Observable report

pyproject.toml

@@ -21,7 +21,7 @@ dependencies = [
     "polarstate==0.1.8",
 ]
 name = "rtichoke"
-version = "0.1.13"
+version = "0.1.14"
 description = "interactive visualizations for performance of predictive models"
 readme = "README.md"
 

src/rtichoke/discrimination/gains.py

@@ -1,20 +1,23 @@
 """
-A module for Gains Curves
+A module for Gains Curves using Plotly helpers
 """
 
-from typing import Dict, List, Optional
-from pandas import DataFrame
+from typing import Dict, List, Sequence, Union
 from plotly.graph_objs._figure import Figure
-from rtichoke.helpers.send_post_request_to_r_rtichoke import create_rtichoke_curve
-from rtichoke.helpers.send_post_request_to_r_rtichoke import plot_rtichoke_curve
+from rtichoke.helpers.plotly_helper_functions import (
+    _create_rtichoke_plotly_curve_binary,
+    _plot_rtichoke_curve_binary,
+)
+import numpy as np
+import polars as pl
 
 
 def create_gains_curve(
-    probs: Dict[str, List[float]],
-    reals: Dict[str, List[int]],
+    probs: Dict[str, np.ndarray],
+    reals: Union[np.ndarray, Dict[str, np.ndarray]],
     by: float = 0.01,
-    stratified_by: str = "probability_threshold",
-    size: Optional[int] = None,
+    stratified_by: Sequence[str] = ["probability_threshold"],
+    size: int = 600,
     color_values: List[str] = [
         "#1b9e77",
         "#d95f02",
@@ -37,78 +40,86 @@ def create_gains_curve(
         "#D1603D",
         "#585123",
     ],
-    url_api: str = "http://localhost:4242/",
 ) -> Figure:
-    """Create Gains Curve
+    """Create Gains Curve.
 
-    Args:
-        probs (Dict[str, List[float]]): _description_
-        reals (Dict[str, List[int]]): _description_
-        by (float, optional): _description_. Defaults to 0.01.
-        stratified_by (str, optional): _description_. Defaults to "probability_threshold".
-        size (Optional[int], optional): _description_. Defaults to None.
-        color_values (List[str], optional): _description_. Defaults to None.
-        url_api (_type_, optional): _description_. Defaults to "http://localhost:4242/".
+    Parameters
+    ----------
+    probs : Dict[str, np.ndarray]
+        Dictionary mapping a label or group name to an array of predicted
+        probabilities for the positive class.
+    reals : Union[np.ndarray, Dict[str, np.ndarray]]
+        Ground-truth binary labels (0/1) as a single array, or a dictionary
+        mapping the same label/group keys used in ``probs`` to arrays of
+        ground-truth labels.
+    by : float, optional
+        Resolution for probability thresholds when computing the curve
+        (step size). Default is 0.01.
+    stratified_by : Sequence[str], optional
+        Sequence of column names to stratify the performance data by.
+        Default is ["probability_threshold"].
+    size : int, optional
+        Plot size in pixels (width and height). Default is 600.
+    color_values : List[str], optional
+        List of color hex strings to use for the plotted lines. If not
+        provided, a default palette is used.
 
-    Returns:
-        Figure: _description_
+    Returns
+    -------
+    Figure
+        A Plotly ``Figure`` containing the Gains curve(s).
+
+    Notes
+    -----
+    The function delegates computation and plotting to
+    ``_create_rtichoke_plotly_curve_binary`` and returns the resulting
+    Plotly figure.
     """
-    fig = create_rtichoke_curve(
+    fig = _create_rtichoke_plotly_curve_binary(
         probs,
         reals,
         by=by,
         stratified_by=stratified_by,
         size=size,
         color_values=color_values,
-        url_api=url_api,
         curve="gains",
     )
     return fig
 
 
 def plot_gains_curve(
-    performance_data: DataFrame,
-    size: Optional[int] = None,
-    color_values: List[str] = [
-        "#1b9e77",
-        "#d95f02",
-        "#7570b3",
-        "#e7298a",
-        "#07004D",
-        "#E6AB02",
-        "#FE5F55",
-        "#54494B",
-        "#006E90",
-        "#BC96E6",
-        "#52050A",
-        "#1F271B",
-        "#BE7C4D",
-        "#63768D",
-        "#08A045",
-        "#320A28",
-        "#82FF9E",
-        "#2176FF",
-        "#D1603D",
-        "#585123",
-    ],
-    url_api: str = "http://localhost:4242/",
+    performance_data: pl.DataFrame,
+    stratified_by: Sequence[str] = ["probability_threshold"],
+    size: int = 600,
 ) -> Figure:
-    """Plot Gains Curve
+    """Plot Gains curve from performance data.
+
+    Parameters
+    ----------
+    performance_data : pl.DataFrame
+        A Polars DataFrame containing performance metrics for the Gains curve.
+        Expected columns include (but may not be limited to)
+        ``probability_threshold`` and gains-related metrics, plus any
+        stratification columns.
+    stratified_by : Sequence[str], optional
+        Sequence of column names used for stratification in the
+        ``performance_data``. Default is ["probability_threshold"].
+    size : int, optional
+        Plot size in pixels (width and height). Default is 600.
 
-    Args:
-        performance_data (DataFrame): _description_
-        size (Optional[int], optional): _description_. Defaults to None.
-        color_values (List[str], optional): _description_. Defaults to None.
-        url_api (_type_, optional): _description_. Defaults to "http://localhost:4242/".
+    Returns
+    -------
+    Figure
+        A Plotly ``Figure`` containing the Gains plot.
 
-    Returns:
-        Figure: _description_
+    Notes
+    -----
+    This function wraps ``_plot_rtichoke_curve_binary`` to produce a
+    ready-to-render Plotly figure from precomputed performance data.
     """
-    fig = plot_rtichoke_curve(
+    fig = _plot_rtichoke_curve_binary(
         performance_data,
         size=size,
-        color_values=color_values,
-        url_api=url_api,
         curve="gains",
     )
     return fig

src/rtichoke/discrimination/lift.py

@@ -1,20 +1,23 @@
 """
-A module for Lift Curves
+A module for Lift Curves using Plotly helpers
 """
 
-from typing import Dict, List, Optional
+from typing import Dict, List, Sequence, Union
 from plotly.graph_objs._figure import Figure
-from pandas import DataFrame
-from rtichoke.helpers.send_post_request_to_r_rtichoke import create_rtichoke_curve
-from rtichoke.helpers.send_post_request_to_r_rtichoke import plot_rtichoke_curve
+from rtichoke.helpers.plotly_helper_functions import (
+    _create_rtichoke_plotly_curve_binary,
+    _plot_rtichoke_curve_binary,
+)
+import numpy as np
+import polars as pl
 
 
 def create_lift_curve(
-    probs: Dict[str, List[float]],
-    reals: Dict[str, List[int]],
+    probs: Dict[str, np.ndarray],
+    reals: Union[np.ndarray, Dict[str, np.ndarray]],
     by: float = 0.01,
-    stratified_by: str = "probability_threshold",
-    size: Optional[int] = None,
+    stratified_by: Sequence[str] = ["probability_threshold"],
+    size: int = 600,
     color_values: List[str] = [
         "#1b9e77",
         "#d95f02",
@@ -37,78 +40,86 @@ def create_lift_curve(
         "#D1603D",
         "#585123",
     ],
-    url_api: str = "http://localhost:4242/",
 ) -> Figure:
-    """Create Lift Curve
+    """Create Lift Curve.
 
-    Args:
-        probs (Dict[str, List[float]]): _description_
-        reals (Dict[str, List[int]]): _description_
-        by (float, optional): _description_. Defaults to 0.01.
-        stratified_by (str, optional): _description_. Defaults to "probability_threshold".
-        size (Optional[int], optional): _description_. Defaults to None.
-        color_values (List[str], optional): _description_. Defaults to None.
-        url_api (_type_, optional): _description_. Defaults to "http://localhost:4242/".
+    Parameters
+    ----------
+    probs : Dict[str, np.ndarray]
+        Dictionary mapping a label or group name to an array of predicted
+        probabilities for the positive class.
+    reals : Union[np.ndarray, Dict[str, np.ndarray]]
+        Ground-truth binary labels (0/1) as a single array, or a dictionary
+        mapping the same label/group keys used in ``probs`` to arrays of
+        ground-truth labels.
+    by : float, optional
+        Resolution for probability thresholds when computing the curve
+        (step size). Default is 0.01.
+    stratified_by : Sequence[str], optional
+        Sequence of column names to stratify the performance data by.
+        Default is ["probability_threshold"].
+    size : int, optional
+        Plot size in pixels (width and height). Default is 600.
+    color_values : List[str], optional
+        List of color hex strings to use for the plotted lines. If not
+        provided, a default palette is used.
 
-    Returns:
-        Figure: _description_
+    Returns
+    -------
+    Figure
+        A Plotly ``Figure`` containing the Lift curve(s).
+
+    Notes
+    -----
+    The function delegates computation and plotting to
+    ``_create_rtichoke_plotly_curve_binary`` and returns the resulting
+    Plotly figure.
     """
-    fig = create_rtichoke_curve(
+    fig = _create_rtichoke_plotly_curve_binary(
         probs,
         reals,
         by=by,
         stratified_by=stratified_by,
         size=size,
         color_values=color_values,
-        url_api=url_api,
         curve="lift",
     )
     return fig
 
 
 def plot_lift_curve(
-    performance_data: DataFrame,
-    size: Optional[int] = None,
-    color_values: List[str] = [
-        "#1b9e77",
-        "#d95f02",
-        "#7570b3",
-        "#e7298a",
-        "#07004D",
-        "#E6AB02",
-        "#FE5F55",
-        "#54494B",
-        "#006E90",
-        "#BC96E6",
-        "#52050A",
-        "#1F271B",
-        "#BE7C4D",
-        "#63768D",
-        "#08A045",
-        "#320A28",
-        "#82FF9E",
-        "#2176FF",
-        "#D1603D",
-        "#585123",
-    ],
-    url_api: str = "http://localhost:4242/",
+    performance_data: pl.DataFrame,
+    stratified_by: Sequence[str] = ["probability_threshold"],
+    size: int = 600,
 ) -> Figure:
-    """Plot Lift Curve
+    """Plot Lift curve from performance data.
+
+    Parameters
+    ----------
+    performance_data : pl.DataFrame
+        A Polars DataFrame containing performance metrics for the Lift curve.
+        Expected columns include (but may not be limited to)
+        ``probability_threshold`` and lift-related metrics, plus any
+        stratification columns.
+    stratified_by : Sequence[str], optional
+        Sequence of column names used for stratification in the
+        ``performance_data``. Default is ["probability_threshold"].
+    size : int, optional
+        Plot size in pixels (width and height). Default is 600.
 
-    Args:
-        performance_data (DataFrame): _description_
-        size (Optional[int], optional): _description_. Defaults to None.
-        color_values (List[str], optional): _description_. Defaults to None.
-        url_api (_type_, optional): _description_. Defaults to "http://localhost:4242/".
+    Returns
+    -------
+    Figure
+        A Plotly ``Figure`` containing the Lift plot.
 
-    Returns:
-        Figure: _description_
+    Notes
+    -----
+    This function wraps ``_plot_rtichoke_curve_binary`` to produce a
+    ready-to-render Plotly figure from precomputed performance data.
     """
-    fig = plot_rtichoke_curve(
+    fig = _plot_rtichoke_curve_binary(
         performance_data,
         size=size,
-        color_values=color_values,
-        url_api=url_api,
         curve="lift",
     )
     return fig