validmind 2.7.5__py3-none-any.whl → 2.7.7__py3-none-any.whl

validmind/tests/__types__.py

@@ -39,6 +39,7 @@ TestID = Union[
         "validmind.data_validation.LaggedCorrelationHeatmap",
         "validmind.data_validation.MissingValues",
         "validmind.data_validation.MissingValuesBarPlot",
+        "validmind.data_validation.MutualInformation",
         "validmind.data_validation.PearsonCorrelationMatrix",
         "validmind.data_validation.PhillipsPerronArch",
         "validmind.data_validation.ProtectedClassesCombination",
@@ -48,6 +49,7 @@ TestID = Union[
         "validmind.data_validation.RollingStatsPlot",
         "validmind.data_validation.RunsTest",
         "validmind.data_validation.ScatterPlot",
+        "validmind.data_validation.ScoreBandDefaultRates",
         "validmind.data_validation.SeasonalDecompose",
         "validmind.data_validation.ShapiroWilk",
         "validmind.data_validation.Skewness",
@@ -121,7 +123,9 @@ TestID = Union[
         "validmind.model_validation.ragas.SemanticSimilarity",
         "validmind.model_validation.sklearn.AdjustedMutualInformation",
         "validmind.model_validation.sklearn.AdjustedRandIndex",
+        "validmind.model_validation.sklearn.CalibrationCurve",
         "validmind.model_validation.sklearn.ClassifierPerformance",
+        "validmind.model_validation.sklearn.ClassifierThresholdOptimization",
         "validmind.model_validation.sklearn.ClusterCosineSimilarity",
         "validmind.model_validation.sklearn.ClusterPerformanceMetrics",
         "validmind.model_validation.sklearn.CompletenessScore",
@@ -134,6 +138,7 @@ TestID = Union[
         "validmind.model_validation.sklearn.MinimumAccuracy",
         "validmind.model_validation.sklearn.MinimumF1Score",
         "validmind.model_validation.sklearn.MinimumROCAUCScore",
+        "validmind.model_validation.sklearn.ModelParameters",
         "validmind.model_validation.sklearn.ModelsPerformanceComparison",
         "validmind.model_validation.sklearn.OverfitDiagnosis",
         "validmind.model_validation.sklearn.PermutationFeatureImportance",
@@ -147,6 +152,7 @@ TestID = Union[
         "validmind.model_validation.sklearn.RegressionR2SquareComparison",
         "validmind.model_validation.sklearn.RobustnessDiagnosis",
         "validmind.model_validation.sklearn.SHAPGlobalImportance",
+        "validmind.model_validation.sklearn.ScoreProbabilityAlignment",
         "validmind.model_validation.sklearn.SilhouettePlot",
         "validmind.model_validation.sklearn.TrainingTestDegradation",
         "validmind.model_validation.sklearn.VMeasure",
@@ -166,9 +172,20 @@ TestID = Union[
         "validmind.model_validation.statsmodels.RegressionModelSummary",
         "validmind.model_validation.statsmodels.RegressionPermutationFeatureImportance",
         "validmind.model_validation.statsmodels.ScorecardHistogram",
+        "validmind.ongoing_monitoring.CalibrationCurveDrift",
+        "validmind.ongoing_monitoring.ClassDiscriminationDrift",
+        "validmind.ongoing_monitoring.ClassImbalanceDrift",
+        "validmind.ongoing_monitoring.ClassificationAccuracyDrift",
+        "validmind.ongoing_monitoring.ConfusionMatrixDrift",
+        "validmind.ongoing_monitoring.CumulativePredictionProbabilitiesDrift",
         "validmind.ongoing_monitoring.FeatureDrift",
         "validmind.ongoing_monitoring.PredictionAcrossEachFeature",
         "validmind.ongoing_monitoring.PredictionCorrelation",
+        "validmind.ongoing_monitoring.PredictionProbabilitiesHistogramDrift",
+        "validmind.ongoing_monitoring.PredictionQuantilesAcrossFeatures",
+        "validmind.ongoing_monitoring.ROCCurveDrift",
+        "validmind.ongoing_monitoring.ScoreBandsDrift",
+        "validmind.ongoing_monitoring.ScorecardHistogramDrift",
         "validmind.ongoing_monitoring.TargetPredictionDistributionPlot",
         "validmind.prompt_validation.Bias",
         "validmind.prompt_validation.Clarity",

validmind/tests/data_validation/ACFandPACFPlot.py

@@ -6,7 +6,7 @@ import pandas as pd
 import plotly.graph_objects as go
 from statsmodels.tsa.stattools import acf, pacf
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.vm_models import VMDataset
 
 
@@ -62,6 +62,8 @@ def ACFandPACFPlot(dataset: VMDataset):
         raise ValueError("Provided 'columns' must exist in the dataset")
 
     figures = []
+    acf_store = {}
+    pacf_store = {}
     for col in df.columns:
         series = df[col]
         max_lags = min(40, len(series) // 2 - 1)
@@ -77,6 +79,7 @@ def ACFandPACFPlot(dataset: VMDataset):
             font=dict(size=18),
         )
         figures.append(acf_fig)
+        acf_store[col] = acf_values
 
         # Create PACF plot using Plotly
         pacf_values = pacf(series, nlags=max_lags)
@@ -89,5 +92,6 @@ def ACFandPACFPlot(dataset: VMDataset):
             font=dict(size=18),
         )
         figures.append(pacf_fig)
+        pacf_store[col] = pacf_values
 
-    return tuple(figures)
+    return (*figures, RawData(acf_values=acf_store, pacf_values=pacf_store))

validmind/tests/data_validation/AutoMA.py

@@ -6,7 +6,7 @@ import pandas as pd
 from statsmodels.tsa.arima.model import ARIMA
 from statsmodels.tsa.stattools import adfuller
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.logging import get_logger
 from validmind.vm_models import VMDataset
 
@@ -116,4 +116,4 @@ def AutoMA(dataset: VMDataset, max_ma_order: int = 3):
     return {
         "Auto MA Analysis Results": summary_ma_analysis,
         "Best MA Order Results": best_ma_order,
-    }
+    }, RawData(raw_series_data=df)

validmind/tests/data_validation/BivariateScatterPlots.py

@@ -6,7 +6,7 @@ import itertools
 
 import plotly.express as px
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 
 
 @tags("tabular_data", "numerical_data", "visualization")
@@ -79,4 +79,6 @@ def BivariateScatterPlots(dataset):
 
         figures.append(fig)
 
-    return tuple(figures)
+    return tuple(figures) + (
+        RawData(selected_numerical_df=df, feature_pairs=features_pairs),
+    )

validmind/tests/data_validation/BoxPierce.py

@@ -5,7 +5,7 @@
 import pandas as pd
 from statsmodels.stats.diagnostic import acorr_ljungbox
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 
 
 @tasks("regression")
@@ -68,4 +68,4 @@ def BoxPierce(dataset):
     box_pierce_df.reset_index(inplace=True)
     box_pierce_df.columns = ["column", "stat", "pvalue"]
 
-    return box_pierce_df
+    return box_pierce_df, RawData(box_pierce_values=box_pierce_values)

validmind/tests/data_validation/ClassImbalance.py

@@ -9,7 +9,7 @@ from typing import Any, Dict, Tuple
 
 import plotly.graph_objs as go
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.errors import SkipTestError
 from validmind.vm_models import VMDataset
 
@@ -104,4 +104,5 @@ def ClassImbalance(
         },
         go.Figure(data=[trace], layout=layout),
         all(row["Pass/Fail"] == "Pass" for row in imbalanced_classes),
+        RawData(imbalance_percentages=imbalance_percentages),
     )

validmind/tests/data_validation/DatasetDescription.py

@@ -9,7 +9,7 @@ import numpy as np
 from ydata_profiling.config import Settings
 from ydata_profiling.model.typeset import ProfilingTypeSet
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.errors import UnsupportedColumnTypeError
 from validmind.logging import get_logger
 from validmind.vm_models import VMDataset
@@ -220,6 +220,15 @@ def DatasetDescription(dataset: VMDataset):
     for column in infer_datatypes(df):
         results.append(describe_column(df, column))
 
+    raw_data = {
+        column["id"]: {
+            "type": column["type"],
+            "statistics": column["statistics"],
+            "histograms": column["histograms"],
+        }
+        for column in results
+    }
+
     return {
         "Dataset Description": [
             {
@@ -233,4 +242,4 @@ def DatasetDescription(dataset: VMDataset):
             }
             for column in results
         ]
-    }
+    }, RawData(raw_data=raw_data)

validmind/tests/data_validation/DatasetSplit.py

@@ -4,7 +4,7 @@
 
 from typing import List
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.vm_models import VMDataset
 
 DATASET_LABELS = {
@@ -98,4 +98,4 @@ def DatasetSplit(datasets: List[VMDataset]):
                 }
             )
 
-    return table
+    return table, RawData(dataset_results=results)

validmind/tests/data_validation/DickeyFullerGLS.py

@@ -6,7 +6,7 @@ import pandas as pd
 from arch.unitroot import DFGLS
 from numpy.linalg import LinAlgError
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.errors import SkipTestError
 from validmind.logging import get_logger
 from validmind.vm_models import VMDataset
@@ -97,4 +97,4 @@ def DickeyFullerGLS(dataset: VMDataset):
 
     return {
         "DFGLS Test Results": dfgls_values,
-    }
+    }, RawData(df=df)

validmind/tests/data_validation/FeatureTargetCorrelationPlot.py

@@ -6,7 +6,7 @@
 import numpy as np
 import plotly.graph_objects as go
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 
 
 @tags("tabular_data", "visualization", "correlation")
@@ -58,7 +58,13 @@ def FeatureTargetCorrelationPlot(dataset, fig_height=600):
 
     fig = _visualize_feature_target_correlation(df, dataset.target_column, fig_height)
 
-    return fig
+    correlations = (
+        df.corr(numeric_only=True)[dataset.target_column]
+        .drop(dataset.target_column)
+        .to_frame()
+    )
+
+    return fig, RawData(correlation_data=correlations)
 
 
 def _visualize_feature_target_correlation(df, target_column, fig_height):

validmind/tests/data_validation/HighCardinality.py

@@ -2,7 +2,7 @@
 # See the LICENSE file in the root of this repository for details.
 # SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.vm_models import VMDataset
 
 
@@ -59,6 +59,8 @@ def HighCardinality(
     table = []
     all_passed = True
 
+    raw_data = {}
+
     for col in dataset.feature_columns_categorical:
         n_distinct = df[col].nunique()
         p_distinct = n_distinct / df.shape[0]
@@ -73,7 +75,12 @@ def HighCardinality(
             }
         )
 
+        raw_data[col] = {
+            "n_distinct": n_distinct,
+            "p_distinct": p_distinct,
+        }
+
         if not passed:
             all_passed = False
 
-    return table, all_passed
+    return table, all_passed, RawData(raw_cardinality_details=raw_data)

validmind/tests/data_validation/HighPearsonCorrelation.py

@@ -2,14 +2,17 @@
 # See the LICENSE file in the root of this repository for details.
 # SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.vm_models import VMDataset
 
 
 @tags("tabular_data", "data_quality", "correlation")
 @tasks("classification", "regression")
 def HighPearsonCorrelation(
-    dataset: VMDataset, max_threshold: float = 0.3, top_n_correlations: int = 10
+    dataset: VMDataset,
+    max_threshold: float = 0.3,
+    top_n_correlations: int = 10,
+    feature_columns: list = None,
 ):
     """
     Identifies highly correlated feature pairs in a dataset suggesting feature redundancy or multicollinearity.
@@ -51,8 +54,15 @@ def HighPearsonCorrelation(
     - Limited to identifying redundancy only within feature pairs; may fail to spot more complex relationships among
     three or more variables.
     """
+
+    # Select features
+    if feature_columns is None:
+        df = dataset.df
+    else:
+        df = dataset.df[feature_columns]
+
     # Get correlation matrix for numeric columns
-    corr = dataset.df.corr(numeric_only=True)
+    corr = df.corr(numeric_only=True)
 
     # Create table of correlation coefficients and column pairs
     pairs = []
@@ -71,4 +81,8 @@ def HighPearsonCorrelation(
     pairs.sort(key=lambda x: abs(x["Coefficient"]), reverse=True)
     pairs = pairs[:top_n_correlations]
 
-    return pairs, all(p["Pass/Fail"] == "Pass" for p in pairs)
+    return (
+        pairs,
+        all(p["Pass/Fail"] == "Pass" for p in pairs),
+        RawData(correlation_matrix=corr),
+    )

validmind/tests/data_validation/IQROutliersBarPlot.py

@@ -4,7 +4,7 @@
 
 import plotly.graph_objects as go
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.vm_models import VMDataset
 
 
@@ -118,4 +118,11 @@ def IQROutliersBarPlot(
         )
         figures.append(fig)
 
-    return tuple(figures)
+    return (
+        *figures,
+        RawData(
+            outlier_counts_by_feature=df[dataset.feature_columns_numeric].apply(
+                lambda col: compute_outliers(col, threshold)
+            )
+        ),
+    )

validmind/tests/data_validation/LaggedCorrelationHeatmap.py

@@ -6,7 +6,7 @@ import numpy as np
 import pandas as pd
 import plotly.figure_factory as ff
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.vm_models import VMDataset
 
 # Define the 'coolwarm' color scale manually
@@ -101,4 +101,4 @@ def LaggedCorrelationHeatmap(dataset: VMDataset, num_lags: int = 10):
         xaxis_title="Lags",
     )
 
-    return fig
+    return fig, RawData(correlation_matrix=correlation_df)

validmind/tests/data_validation/MissingValuesBarPlot.py

@@ -4,7 +4,7 @@
 
 import plotly.graph_objects as go
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.vm_models import VMDataset
 
 
@@ -106,13 +106,16 @@ def MissingValuesBarPlot(
         line=dict(color="red", dash="dash"),
     )
 
-    return go.Figure(
-        data=[trace_below_threshold, trace_above_threshold, threshold_line],
-        layout=go.Layout(
-            title="Missing Values",
-            yaxis=dict(title="Columns"),
-            xaxis=dict(title="Missing Value Percentage (%)", range=[0, 100]),
-            barmode="stack",
-            height=fig_height,
+    return (
+        go.Figure(
+            data=[trace_below_threshold, trace_above_threshold, threshold_line],
+            layout=go.Layout(
+                title="Missing Values",
+                yaxis=dict(title="Columns"),
+                xaxis=dict(title="Missing Value Percentage (%)", range=[0, 100]),
+                barmode="stack",
+                height=fig_height,
+            ),
         ),
+        RawData(missing_percentages=missing_percentages_sorted),
     )

validmind/tests/data_validation/MutualInformation.py

@@ -4,6 +4,7 @@
 
 import plotly.graph_objects as go
 from sklearn.feature_selection import mutual_info_classif, mutual_info_regression
+
 from validmind import tags, tasks
 from validmind.vm_models import VMDataset
 from validmind.vm_models.result import RawData
@@ -76,13 +77,6 @@ def MutualInformation(
     else:
         mi_scores = mutual_info_regression(X, y)
 
-    # Create DataFrame for raw data
-    raw_data = RawData(
-        feature=dataset.feature_columns,
-        mutual_information_score=mi_scores.tolist(),
-        pass_fail=["Pass" if score >= min_threshold else "Fail" for score in mi_scores],
-    )
-
     # Create Plotly figure
     fig = go.Figure()
 
@@ -126,4 +120,8 @@ def MutualInformation(
         template="plotly_white",
     )
 
-    return raw_data, fig
+    return fig, RawData(
+        mutual_information_scores={
+            feature: score for feature, score in zip(sorted_features, sorted_scores)
+        }
+    )

validmind/tests/data_validation/PearsonCorrelationMatrix.py

@@ -5,7 +5,7 @@
 
 import plotly.graph_objects as go
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 
 
 @tags("tabular_data", "numerical_data", "correlation")
@@ -88,4 +88,4 @@ def PearsonCorrelationMatrix(dataset):
 
     fig = go.Figure(data=[heatmap], layout=layout)
 
-    return fig
+    return fig, RawData(correlation_matrix=corr_matrix)

validmind/tests/data_validation/ProtectedClassesCombination.py

@@ -8,7 +8,7 @@ import pandas as pd
 import plotly.graph_objects as go
 import plotly.subplots as sp
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.errors import MissingDependencyError
 from validmind.logging import get_logger
 
@@ -202,4 +202,9 @@ def ProtectedClassesCombination(dataset, model, protected_classes=None):
         {"Class Combination Table": metrics_by_group},
         {"DPR and EOR table": dpr_eor_df},
         fig,
+        RawData(
+            metrics_frame=mf,
+            demographic_parity_ratios=m_dpr,
+            equalized_odds_ratios=m_eqo,
+        ),
     )

validmind/tests/data_validation/ProtectedClassesDescription.py

@@ -127,4 +127,4 @@ def ProtectedClassesDescription(dataset, protected_classes=None):
         ["Protected Class", "Count"], ascending=[True, False]
     )
 
-    return (stats_df, *tuple(figures))
+    return (stats_df, *figures)

validmind/tests/data_validation/ProtectedClassesDisparity.py

@@ -119,7 +119,7 @@ def ProtectedClassesDisparity(
         mask_significance=True,
     )
 
-    plots = []
+    figures = []
     for protected_class in protected_classes:
         plot = ap.disparity(
             bdf, metrics, protected_class, fairness_threshold=disparity_tolerance
@@ -129,13 +129,12 @@ def ProtectedClassesDisparity(
         plot.save(
             buf, format="png"
         )  # as long as the above library is installed, this will work
-        plots.append(buf.getvalue())
+        figures.append(buf.getvalue())
 
     string = "_disparity"
     metrics_adj = [x + string for x in metrics]
 
     table = bdf[["attribute_name", "attribute_value"] + b.list_disparities(bdf)]
-    plots.append(aqp.plot_disparity_all(bdf, metrics=metrics_adj))
-    plots_return = tuple(plots)
+    figures.append(aqp.plot_disparity_all(bdf, metrics=metrics_adj))
 
-    return (table, *plots_return)
+    return (table, *figures)

validmind/tests/data_validation/ProtectedClassesThresholdOptimizer.py

@@ -103,10 +103,7 @@ def ProtectedClassesThresholdOptimizer(
         test_df, target, y_pred_opt, protected_classes
     )
 
-    return (
-        {"DPR and EOR Table": fairness_metrics.reset_index()},
-        fig,
-    )
+    return {"DPR and EOR Table": fairness_metrics.reset_index()}, fig
 
 
 def initialize_and_fit_optimizer(pipeline, X_train, y_train, protected_classes_df):

validmind/tests/data_validation/RollingStatsPlot.py

@@ -5,7 +5,7 @@
 import matplotlib.pyplot as plt
 import pandas as pd
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.errors import SkipTestError
 from validmind.vm_models import VMDataset
 
@@ -95,13 +95,24 @@ def RollingStatsPlot(dataset: VMDataset, window_size: int = 12):
     if not pd.api.types.is_datetime64_any_dtype(dataset.df.index):
         raise SkipTestError("Index must be a datetime type")
 
-    return tuple(
-        [
-            plot_rolling_statistics(
-                df=dataset.df.dropna(),
-                col=col,
-                window_size=window_size,
-            )
-            for col in dataset.feature_columns
-        ]
+    figures = [
+        plot_rolling_statistics(
+            df=dataset.df.dropna(),
+            col=col,
+            window_size=window_size,
+        )
+        for col in dataset.feature_columns
+    ]
+
+    return (
+        *figures,
+        RawData(
+            rolling_means_stds={
+                col: {
+                    "rolling_mean": dataset.df[col].rolling(window=window_size).mean(),
+                    "rolling_std": dataset.df[col].rolling(window=window_size).std(),
+                }
+                for col in dataset.feature_columns
+            }
+        ),
     )

validmind/tests/data_validation/ScatterPlot.py

@@ -55,8 +55,8 @@ def ScatterPlot(dataset):
     - Assumes that the dataset can fit into the computer's memory, which might not be valid for extremely large
     datasets.
     """
-
     g = sns.pairplot(data=dataset.df, diag_kind="kde")
+
     for ax in g.axes.flatten():
         # rotate x axis labels
         ax.set_xlabel(ax.get_xlabel(), rotation=45)
@@ -64,12 +64,10 @@ def ScatterPlot(dataset):
         ax.set_ylabel(ax.get_ylabel(), rotation=45)
         # set y labels alignment
         ax.yaxis.get_label().set_horizontalalignment("right")
+
     # Get the current figure
     fig = plt.gcf()
 
-    figures = []
-    figures.append(fig)
-
     plt.close("all")
 
-    return tuple(figures)
+    return fig

validmind/tests/data_validation/ScoreBandDefaultRates.py

@@ -2,8 +2,9 @@
 # See the LICENSE file in the root of this repository for details.
 # SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial
 
-import pandas as pd
 import numpy as np
+import pandas as pd
+
 from validmind import tags, tasks
 from validmind.vm_models import VMDataset, VMModel
 

validmind/tests/data_validation/SeasonalDecompose.py

@@ -9,7 +9,7 @@ from plotly.subplots import make_subplots
 from scipy import stats
 from statsmodels.tsa.seasonal import seasonal_decompose
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.errors import SkipTestError
 from validmind.logging import get_logger
 from validmind.vm_models import VMDataset
@@ -65,6 +65,8 @@ def SeasonalDecompose(dataset: VMDataset, seasonal_model: str = "additive"):
 
     figures = []
 
+    raw_data = {}
+
     for col in df.columns:
         series = df[col].dropna()
 
@@ -153,7 +155,15 @@ def SeasonalDecompose(dataset: VMDataset, seasonal_model: str = "additive"):
 
         figures.append(fig)
 
+        # Add the decomposed components to raw_data
+        raw_data[col] = {
+            "observed": sd.observed,
+            "trend": sd.trend,
+            "seasonal": sd.seasonal,
+            "residuals": sd.resid,
+        }
+
     if not figures:
         raise SkipTestError("No valid features found for seasonal decomposition")
 
-    return tuple(figures)
+    return (*figures, RawData(decomposed_components=raw_data))

validmind/tests/data_validation/Skewness.py

@@ -73,6 +73,9 @@ def Skewness(dataset, max_threshold=1):
             }
         )
 
-    return {
-        "Skewness Results for Dataset": results_table,
-    }, passed
+    return (
+        {
+            "Skewness Results for Dataset": results_table,
+        },
+        passed,
+    )