Improve setup

ceruleo/__init__.py

@@ -9,4 +9,4 @@
 CACHE_PATH.mkdir(parents=True, exist_ok=True)
 
 
-__version__ = "2.0.3"
+__version__ = "2.0.4"

ceruleo/dataset/analysis/numerical_features.py

@@ -4,15 +4,14 @@
 import antropy as ant
 import numpy as np
 import pandas as pd
-
-from pyparsing import col
 from scipy.stats import spearmanr
-from ceruleo.dataset.transformed import TransformedDataset
-from tqdm.auto import tqdm
 from sklearn.feature_selection import mutual_info_regression
+from tqdm.auto import tqdm
+from uncertainties import ufloat
+
+from ceruleo.dataset.transformed import TransformedDataset
 from ceruleo.dataset.ts_dataset import AbstractLivesDataset
 from ceruleo.dataset.utils import iterate_over_features_and_target
-from uncertainties import ufloat
 
 
 def entropy(s: np.ndarray)-> float:

ceruleo/graphics/results.py

@@ -11,13 +11,18 @@
 import matplotlib
 import matplotlib.pyplot as plt
 import numpy as np
-
 import seaborn as sns
+
 from ceruleo.dataset.transformed import TransformedDataset
 from ceruleo.graphics.utils.curly_brace import curlyBrace
-from ceruleo.results.results import (FittedLife, PredictionResult,
-                                     models_cv_results, split_lives,
-                                     unexpected_breaks, unexploited_lifetime)
+from ceruleo.results.results import (
+    FittedLife,
+    PredictionResult,
+    models_cv_results,
+    split_lives,
+    unexpected_breaks,
+    unexploited_lifetime,
+)
 
 
 def plot_lives(ds: TransformedDataset):
@@ -116,7 +121,7 @@ def set_box_color(bp, color):
 
     max_x = np.max(ticks) + 1
     ax.set_xlabel("RUL" + ("" if x_axis_label is None else x_axis_label))
-    ax.set_ylabel("$y - \hat{y}$" + ("" if y_axis_label is None else y_axis_label))
+    ax.set_ylabel(r"$y - \hat{y}$" + ("" if y_axis_label is None else y_axis_label))
     ax.set_xticks(ticks)
     ax.set_xticklabels(labels)
     ax.legend()
@@ -263,7 +268,7 @@ def _cv_barplot_errors_wrt_RUL_multiple_models(
         dx += bar_group_width + group_separation
 
     ax.set_xlabel("RUL" + ("" if x_axis_label is None else x_axis_label))
-    ax.set_ylabel("$y - \hat{y}$" + ("" if y_axis_label is None else y_axis_label))
+    ax.set_ylabel(r"$y - \hat{y}$" + ("" if y_axis_label is None else y_axis_label))
     ax.set_xticks(ticks)
     ax.set_xticklabels(labels)
     ax.legend()
@@ -370,7 +375,7 @@ def _cv_shadedline_plot_errors_wrt_RUL_multiple_models(
         ticks.append(x)
 
     ax.set_xlabel("RUL" + ("" if x_axis_label is None else x_axis_label))
-    ax.set_ylabel("$y - \hat{y}$" + ("" if y_axis_label is None else y_axis_label))
+    ax.set_ylabel(r"$y - \hat{y}$" + ("" if y_axis_label is None else y_axis_label))
     ax.set_xticks(ticks)
     ax.set_xticklabels(labels)
     ax.legend()
@@ -737,7 +742,7 @@ def plot_predictions(
     ax.set_ylabel(units)
     ax.set_xlabel(units)
     legend = ax.legend()
-    for l in legend.legendHandles:
+    for l in legend.legend_handles:
         l.set_markersize(6)
 
 

ceruleo/results/results.py

@@ -26,10 +26,9 @@
 """
 import logging
 from dataclasses import dataclass
-from typing import Callable, Dict, List, Optional, Tuple, Union
+from typing import  Dict, List, Optional, Tuple, Union
 
 import numpy as np
-import pandas as pd
 from ceruleo.results.picewise_regression import (PiecewesieLinearFunction,
                                                 PiecewiseLinearRegression)
 from sklearn.metrics import mean_absolute_error as mae

ceruleo/transformation/features/denoising.py

@@ -10,14 +10,13 @@
 class SavitzkyGolayTransformer(TransformerStep):
     """Filter each feature using LOESS
 
-    Parameters:    
+    Parameters:
         window: window size of the filter
         order:  Order of the filter, by default 2
         name: Step name
     """
 
     def __init__(self, window: int, order: int = 2, name: Optional[str] = None):
-
         super().__init__(name=name)
         self.window = window
         self.order = order
@@ -66,16 +65,17 @@ def __init__(
         min_periods: int = 15,
         name: Optional[str] = None,
     ):
-
         super().__init__(name=name)
         self.window = window
         self.min_periods = min_periods
         self.center = center
 
     def transform(self, X: pd.DataFrame, y=None) -> pd.DataFrame:
         return X.rolling(
-            self.window, min_periods=self.min_periods, center=self.center
-        ).mean(skip_na=True)
+            self.window,
+            min_periods=self.min_periods,
+            center=self.center,
+        ).mean(numeric_only=True)
 
 
 class MedianFilter(TransformerStep):
@@ -97,7 +97,9 @@ def __init__(self, window: int, min_periods: int = 15, name: Optional[str] = Non
         self.min_periods = min_periods
 
     def transform(self, X: pd.DataFrame, y=None) -> pd.DataFrame:
-        return X.rolling(self.window, min_periods=self.min_periods).median(skip_na=True)
+        return X.rolling(self.window, min_periods=self.min_periods).median(
+            numeric_only=True
+        )
 
 
 class OneDimensionalKMeans(TransformerStep):
@@ -117,7 +119,9 @@ def __init__(self, n_clusters: int = 5, name: Optional[str] = None):
     def partial_fit(self, X):
         if len(self.clusters) == 0:
             for c in X.columns:
-                self.clusters[c] = MiniBatchKMeans(n_clusters=self.n_clusters)
+                self.clusters[c] = MiniBatchKMeans(
+                    n_clusters=self.n_clusters, n_init="auto"
+                )
 
         for c in X.columns:
             self.clusters[c].partial_fit(np.atleast_2d(X[c]).T)
@@ -160,7 +164,7 @@ class MultiDimensionalKMeans(TransformerStep):
     def __init__(self, n_clusters: int = 5, name: Optional[str] = None):
         super().__init__(name=name)
         self.n_clusters = n_clusters
-        self.clusters = MiniBatchKMeans(n_clusters=self.n_clusters)
+        self.clusters = MiniBatchKMeans(n_clusters=self.n_clusters, n_init="auto")
 
     def partial_fit(self, X):
         self.clusters.partial_fit(X)
@@ -203,7 +207,7 @@ def __init__(self, span: float, name: Optional[str] = None):
         self.span = span
 
     def transform(self, X: pd.DataFrame, y=None) -> pd.DataFrame:
-        return X.ewm(span=self.span).mean(skip_na=True)
+        return X.ewm(span=self.span, ignore_na=True).mean()
 
 
 class GaussianFilter(TransformerStep):
@@ -228,7 +232,7 @@ def __init__(
         min_points: int = 1,
         center: bool = False,
         *args,
-        **kwargs
+        **kwargs,
     ):
         super().__init__(**kwargs)
         self.window_size = window_size