+                    else self._max_train_timeout.total_seconds(),
                 )
 
             train_len = len(df[dt_index < start_test_index])
@@ -319,6 +333,7 @@ class Trainer(Fit):
                         found = True
                         break
                 if found:
+                    last_processed_dt = test_dt
                     continue
 
                 test_df = df.iloc[: train_len + count + test_len]
@@ -336,6 +351,9 @@ class Trainer(Fit):
                             validate_objctive, idx=test_idx, series=test_series
                         ),
                         n_trials=1,
+                        timeout=None
+                        if self._max_train_timeout is None
+                        else self._max_train_timeout.total_seconds(),
                     )
 
                 _fit(study.best_trial, test_df, test_series, True, test_idx)

{wavetrainer-0.0.10 → wavetrainer-0.0.12/wavetrainer.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: wavetrainer
-Version: 0.0.10
+Version: 0.0.12
 Summary: A library for automatically finding the optimal model within feature and hyperparameter space.
 Home-page: https://github.com/8W9aG/wavetrainer
 Author: Will Sackfield