wavetrainer 0.0.6__tar.gz → 0.0.8__tar.gz

{wavetrainer-0.0.6/wavetrainer.egg-info → wavetrainer-0.0.8}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: wavetrainer
-Version: 0.0.6
+Version: 0.0.8
 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-0.0.6 → wavetrainer-0.0.8}/setup.py

@@ -23,7 +23,7 @@ def install_requires() -> typing.List[str]:
 
 setup(
     name='wavetrainer',
-    version='0.0.6',
+    version='0.0.8',
     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-0.0.8/tests/model/catboost_kwargs_test.py

@@ -0,0 +1,28 @@
+"""Tests for the catboost kwargs handler class."""
+import unittest
+
+import pandas as pd
+
+from wavetrainer.model.catboost_kwargs import handle_fit_kwargs
+
+
+class TestCatboostKwargs(unittest.TestCase):
+
+    def test_handle_fit_kwargs(self):
+        x_train = pd.DataFrame(data={
+            "thing": [0.0, 1.0, 2.0, 3.0, 4.0],
+        })
+        x_train["thing"] = x_train["thing"].astype('category')
+        y_train = pd.Series(data=[1.0, 2.0, 3.0, 4.0])
+        x_test = pd.DataFrame(data={
+            "thing": [0.0, 1.0, 2.0, 3.0, 4.0],
+        })
+        x_test["thing"] = x_test["thing"].astype('category')
+        y_test = pd.Series(data=[1.0, 2.0, 3.0, 4.0])
+        args, _ = handle_fit_kwargs(
+            x_train,
+            y_train,
+            eval_set=(x_test, y_test),
+            cat_features=x_train.select_dtypes(include="category").columns.tolist(),
+        )
+        assert len(args) == 2

{wavetrainer-0.0.6 → wavetrainer-0.0.8}/tests/trainer_test.py

@@ -20,10 +20,11 @@ class TestTrainer(unittest.TestCase):
                 data={
                     "column1": x_data,
                     "column2": [(x * random.random()) + random.random() for x in x_data],
-                    "column3": [(x / random.random()) - random.random() for x in x_data],
+                    "column3": [int(((x / random.random()) - random.random()) * 1000.0) for x in x_data],
                 },
                 index=x_index,
             )
+            df["column3"] = df["column3"].astype('category')
             y = pd.DataFrame(
                 data={
                     "y": [x % 2 == 0 for x in x_data],

{wavetrainer-0.0.6 → wavetrainer-0.0.8}/wavetrainer/__init__.py

@@ -2,5 +2,5 @@
 
 from .create import create
 
-__VERSION__ = "0.0.6"
+__VERSION__ = "0.0.8"
 __all__ = ("create",)

{wavetrainer-0.0.6 → wavetrainer-0.0.8}/wavetrainer/calibrator/mapie_calibrator.py

@@ -60,7 +60,7 @@ class MAPIECalibrator(Calibrator):
         try:
             alpha = []
             for potential_alpha in [0.05, 0.32]:
-                if len(df) > int(1.0 / potential_alpha):
+                if len(df) > int(1.0 / potential_alpha) + 1:
                     alpha.append(potential_alpha)
             if alpha:
                 _, y_pis = self._mapie.predict(df, alpha=alpha)

{wavetrainer-0.0.6 → wavetrainer-0.0.8}/wavetrainer/model/catboost_classifier_wrap.py

@@ -11,5 +11,5 @@ class CatBoostClassifierWrapper(CatBoostClassifier):
     """A wrapper for the catboost classifier."""
 
     def fit(self, *args, **kwargs):
-        kwargs = handle_fit_kwargs(*args, **kwargs)
+        args, kwargs = handle_fit_kwargs(*args, **kwargs)
         return super().fit(*args, **kwargs)

wavetrainer-0.0.8/wavetrainer/model/catboost_kwargs.py

@@ -0,0 +1,50 @@
+"""A list of constant catboost kwargs."""
+
+from typing import Any
+
+import numpy as np
+from catboost import Pool  # type: ignore
+
+ORIGINAL_X_ARG_KEY = "original_x"
+EVAL_SET_ARG_KEY = "eval_set"
+CAT_FEATURES_ARG_KEY = "cat_features"
+
+
+def handle_fit_kwargs(*args, **kwargs) -> tuple[tuple[Any, ...], dict[str, Any]]:
+    """Handles keyword args coming into a catboost fit method."""
+    if ORIGINAL_X_ARG_KEY in kwargs:
+        df = kwargs[ORIGINAL_X_ARG_KEY]
+        eval_x, eval_y = kwargs[EVAL_SET_ARG_KEY]
+        cat_features = kwargs[CAT_FEATURES_ARG_KEY]
+        args_list = list(args)
+        fit_x = args_list[0]
+        fix_x_cp = fit_x.copy()
+
+        # Stupid code to ensure eval is feature equivalent to train data
+        included_columns = []
+        for i in range(fix_x_cp.shape[1]):
+            arr_col_values = fix_x_cp[:, i]
+            for col in df.columns:
+                if col in included_columns:
+                    continue
+                df_col_values = df[col].values
+                if np.allclose(df_col_values, arr_col_values, equal_nan=True):
+                    included_columns.append(col)
+                    break
+        # We also need to update cat_features or catboost will yell at us
+        cat_features = list(
+            set(list(kwargs.get(CAT_FEATURES_ARG_KEY, []))) & set(included_columns)
+        )
+        args_list[0] = df[included_columns]
+        args = tuple(args_list)
+
+        eval_x = eval_x[included_columns]
+        kwargs[EVAL_SET_ARG_KEY] = Pool(
+            eval_x,
+            label=eval_y,
+            cat_features=cat_features,
+        )
+        kwargs[CAT_FEATURES_ARG_KEY] = cat_features
+
+        del kwargs[ORIGINAL_X_ARG_KEY]
+    return args, kwargs

{wavetrainer-0.0.6 → wavetrainer-0.0.8}/wavetrainer/model/catboost_model.py

@@ -10,7 +10,8 @@ from catboost import CatBoost, Pool  # type: ignore
 
 from ..model_type import ModelType, determine_model_type
 from .catboost_classifier_wrap import CatBoostClassifierWrapper
-from .catboost_kwargs import EVAL_SET, ORIGINAL_X
+from .catboost_kwargs import (CAT_FEATURES_ARG_KEY, EVAL_SET_ARG_KEY,
+                              ORIGINAL_X_ARG_KEY)
 from .catboost_regressor_wrap import CatBoostRegressorWrapper
 from .model import PREDICTION_COLUMN, PROBABILITY_COLUMN_PREFIX, Model
 
@@ -66,9 +67,9 @@ class CatboostModel(Model):
             raise ValueError("y is null.")
         self._model_type = determine_model_type(y)
         return {
-            EVAL_SET: (eval_x, eval_y),
-            "cat_features": df.select_dtypes(include="category").columns.tolist(),
-            ORIGINAL_X: df,
+            EVAL_SET_ARG_KEY: (eval_x, eval_y),
+            CAT_FEATURES_ARG_KEY: df.select_dtypes(include="category").columns.tolist(),
+            ORIGINAL_X_ARG_KEY: df,
         }
 
     def set_options(self, trial: optuna.Trial | optuna.trial.FrozenTrial) -> None:
@@ -122,6 +123,8 @@ class CatboostModel(Model):
     ) -> Self:
         if y is None:
             raise ValueError("y is null.")
+        if eval_x is None:
+            raise ValueError("eval_x is null.")
         self._model_type = determine_model_type(y)
         catboost = self._provide_catboost()
 
@@ -129,10 +132,12 @@ class CatboostModel(Model):
             df,
             label=y,
             weight=w,
+            cat_features=df.select_dtypes(include="category").columns.tolist(),
         )
         eval_pool = Pool(
             eval_x,
             label=eval_y,
+            cat_features=eval_x.select_dtypes(include="category").columns.tolist(),
         )
         catboost.fit(
             train_pool,

{wavetrainer-0.0.6 → wavetrainer-0.0.8}/wavetrainer/model/catboost_regressor_wrap.py

@@ -9,5 +9,5 @@ class CatBoostRegressorWrapper(CatBoostRegressor):
     """A wrapper for the catboost regressor."""
 
     def fit(self, *args, **kwargs):
-        kwargs = handle_fit_kwargs(*args, **kwargs)
+        args, kwargs = handle_fit_kwargs(*args, **kwargs)
         return super().fit(*args, **kwargs)

{wavetrainer-0.0.6 → wavetrainer-0.0.8/wavetrainer.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: wavetrainer
-Version: 0.0.6
+Version: 0.0.8
 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-0.0.6 → wavetrainer-0.0.8}/wavetrainer.egg-info/SOURCES.txt

@@ -5,6 +5,8 @@ requirements.txt
 setup.py
 tests/__init__.py
 tests/trainer_test.py
+tests/model/__init__.py
+tests/model/catboost_kwargs_test.py
 wavetrainer/__init__.py
 wavetrainer/create.py
 wavetrainer/exceptions.py

wavetrainer-0.0.6/wavetrainer/model/catboost_kwargs.py

@@ -1,35 +0,0 @@
-"""A list of constant catboost kwargs."""
-
-from typing import Any
-
-import numpy as np
-from catboost import Pool  # type: ignore
-
-ORIGINAL_X = "original_x"
-EVAL_SET = "eval_set"
-
-
-def handle_fit_kwargs(*args, **kwargs) -> dict[str, Any]:
-    """Handles keyword args coming into a catboost fit method."""
-    if ORIGINAL_X in kwargs:
-        df = kwargs[ORIGINAL_X]
-        eval_x, eval_y = kwargs[EVAL_SET]
-        fit_x = args[0]
-        fix_x_cp = fit_x.copy()
-
-        # Stupid code to ensure eval is feature equivalent to train data
-        included_columns = []
-        for i in range(fix_x_cp.shape[1]):
-            arr_col_values = fix_x_cp[:, i]
-            for col in df.columns:
-                df_col_values = df[col].values
-                if np.allclose(df_col_values, arr_col_values, equal_nan=True):
-                    included_columns.append(col)
-                    df = df.drop(col, axis=1)
-                    break
-
-        eval_x = eval_x[included_columns]
-        kwargs[EVAL_SET] = Pool(eval_x, label=eval_y)
-
-        del kwargs[ORIGINAL_X]
-    return kwargs