wavetrainer 0.0.3__tar.gz → 0.0.4__tar.gz

{wavetrainer-0.0.3/wavetrainer.egg-info → wavetrainer-0.0.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: wavetrainer
-Version: 0.0.3
+Version: 0.0.4
 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-0.0.3 → wavetrainer-0.0.4}/setup.py

@@ -23,7 +23,7 @@ def install_requires() -> typing.List[str]:
 
 setup(
     name='wavetrainer',
-    version='0.0.3',
+    version='0.0.4',
     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-0.0.3 → wavetrainer-0.0.4}/wavetrainer/__init__.py

@@ -3,7 +3,7 @@
 from .create import create
 from .load import load
 
-__VERSION__ = "0.0.3"
+__VERSION__ = "0.0.4"
 __all__ = (
     "create",
     "load",

{wavetrainer-0.0.3 → wavetrainer-0.0.4}/wavetrainer/reducer/combined_reducer.py

@@ -10,6 +10,7 @@ import pandas as pd
 from .constant_reducer import ConstantReducer
 from .correlation_reducer import CorrelationReducer
 from .duplicate_reducer import DuplicateReducer
+from .nonnumeric_reducer import NonNumericReducer
 from .reducer import Reducer
 
 _COMBINED_REDUCER_FILE = "combined_reducer.json"
@@ -21,7 +22,12 @@ class CombinedReducer(Reducer):
 
     def __init__(self):
         super().__init__()
-        self._reducers = [ConstantReducer(), DuplicateReducer(), CorrelationReducer()]
+        self._reducers = [
+            ConstantReducer(),
+            DuplicateReducer(),
+            CorrelationReducer(),
+            NonNumericReducer(),
+        ]
 
     @classmethod
     def name(cls) -> str:
@@ -44,6 +50,8 @@ class CombinedReducer(Reducer):
                     self._reducers.append(DuplicateReducer())
                 elif reducer_name == CorrelationReducer.name():
                     self._reducers.append(CorrelationReducer())
+                elif reducer_name == NonNumericReducer.name():
+                    self._reducers.append(NonNumericReducer())
         for reducer in self._reducers:
             reducer.load(folder)
 
@@ -67,7 +75,7 @@ class CombinedReducer(Reducer):
         w: pd.Series | None = None,
     ) -> Self:
         for reducer in self._reducers:
-            reducer.fit(df)
+            df = reducer.fit_transform(df)
         return self
 
     def transform(self, df: pd.DataFrame) -> pd.DataFrame:

wavetrainer-0.0.4/wavetrainer/reducer/nonnumeric_reducer.py

@@ -0,0 +1,39 @@
+"""A reducer that removes non-numeric columns."""
+
+from typing import Self
+
+import optuna
+import pandas as pd
+
+from .reducer import Reducer
+
+
+class NonNumericReducer(Reducer):
+    """A class that removes non numeric columns from a dataframe."""
+
+    @classmethod
+    def name(cls) -> str:
+        return "nonnumeric"
+
+    def set_options(self, trial: optuna.Trial | optuna.trial.FrozenTrial) -> None:
+        pass
+
+    def load(self, folder: str) -> None:
+        pass
+
+    def save(self, folder: str) -> None:
+        pass
+
+    def fit(
+        self,
+        df: pd.DataFrame,
+        y: pd.Series | pd.DataFrame | None = None,
+        w: pd.Series | None = None,
+    ) -> Self:
+        return self
+
+    def transform(self, df: pd.DataFrame) -> pd.DataFrame:
+        categorical_cols = df.select_dtypes(include="category").columns.tolist()
+        numeric_cols = df.select_dtypes(include="number").columns.tolist()
+        keep_cols = categorical_cols + numeric_cols
+        return df[keep_cols]

{wavetrainer-0.0.3 → wavetrainer-0.0.4}/wavetrainer/trainer.py

@@ -252,14 +252,16 @@ class Trainer(Fit):
             start_validation_index = (
                 dt_index[-int(len(dt_index) * self._validation_size) - 1]
                 if isinstance(self._validation_size, float)
-                else dt_index[dt_index >= self._validation_size][0]
+                else dt_index[
+                    dt_index >= (dt_index.to_list()[-1] - self._validation_size)
+                ].to_list()[0]
             )
             test_df = df[dt_index < start_validation_index]
             test_dt_index = (
                 test_df.index if self._dt_column is None else test_df[self._dt_column]
             )
             start_test_index = (
-                test_dt_index[-int(len(test_dt_index) * self._test_size)]
+                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]
             )

{wavetrainer-0.0.3 → wavetrainer-0.0.4}/wavetrainer/weights/class_weights.py

@@ -47,7 +47,7 @@ class ClassWeights(Weights):
             self._class_weights = {}
             return self
 
-        arr = df.astype(int).to_numpy().flatten().astype(float)
+        arr = y.astype(int).to_numpy().flatten().astype(float)
         unique_vals = np.unique(arr)
         w_arr = compute_class_weight(
             class_weight="balanced", classes=unique_vals, y=arr

{wavetrainer-0.0.3 → wavetrainer-0.0.4}/wavetrainer/windower/windower.py

@@ -52,7 +52,9 @@ class Windower(Params, Fit):
         lookback_ratio = self._lookback_ratio
         if lookback_ratio is None:
             raise ValueError("lookback_ratio is null")
-        dt_index = df.index if self._dt_column is None else df[self._dt_column]
+        dt_index = (
+            df.index if self._dt_column is None else df[self._dt_column].to_list()
+        )
         start_idx = dt_index[int(len(df) * lookback_ratio)]
         end_idx = dt_index[-1]
         td = end_idx.to_pydatetime() - start_idx.to_pydatetime()
@@ -66,5 +68,6 @@ class Windower(Params, Fit):
         dt_index = df.index if self._dt_column is None else df[self._dt_column]
         return df[
             dt_index
-            >= dt_index[-1].to_pydatetime() - datetime.timedelta(seconds=lookback)
+            >= dt_index.to_list()[-1].to_pydatetime()
+            - datetime.timedelta(seconds=lookback)
         ]

{wavetrainer-0.0.3 → wavetrainer-0.0.4/wavetrainer.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: wavetrainer
-Version: 0.0.3
+Version: 0.0.4
 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-0.0.3 → wavetrainer-0.0.4}/wavetrainer.egg-info/SOURCES.txt

@@ -34,6 +34,7 @@ wavetrainer/reducer/combined_reducer.py
 wavetrainer/reducer/constant_reducer.py
 wavetrainer/reducer/correlation_reducer.py
 wavetrainer/reducer/duplicate_reducer.py
+wavetrainer/reducer/nonnumeric_reducer.py
 wavetrainer/reducer/reducer.py
 wavetrainer/selector/__init__.py
 wavetrainer/selector/selector.py