validmind 2.7.6__py3-none-any.whl → 2.7.7__py3-none-any.whl

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,
+    )

validmind/tests/data_validation/SpreadPlot.py

@@ -6,7 +6,7 @@ import matplotlib.pyplot as plt
 import pandas as pd
 import seaborn as sns
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.errors import SkipTestError
 from validmind.vm_models import VMDataset
 
@@ -70,6 +70,7 @@ def SpreadPlot(dataset: VMDataset):
     ]
 
     figures = []
+    spread_data = {}
 
     for var1, var2 in feature_pairs:
         fig, ax = plt.subplots()
@@ -80,8 +81,9 @@ def SpreadPlot(dataset: VMDataset):
             y=0.95,
         )
 
+        spread_series = df[var1] - df[var2]
         sns.lineplot(
-            data=df[var1] - df[var2],
+            data=spread_series,
             ax=ax,
         )
 
@@ -89,5 +91,8 @@ def SpreadPlot(dataset: VMDataset):
         ax.tick_params(axis="both", labelsize=18)
 
         figures.append(fig)
+        spread_data[f"{var1}_{var2}_spread"] = spread_series.to_frame(
+            name=f"spread_{var1}_{var2}"
+        )
 
-    return tuple(figures)
+    return (*figures, RawData(spread_data=spread_data))

validmind/tests/data_validation/TabularCategoricalBarPlots.py

@@ -4,7 +4,7 @@
 
 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
 
@@ -66,9 +66,11 @@ def TabularCategoricalBarPlots(dataset: VMDataset):
     ]
 
     figures = []
+    counts_dict = {}
 
     for col in dataset.feature_columns_categorical:
         counts = dataset.df[col].value_counts()
+        counts_dict[col] = counts
 
         fig = go.Figure()
         fig.add_trace(
@@ -90,4 +92,4 @@ def TabularCategoricalBarPlots(dataset: VMDataset):
         )
         figures.append(fig)
 
-    return tuple(figures)
+    return (*figures, RawData(category_counts=counts_dict))

validmind/tests/data_validation/TabularDateTimeHistograms.py

@@ -5,7 +5,7 @@
 import pandas as pd
 import plotly.graph_objects as go
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.errors import SkipTestError
 from validmind.vm_models import VMDataset
 
@@ -72,4 +72,4 @@ def TabularDateTimeHistograms(dataset: VMDataset):
         font=dict(size=18),
     )
 
-    return fig
+    return fig, RawData(date_differences=date_diffs)

validmind/tests/data_validation/TargetRateBarPlots.py

@@ -6,7 +6,7 @@ import numpy as np
 import plotly.graph_objs as go
 from plotly.subplots import make_subplots
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.errors import SkipTestError
 from validmind.vm_models import VMDataset
 
@@ -62,12 +62,13 @@ def TargetRateBarPlots(dataset: VMDataset):
 
     df = dataset.df
     figures = []
+    raw_data = []
 
     for col in dataset.feature_columns_categorical:
-
         # Calculate counts and default rate for each category
         counts = df[col].value_counts()
         default_rate = df.groupby(col)[dataset.target_column].mean()
+        raw_data.append({"column": col, "counts": counts, "default_rate": default_rate})
 
         fig = make_subplots(
             rows=1,
@@ -107,4 +108,4 @@ def TargetRateBarPlots(dataset: VMDataset):
 
         figures.append(fig)
 
-    return tuple(figures)
+    return (*figures, RawData(target_rates_by_category=raw_data))

validmind/tests/data_validation/TimeSeriesFrequency.py

@@ -5,7 +5,7 @@
 import pandas as pd
 import plotly.graph_objects as go
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.errors import SkipTestError
 from validmind.vm_models import VMDataset
 
@@ -103,4 +103,9 @@ def TimeSeriesFrequency(dataset: VMDataset):
         ),
     )
 
-    return frequencies, fig, len(set(item["Frequency"] for item in frequencies)) == 1
+    return (
+        frequencies,
+        fig,
+        len(set(item["Frequency"] for item in frequencies)) == 1,
+        RawData(time_diff_days=time_diff_days),
+    )

validmind/tests/data_validation/TimeSeriesMissingValues.py

@@ -6,7 +6,7 @@ import pandas as pd
 import plotly.express as px
 import plotly.figure_factory as ff
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.errors import SkipTestError
 from validmind.vm_models import VMDataset
 
@@ -62,15 +62,18 @@ def TimeSeriesMissingValues(dataset: VMDataset, min_threshold: int = 1):
 
     if sum(missing.values) == 0:
         # if theres no missing values, no need to plot anything
-        return [
-            {
-                "Column": col,
-                "Number of Missing Values": missing[col],
-                "Percentage of Missing Values (%)": 0,
-                "Pass/Fail": "Pass",
-            }
-            for col in missing.index
-        ], True
+        return (
+            [
+                {
+                    "Column": col,
+                    "Number of Missing Values": missing[col],
+                    "Percentage of Missing Values (%)": 0,
+                    "Pass/Fail": "Pass",
+                }
+                for col in missing.index
+            ],
+            True,
+        )
 
     barplot = px.bar(
         missing,
@@ -110,4 +113,5 @@ def TimeSeriesMissingValues(dataset: VMDataset, min_threshold: int = 1):
         barplot,
         heatmap,
         all(missing[col] < min_threshold for col in missing.index),
+        RawData(missing_values_count=missing, missing_values_mask=missing_mask),
     )

validmind/tests/data_validation/TimeSeriesOutliers.py

@@ -111,8 +111,4 @@ def TimeSeriesOutliers(dataset: VMDataset, zscore_threshold: int = 3):
 
         figures.append(fig)
 
-    return (
-        outlier_df.sort_values(["Column", "Date"]),
-        figures,
-        len(outlier_df) == 0,
-    )
+    return (outlier_df.sort_values(["Column", "Date"]), figures, len(outlier_df) == 0)

validmind/tests/data_validation/WOEBinPlots.py

@@ -9,7 +9,7 @@ import plotly.graph_objects as go
 import scorecardpy as sc
 from plotly.subplots import make_subplots
 
-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
@@ -140,4 +140,4 @@ def WOEBinPlots(
 
         figures.append(fig)
 
-    return tuple(figures)
+    return (*figures, RawData(woe_iv_data=woe_iv_df))

validmind/tests/data_validation/WOEBinTable.py

@@ -5,7 +5,7 @@
 import pandas as pd
 import scorecardpy as sc
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.errors import SkipTestError
 from validmind.vm_models import VMDataset
 
@@ -61,12 +61,14 @@ def WOEBinTable(dataset: VMDataset, breaks_adj: list = None):
     except Exception as e:
         raise SkipTestError(f"Error during binning: {e}")
 
+    result_table = (
+        pd.concat(bins.values(), keys=bins.keys())
+        .reset_index()
+        .drop(columns=["variable"])
+        .rename(columns={"level_0": "variable"})
+        .assign(bin_number=lambda x: x.groupby("variable").cumcount())
+    )
+
     return {
-        "Weight of Evidence (WoE) and Information Value (IV)": (
-            pd.concat(bins.values(), keys=bins.keys())
-            .reset_index()
-            .drop(columns=["variable"])
-            .rename(columns={"level_0": "variable"})
-            .assign(bin_number=lambda x: x.groupby("variable").cumcount())
-        )
-    }
+        "Weight of Evidence (WoE) and Information Value (IV)": result_table
+    }, RawData(woe_bins=bins)

validmind/tests/data_validation/nlp/CommonWords.py

@@ -8,7 +8,7 @@ import nltk
 import plotly.graph_objects as go
 from nltk.corpus import stopwords
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.vm_models import VMDataset
 
 
@@ -94,4 +94,4 @@ def CommonWords(dataset: VMDataset):
         xaxis_tickangle=-45,
     )
 
-    return fig
+    return fig, RawData(words=x, frequencies=y)

validmind/tests/data_validation/nlp/Hashtags.py

@@ -6,7 +6,7 @@ import re
 
 import plotly.graph_objects as go
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.errors import SkipTestError
 from validmind.vm_models import VMDataset
 
@@ -76,4 +76,4 @@ def Hashtags(dataset: VMDataset, top_hashtags: int = 25):
         xaxis_tickangle=-45,
     )
 
-    return fig
+    return fig, RawData(top_hashtag_counts=top_hashtag_counts)

validmind/tests/data_validation/nlp/LanguageDetection.py

@@ -5,7 +5,7 @@
 import plotly.express as px
 from langdetect import LangDetectException, detect
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 
 
 @tags("nlp", "text_data", "visualization")
@@ -64,9 +64,12 @@ def LanguageDetection(dataset):
 
     languages = dataset.df[dataset.text_column].apply(detect_language)
 
-    return px.histogram(
-        languages,
-        x=languages,
-        title="Language Distribution",
-        labels={"x": "Language Codes"},
+    return (
+        px.histogram(
+            languages,
+            x=languages,
+            title="Language Distribution",
+            labels={"x": "Language Codes"},
+        ),
+        RawData(detected_languages=languages),
     )

validmind/tests/data_validation/nlp/Mentions.py

@@ -7,7 +7,7 @@ import re
 import pandas as pd
 import plotly.express as px
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.errors import SkipTestError
 from validmind.vm_models import VMDataset
 
@@ -75,9 +75,12 @@ def Mentions(dataset: VMDataset, top_mentions: int = 25):
         }
     )
 
-    return px.treemap(
-        mention_frequencies_df,
-        path=["Scenario"],
-        values="Percentage",
-        title="Tree of Mentions",
+    return (
+        px.treemap(
+            mention_frequencies_df,
+            path=["Scenario"],
+            values="Percentage",
+            title="Tree of Mentions",
+        ),
+        RawData(mention_counts=mention_counts),
     )

validmind/tests/data_validation/nlp/PolarityAndSubjectivity.py

@@ -7,7 +7,7 @@ import pandas as pd
 import plotly.express as px
 from textblob import TextBlob
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 
 
 @tags("nlp", "text_data", "data_validation")
@@ -144,4 +144,4 @@ def PolarityAndSubjectivity(dataset, threshold_subjectivity=0.5, threshold_polar
 
     statistics_tables = {"Quadrant Distribution": quadrant_df, "Statistics": stats_df}
 
-    return fig, statistics_tables
+    return fig, statistics_tables, RawData(sentiment_data=data)

validmind/tests/data_validation/nlp/Punctuations.py

@@ -11,7 +11,7 @@ from collections import defaultdict
 
 import plotly.graph_objects as go
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 
 
 @tags("nlp", "text_data", "visualization", "frequency_analysis")
@@ -63,7 +63,9 @@ def Punctuations(dataset, count_mode="token"):
 
     corpus = _create_corpus(dataset.df, dataset.text_column)
     punctuation_counts = _count_punctuations(corpus, count_mode)
-    return _create_punctuation_plot(punctuation_counts)
+    fig = _create_punctuation_plot(punctuation_counts)
+
+    return fig, RawData(punctuation_counts=punctuation_counts)
 
 
 def _create_punctuation_plot(punctuation_counts):

validmind/tests/data_validation/nlp/Sentiment.py

@@ -8,7 +8,7 @@ import nltk
 import seaborn as sns
 from nltk.sentiment import SentimentIntensityAnalyzer
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 
 
 @tags("nlp", "text_data", "data_validation")
@@ -77,4 +77,4 @@ def Sentiment(dataset):
 
     plt.close("all")
 
-    return fig
+    return fig, RawData(sentiment_scores=vader_sentiment.tolist())

validmind/tests/data_validation/nlp/StopWords.py

@@ -13,7 +13,7 @@ import pandas as pd
 import plotly.graph_objects as go
 from nltk.corpus import stopwords
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.vm_models import VMDataset
 
 
@@ -84,17 +84,17 @@ def StopWords(
     nltk.download("stopwords", quiet=True)
 
     stop = set(stopwords.words("english"))
-    dic = defaultdict(int)
+    stop_word_frequencies = defaultdict(int)
     for word in corpus:
         if word in stop:
-            dic[word] += 1
+            stop_word_frequencies[word] += 1
 
     # Calculate the total number of words in the corpus
     total_words = len(corpus)
 
     # Calculate the percentage of each word in the corpus
     word_percentages = {}
-    for word, count in dic.items():
+    for word, count in stop_word_frequencies.items():
         percentage = (count / total_words) * 100
         word_percentages[word] = percentage
 
@@ -124,4 +124,5 @@ def StopWords(
         },
         fig,
         passed,
+        RawData(stop_word_frequencies=stop_word_frequencies, total_words=total_words),
     )

validmind/tests/data_validation/nlp/TextDescription.py

@@ -9,7 +9,7 @@ import pandas as pd
 import plotly.express as px
 from nltk.corpus import stopwords
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.vm_models import VMDataset
 
 
@@ -173,4 +173,4 @@ def TextDescription(
             )
         )
 
-    return tuple(figures)
+    return (*figures, RawData(metrics_dataframe=metrics_df))

validmind/tests/data_validation/nlp/Toxicity.py

@@ -6,7 +6,7 @@ import evaluate
 import matplotlib.pyplot as plt
 import seaborn as sns
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 
 
 @tags("nlp", "text_data", "data_validation")
@@ -73,4 +73,4 @@ def Toxicity(dataset):
 
     plt.close()
 
-    return fig
+    return fig, RawData(toxicity_scores=toxicity_scores)

validmind/tests/model_validation/BertScore.py

@@ -6,7 +6,7 @@ import evaluate
 import pandas as pd
 import plotly.graph_objects as go
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.tests.utils import validate_prediction
 
 
@@ -131,4 +131,4 @@ def BertScore(
     # Create a DataFrame from all collected statistics
     result_df = pd.DataFrame(stats_df).reset_index().rename(columns={"index": "Metric"})
 
-    return (result_df, *tuple(figures))
+    return (result_df, *figures, RawData(bert_scores_df=metrics_df))

validmind/tests/model_validation/BleuScore.py

@@ -6,7 +6,7 @@ import evaluate
 import pandas as pd
 import plotly.graph_objects as go
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.tests.utils import validate_prediction
 
 
@@ -114,4 +114,4 @@ def BleuScore(dataset, model):
     # Create a DataFrame from all collected statistics
     result_df = pd.DataFrame(stats_df).reset_index().rename(columns={"index": "Metric"})
 
-    return (result_df, *tuple(figures))
+    return (result_df, *figures, RawData(bleu_scores_df=metrics_df))

validmind/tests/model_validation/ClusterSizeDistribution.py

@@ -5,7 +5,7 @@
 import pandas as pd
 import plotly.graph_objects as go
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.vm_models import VMDataset, VMModel
 
 
@@ -72,4 +72,4 @@ def ClusterSizeDistribution(dataset: VMDataset, model: VMModel):
     fig.update_yaxes(title_text="Counts", showgrid=False)
     fig.update_layout(title_text="Cluster distribution", title_x=0.5, barmode="group")
 
-    return fig
+    return fig, RawData(cluster_counts=df_counts)

validmind/tests/model_validation/ContextualRecall.py

@@ -6,7 +6,7 @@ import nltk
 import pandas as pd
 import plotly.graph_objects as go
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.tests.utils import validate_prediction
 
 
@@ -118,4 +118,4 @@ def ContextualRecall(dataset, model):
     # Create a DataFrame from all collected statistics
     result_df = pd.DataFrame(stats_df).reset_index().rename(columns={"index": "Metric"})
 
-    return (result_df, *tuple(figures))
+    return (result_df, *tuple(figures), RawData(contextual_recall_scores=metrics_df))

validmind/tests/model_validation/FeaturesAUC.py

@@ -7,7 +7,7 @@ import pandas as pd
 import plotly.graph_objects as go
 from sklearn.metrics import roc_auc_score
 
-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
@@ -95,4 +95,4 @@ def FeaturesAUC(dataset: VMDataset, fontsize: int = 12, figure_height: int = 500
         height=figure_height,
     )
 
-    return fig
+    return fig, RawData(feature_aucs=aucs)

validmind/tests/model_validation/MeteorScore.py

@@ -6,7 +6,7 @@ import evaluate
 import pandas as pd
 import plotly.graph_objects as go
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.tests.utils import validate_prediction
 
 
@@ -117,4 +117,4 @@ def MeteorScore(dataset, model):
     # Create a DataFrame from all collected statistics
     result_df = pd.DataFrame(stats_df).reset_index().rename(columns={"index": "Metric"})
 
-    return (result_df, *tuple(figures))
+    return (result_df, *tuple(figures), RawData(meteor_scores=metrics_df))

validmind/tests/model_validation/ModelPredictionResiduals.py

@@ -6,7 +6,7 @@ import pandas as pd
 import plotly.graph_objects as go
 from scipy.stats import kstest
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 
 
 @tags("regression")
@@ -102,4 +102,4 @@ def ModelPredictionResiduals(
     # Create a summary DataFrame for the KS normality test results
     summary_df = pd.DataFrame([summary])
 
-    return (summary_df, *figures)
+    return (summary_df, *figures, RawData(residuals=residuals))

validmind/tests/model_validation/RegardScore.py

@@ -6,7 +6,7 @@ import evaluate
 import pandas as pd
 import plotly.graph_objects as go
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.tests.utils import validate_prediction
 
 
@@ -142,4 +142,8 @@ def RegardScore(dataset, model):
         ]
     ]
 
-    return (result_df, *tuple(figures))
+    return (
+        result_df,
+        *figures,
+        RawData(true_regard=true_df, pred_regard=pred_df),
+    )

validmind/tests/model_validation/RegressionResidualsPlot.py

@@ -6,7 +6,7 @@ import numpy as np
 import plotly.figure_factory as ff
 import plotly.graph_objects as go
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 from validmind.vm_models import VMDataset, VMModel
 
 
@@ -60,8 +60,9 @@ def RegressionResidualsPlot(model: VMModel, dataset: VMDataset, bin_size: float
     figures = []
 
     # Residuals plot
+    residuals = y_true.flatten() - y_pred.flatten()
     fig = ff.create_distplot(
-        hist_data=[y_true.flatten() - y_pred.flatten()],
+        hist_data=[residuals],
         group_labels=["Residuals"],
         bin_size=[bin_size],
         show_hist=True,
@@ -104,4 +105,4 @@ def RegressionResidualsPlot(model: VMModel, dataset: VMDataset, bin_size: float
         )
     )
 
-    return tuple(figures)
+    return (*figures, RawData(residuals=residuals, y_true=y_true, y_pred=y_pred))

validmind/tests/model_validation/RougeScore.py

@@ -6,7 +6,7 @@ import pandas as pd
 import plotly.graph_objects as go
 from rouge import Rouge
 
-from validmind import tags, tasks
+from validmind import RawData, tags, tasks
 
 
 @tags("nlp", "text_data", "visualization")
@@ -118,7 +118,8 @@ def RougeScore(dataset, model, metric="rouge-1"):
         {"p": "Precision", "r": "Recall", "f": "F1 Score"}
     )
 
-    # Create a DataFrame from all collected statistics
-    result_df = pd.DataFrame(stats_df).reset_index().rename(columns={"index": "Metric"})
-
-    return (result_df, *tuple(figures))
+    return (
+        pd.DataFrame(stats_df).reset_index().rename(columns={"index": "Metric"}),
+        *figures,
+        RawData(rouge_scores_df=df_scores),
+    )