validmind 2.4.0__py3-none-any.whl → 2.4.4__py3-none-any.whl

validmind/models/huggingface.py

@@ -56,7 +56,6 @@ class HFModel(VMModel):
             return [result["label"] for result in results]
         elif tasks[-1] == "feature_extraction":
             # Extract [CLS] token embedding for each input and return as list of lists
-            print(f"len(results): {len(results)}")
             return [embedding[0][0] for embedding in results]
         else:
             return results

validmind/template.py

@@ -22,6 +22,8 @@ CONTENT_TYPE_MAP = {
     "dynamic": "Dynamic Content",
     "text": "Text",
     "risk_assessment": "Risk Assessment",
+    "assessment_summary": "Assessment Summary",
+    "guideline": "Guideline Assessment",
 }
 
 

validmind/tests/data_validation/TabularDescriptionTables.py

@@ -2,15 +2,14 @@
 # See the LICENSE file in the root of this repository for details.
 # SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial
 
-from dataclasses import dataclass
-
 import pandas as pd
 
-from validmind.vm_models import Metric, ResultSummary, ResultTable, ResultTableMetadata
+from validmind import tags, tasks
 
 
-@dataclass
-class TabularDescriptionTables(Metric):
+@tags("tabular_data")
+@tasks("classification", "regression")
+def TabularDescriptionTables(dataset):
     """
     Summarizes key descriptive statistics for numerical, categorical, and datetime variables in a dataset.
 
@@ -54,155 +53,104 @@ class TabularDescriptionTables(Metric):
     chosen algorithm.
     """
 
-    name = "tabular_description_tables"
-    required_inputs = ["dataset"]
-
-    tasks = ["classification", "regression"]
-    tags = ["tabular_data"]
-
-    def get_summary_statistics_numerical(self, numerical_fields):
-        summary_stats = self.inputs.dataset.df[numerical_fields].describe().T
-        summary_stats["Missing Values (%)"] = (
-            self.inputs.dataset.df[numerical_fields].isnull().mean() * 100
-        )
-        summary_stats["Data Type"] = self.inputs.dataset.df[
-            numerical_fields
-        ].dtypes.astype(str)
-        summary_stats = summary_stats[
-            ["count", "mean", "min", "max", "Missing Values (%)", "Data Type"]
-        ]
-        summary_stats.columns = [
-            "Num of Obs",
-            "Mean",
-            "Min",
-            "Max",
-            "Missing Values (%)",
-            "Data Type",
-        ]
-        summary_stats["Num of Obs"] = summary_stats["Num of Obs"].astype(int)
-        summary_stats = summary_stats.sort_values(
-            by="Missing Values (%)", ascending=False
-        )
-        summary_stats.reset_index(inplace=True)
-        summary_stats.rename(columns={"index": "Numerical Variable"}, inplace=True)
-        return summary_stats
-
-    def get_summary_statistics_categorical(self, categorical_fields):
-        summary_stats = pd.DataFrame()
-        if categorical_fields:  # check if the list is not empty
-            for column in self.inputs.dataset.df[categorical_fields].columns:
-                summary_stats.loc[column, "Num of Obs"] = int(
-                    self.inputs.dataset.df[column].count()
-                )
-                summary_stats.loc[
-                    column, "Num of Unique Values"
-                ] = self.inputs.dataset.df[column].nunique()
-                summary_stats.loc[column, "Unique Values"] = str(
-                    self.inputs.dataset.df[column].unique()
-                )
-                summary_stats.loc[column, "Missing Values (%)"] = (
-                    self.inputs.dataset.df[column].isnull().mean() * 100
-                )
-                summary_stats.loc[column, "Data Type"] = str(
-                    self.inputs.dataset.df[column].dtype
-                )
-
-            summary_stats = summary_stats.sort_values(
-                by="Missing Values (%)", ascending=False
-            )
-            summary_stats.reset_index(inplace=True)
-            summary_stats.rename(
-                columns={"index": "Categorical Variable"}, inplace=True
-            )
-        return summary_stats
-
-    def get_summary_statistics_datetime(self, datetime_fields):
-        summary_stats = pd.DataFrame()
-        for column in self.inputs.dataset.df[datetime_fields].columns:
-            summary_stats.loc[column, "Num of Obs"] = int(
-                self.inputs.dataset.df[column].count()
-            )
-            summary_stats.loc[column, "Num of Unique Values"] = self.inputs.dataset.df[
+    numerical_fields = get_numerical_columns(dataset)
+    categorical_fields = get_categorical_columns(dataset)
+    datetime_fields = get_datetime_columns(dataset)
+
+    summary_stats_numerical = get_summary_statistics_numerical(
+        dataset, numerical_fields
+    )
+    summary_stats_categorical = get_summary_statistics_categorical(
+        dataset, categorical_fields
+    )
+    summary_stats_datetime = get_summary_statistics_datetime(dataset, datetime_fields)
+
+    return (summary_stats_numerical, summary_stats_categorical, summary_stats_datetime)
+
+
+def get_summary_statistics_numerical(dataset, numerical_fields):
+    summary_stats = dataset.df[numerical_fields].describe().T
+    summary_stats["Missing Values (%)"] = (
+        dataset.df[numerical_fields].isnull().mean() * 100
+    )
+    summary_stats["Data Type"] = dataset.df[numerical_fields].dtypes.astype(str)
+    summary_stats = summary_stats[
+        ["count", "mean", "min", "max", "Missing Values (%)", "Data Type"]
+    ]
+    summary_stats.columns = [
+        "Num of Obs",
+        "Mean",
+        "Min",
+        "Max",
+        "Missing Values (%)",
+        "Data Type",
+    ]
+    summary_stats["Num of Obs"] = summary_stats["Num of Obs"].astype(int)
+    summary_stats = summary_stats.sort_values(by="Missing Values (%)", ascending=False)
+    summary_stats.reset_index(inplace=True)
+    summary_stats.rename(columns={"index": "Numerical Variable"}, inplace=True)
+    return summary_stats
+
+
+def get_summary_statistics_categorical(dataset, categorical_fields):
+    summary_stats = pd.DataFrame()
+    if categorical_fields:  # check if the list is not empty
+        for column in dataset.df[categorical_fields].columns:
+            summary_stats.loc[column, "Num of Obs"] = int(dataset.df[column].count())
+            summary_stats.loc[column, "Num of Unique Values"] = dataset.df[
                 column
             ].nunique()
-            summary_stats.loc[column, "Earliest Date"] = self.inputs.dataset.df[
-                column
-            ].min()
-            summary_stats.loc[column, "Latest Date"] = self.inputs.dataset.df[
-                column
-            ].max()
-            summary_stats.loc[column, "Missing Values (%)"] = (
-                self.inputs.dataset.df[column].isnull().mean() * 100
+            summary_stats.loc[column, "Unique Values"] = str(
+                dataset.df[column].unique()
             )
-            summary_stats.loc[column, "Data Type"] = str(
-                self.inputs.dataset.df[column].dtype
+            summary_stats.loc[column, "Missing Values (%)"] = (
+                dataset.df[column].isnull().mean() * 100
             )
+            summary_stats.loc[column, "Data Type"] = str(dataset.df[column].dtype)
 
-        if not summary_stats.empty:
-            summary_stats = summary_stats.sort_values(
-                by="Missing Values (%)", ascending=False
-            )
+        summary_stats = summary_stats.sort_values(
+            by="Missing Values (%)", ascending=False
+        )
         summary_stats.reset_index(inplace=True)
-        summary_stats.rename(columns={"index": "Datetime Variable"}, inplace=True)
-        return summary_stats
-
-    def summary(self, metric_value):
-        summary_stats_numerical = metric_value["numerical"]
-        summary_stats_categorical = metric_value["categorical"]
-        summary_stats_datetime = metric_value["datetime"]
-
-        return ResultSummary(
-            results=[
-                ResultTable(
-                    data=summary_stats_numerical,
-                    metadata=ResultTableMetadata(title="Numerical Variables"),
-                ),
-                ResultTable(
-                    data=summary_stats_categorical,
-                    metadata=ResultTableMetadata(title="Categorical Variables"),
-                ),
-                ResultTable(
-                    data=summary_stats_datetime,
-                    metadata=ResultTableMetadata(title="Datetime Variables"),
-                ),
-            ]
+        summary_stats.rename(columns={"index": "Categorical Variable"}, inplace=True)
+    return summary_stats
+
+
+def get_summary_statistics_datetime(dataset, datetime_fields):
+    summary_stats = pd.DataFrame()
+    for column in dataset.df[datetime_fields].columns:
+        summary_stats.loc[column, "Num of Obs"] = int(dataset.df[column].count())
+        summary_stats.loc[column, "Num of Unique Values"] = dataset.df[column].nunique()
+        summary_stats.loc[column, "Earliest Date"] = dataset.df[column].min()
+        summary_stats.loc[column, "Latest Date"] = dataset.df[column].max()
+        summary_stats.loc[column, "Missing Values (%)"] = (
+            dataset.df[column].isnull().mean() * 100
         )
+        summary_stats.loc[column, "Data Type"] = str(dataset.df[column].dtype)
 
-    def get_categorical_columns(self):
-        categorical_columns = self.inputs.dataset.df.select_dtypes(
-            include=["object", "category"]
-        ).columns.tolist()
-        return categorical_columns
-
-    def get_numerical_columns(self):
-        numerical_columns = self.inputs.dataset.df.select_dtypes(
-            include=["int", "float", "uint8"]
-        ).columns.tolist()
-        return numerical_columns
-
-    def get_datetime_columns(self):
-        datetime_columns = self.inputs.dataset.df.select_dtypes(
-            include=["datetime"]
-        ).columns.tolist()
-        return datetime_columns
-
-    def run(self):
-        numerical_fields = self.get_numerical_columns()
-        categorical_fields = self.get_categorical_columns()
-        datetime_fields = self.get_datetime_columns()
-
-        summary_stats_numerical = self.get_summary_statistics_numerical(
-            numerical_fields
-        )
-        summary_stats_categorical = self.get_summary_statistics_categorical(
-            categorical_fields
-        )
-        summary_stats_datetime = self.get_summary_statistics_datetime(datetime_fields)
-
-        return self.cache_results(
-            {
-                "numerical": summary_stats_numerical.to_dict(orient="records"),
-                "categorical": summary_stats_categorical.to_dict(orient="records"),
-                "datetime": summary_stats_datetime.to_dict(orient="records"),
-            }
+    if not summary_stats.empty:
+        summary_stats = summary_stats.sort_values(
+            by="Missing Values (%)", ascending=False
         )
+    summary_stats.reset_index(inplace=True)
+    summary_stats.rename(columns={"index": "Datetime Variable"}, inplace=True)
+    return summary_stats
+
+
+def get_categorical_columns(dataset):
+    categorical_columns = dataset.df.select_dtypes(
+        include=["object", "category"]
+    ).columns.tolist()
+    return categorical_columns
+
+
+def get_numerical_columns(dataset):
+    numerical_columns = dataset.df.select_dtypes(
+        include=["int", "float", "uint8"]
+    ).columns.tolist()
+    return numerical_columns
+
+
+def get_datetime_columns(dataset):
+    datetime_columns = dataset.df.select_dtypes(include=["datetime"]).columns.tolist()
+    return datetime_columns

validmind/tests/model_validation/embeddings/ClusterDistribution.py

@@ -52,7 +52,7 @@ class ClusterDistribution(Metric):
         "num_clusters": 5,
     }
     tasks = ["feature_extraction"]
-    tags = ["llm", "text_data", "text_embeddings", "visualization"]
+    tags = ["llm", "text_data", "embeddings", "visualization"]
 
     def run(self):
         # run kmeans clustering on embeddings

validmind/tests/model_validation/embeddings/CosineSimilarityDistribution.py

@@ -51,7 +51,7 @@ class CosineSimilarityDistribution(Metric):
     name = "Text Embeddings Cosine Similarity Distribution"
     required_inputs = ["model", "dataset"]
     tasks = ["feature_extraction"]
-    tags = ["llm", "text_data", "text_embeddings", "visualization"]
+    tags = ["llm", "text_data", "embeddings", "visualization"]
 
     def run(self):
         # Compute cosine similarity

validmind/tests/model_validation/embeddings/DescriptiveAnalytics.py

@@ -54,7 +54,7 @@ class DescriptiveAnalytics(Metric):
     name = "Descriptive Analytics for Text Embeddings Models"
     required_inputs = ["model", "dataset"]
     tasks = ["feature_extraction"]
-    tags = ["llm", "text_data", "text_embeddings", "visualization"]
+    tags = ["llm", "text_data", "embeddings", "visualization"]
 
     def run(self):
         # Assuming y_pred returns a 2D array of embeddings [samples, features]

validmind/tests/model_validation/embeddings/EmbeddingsVisualization2D.py

@@ -54,7 +54,7 @@ class EmbeddingsVisualization2D(Metric):
         "perplexity": 30,
     }
     tasks = ["feature_extraction"]
-    tags = ["llm", "text_data", "text_embeddings", "visualization"]
+    tags = ["llm", "text_data", "embeddings", "visualization"]
 
     def run(self):
         cluster_column = self.params.get("cluster_column")

validmind/tests/model_validation/embeddings/StabilityAnalysis.py

@@ -30,7 +30,7 @@ class StabilityAnalysis(ThresholdTest):
         "mean_similarity_threshold": 0.7,
     }
     tasks = ["feature_extraction"]
-    tags = ["llm", "text_data", "text_embeddings", "visualization"]
+    tags = ["llm", "text_data", "embeddings", "visualization"]
 
     @abstractmethod
     def perturb_data(self, data: str) -> str:
@@ -62,7 +62,8 @@ class StabilityAnalysis(ThresholdTest):
 
     def run(self):
         # Perturb the test dataset
-        original = self.inputs.dataset.df
+        text_column = self.inputs.dataset.text_column
+        original = self.inputs.dataset.df[[text_column]]
         perturbed = original.copy()
         perturbed.update(
             perturbed.select_dtypes(include="object").applymap(self.perturb_data)

validmind/tests/model_validation/embeddings/StabilityAnalysisTranslation.py

@@ -4,8 +4,12 @@
 
 from transformers import MarianMTModel, MarianTokenizer
 
+from validmind.logging import get_logger
+
 from .StabilityAnalysis import StabilityAnalysis
 
+logger = get_logger(__name__)
+
 
 class StabilityAnalysisTranslation(StabilityAnalysis):
     """
@@ -61,8 +65,11 @@ class StabilityAnalysisTranslation(StabilityAnalysis):
     }
 
     def perturb_data(self, data: str):
-        if not isinstance(data, str):
-            return data
+        if len(data) > 512:
+            logger.info(
+                "Data length exceeds 512 tokens. Truncating data to 512 tokens."
+            )
+            data = data[:512]
 
         source_lang = self.params["source_lang"]
         target_lang = self.params["target_lang"]

validmind/tests/model_validation/embeddings/TSNEComponentsPairwisePlots.py

@@ -53,7 +53,6 @@ def TSNEComponentsPairwisePlots(
     - t-SNE visualizations can be misleading if interpreted without considering the stochastic nature of the algorithm;
     two runs with the same parameters might yield different visual outputs, necessitating multiple runs for a consistent interpretation.
     """
-
     # Get embeddings from the dataset using the model
     embeddings = np.stack(dataset.y_pred(model))
 

validmind/tests/model_validation/sklearn/HyperParametersTuning.py

@@ -60,8 +60,9 @@ class HyperParametersTuning(Metric):
         param_grid = self.params["param_grid"]
         if param_grid is None:
             raise SkipTestError(
-                "param_grid in dictnary format must be provide to run hyper parameter tuning"
+                "param_grid in dictonary format must be provided to run this test"
             )
+
         model = self.inputs.model.model
         estimators = GridSearchCV(
             model, param_grid=param_grid, scoring=self.params["scoring"]

validmind/tests/run.py

@@ -118,7 +118,7 @@ def _combine_figures(figure_lists: List[List[Any]], input_groups: List[Dict[str,
 
     title_template = "{current_title}({input_description})"
 
-    for i, figures in enumerate(list(zip(*figure_lists))):
+    for figures in list(zip(*figure_lists)):
         if is_plotly_figure(figures[0].figure):
             _update_plotly_titles(figures, input_groups, title_template)
         elif is_matplotlib_figure(figures[0].figure):

validmind/vm_models/dataset/dataset.py

@@ -139,13 +139,25 @@ class VMDataset:
         )
 
     def _add_column(self, column_name, column_values):
-        if len(column_values) != len(self.df):
-            raise ValueError(
-                "Length of values doesn't match number of rows in the DataFrame."
-            )
+        column_values = np.array(column_values)
+
+        if column_values.ndim == 1:
+            if len(column_values) != len(self.df):
+                raise ValueError(
+                    "Length of values doesn't match number of rows in the DataFrame."
+                )
+            self.columns.append(column_name)
+            self.df[column_name] = column_values
+        elif column_values.ndim == 2:
+            if column_values.shape[0] != len(self.df):
+                raise ValueError(
+                    "Number of rows in values doesn't match number of rows in the DataFrame."
+                )
+            self.columns.append(column_name)
+            self.df[column_name] = column_values.tolist()
 
-        self.columns.append(column_name)
-        self.df[column_name] = column_values
+        else:
+            raise ValueError("Only 1D and 2D arrays are supported for column_values.")
 
     def _validate_assign_predictions(
         self,

validmind/vm_models/test_suite/summary.py

@@ -93,10 +93,10 @@ class TestSuiteSummary:
 
     def _add_results_link(self):
         # avoid circular import
-        from ...api_client import get_api_host, get_api_project
+        from ...api_client import get_api_host, get_api_model
 
         ui_host = get_api_host().replace("/api/v1/tracking", "").replace("api", "app")
-        link = f"{ui_host}/projects/{get_api_project()}/project-overview"
+        link = f"{ui_host}/projects/{get_api_model()}/project-overview"
         results_link = f"""
         <h3>
             Check out the updated documentation in your

{validmind-2.4.0.dist-info → validmind-2.4.4.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: validmind
-Version: 2.4.0
+Version: 2.4.4
 Summary: ValidMind Developer Framework
 License: Commercial License
 Author: Andres Rodriguez
@@ -46,6 +46,7 @@ Requires-Dist: scikit-learn
 Requires-Dist: scipy
 Requires-Dist: scorecardpy (>=0.1.9.6,<0.2.0.0)
 Requires-Dist: seaborn
+Requires-Dist: sentencepiece (>=0.2.0,<0.3.0) ; extra == "all" or extra == "huggingface" or extra == "llm"
 Requires-Dist: sentry-sdk (>=1.24.0,<2.0.0)
 Requires-Dist: shap (>=0.42.0,<0.43.0)
 Requires-Dist: statsmodels
@@ -53,7 +54,7 @@ Requires-Dist: tabulate (>=0.8.9,<0.9.0)
 Requires-Dist: textblob (>=0.18.0.post0,<0.19.0)
 Requires-Dist: torch (>=1.10.0) ; extra == "all" or extra == "llm" or extra == "pytorch"
 Requires-Dist: tqdm
-Requires-Dist: transformers (>=4.32.0,<5.0.0) ; extra == "all" or extra == "llm" or extra == "huggingface"
+Requires-Dist: transformers (>=4.32.0,<5.0.0) ; extra == "all" or extra == "huggingface" or extra == "llm"
 Requires-Dist: xgboost (>=1.5.2,<3)
 Requires-Dist: ydata-profiling
 Description-Content-Type: text/markdown

{validmind-2.4.0.dist-info → validmind-2.4.4.dist-info}/RECORD

@@ -1,8 +1,8 @@
 validmind/__init__.py,sha256=UfmzPwUCdUWbWq3zPqqmq4jw0_kfl3hX4U72p_seE4I,3700
-validmind/__version__.py,sha256=yLaFvd-K80rs_ClRVYULStijkok4RfYSaanIt_E-aKM,22
+validmind/__version__.py,sha256=zUyWRJ2idk3EkNSUQlyMfzQtAlpD5CCHF2ifejOcf-w,22
 validmind/ai/test_descriptions.py,sha256=Q1Ftus4x5eiVLKWJu7hqPLukBQZzhy-dARqq_6_JWtk,9464
 validmind/ai/utils.py,sha256=TEXII_S5CpkpczzSyHwTlqLcPMLnPBJWEBR6QFMKh1U,3421
-validmind/api_client.py,sha256=0IR8MpH_GxBykOs4Egz7oEKZLoOEwoli81X1oFL0DD8,16893
+validmind/api_client.py,sha256=xr9VNqCmA_WFf8rVm-0M0pmzVyLAPFOnfEe4dAog1LA,17144
 validmind/client.py,sha256=UnsEwWK_s3nuktr6i2U3haLjjlWRGR6H431jsZpKEDA,18649
 validmind/client_config.py,sha256=58L6s6-9vFWC9vkSs_98CjV1YWmlksdhblJtPQxQsAk,1611
 validmind/datasets/__init__.py,sha256=oYfcvW7BAyUgpghBOnTeGbQF6tpFAWg38rRirdLr8m8,262
@@ -23,10 +23,10 @@ validmind/datasets/llm/rag/datasets/rfp_existing_questions_client_4.csv,sha256=H
 validmind/datasets/llm/rag/datasets/rfp_existing_questions_client_5.csv,sha256=jJuRVsDbtv3ky8mJVHzpK_4bSlnIZH-hDW6a8DGGvfY,24551
 validmind/datasets/llm/rag/rfp.py,sha256=mlx4I1ipMoPplZOqRPza6XEN5D4FE2xc6_SSuGOkgqk,1234
 validmind/datasets/nlp/__init__.py,sha256=lu2-SGOSECgrxhDtHdlh_FsbpsFtuZ4DbLSL1sww5nc,323
-validmind/datasets/nlp/cnn_dailymail.py,sha256=9gAHFWQhSJJ2JF_zarc2oLErZwPtRyfvXV4molhVzXM,3471
+validmind/datasets/nlp/cnn_dailymail.py,sha256=_N_a19nk7uaGQdDAEpSO-XHshV_8U6BH6s76Pmmi3eI,3692
 validmind/datasets/nlp/datasets/Covid_19.csv,sha256=cwAJHgek3JWUhtZQIiC9-wRWycxbiod6nyZikf09iKE,11545066
-validmind/datasets/nlp/datasets/cnn_dailymail_100_with_predictions.csv,sha256=MSaj0RJ1lbgePDTfpmgganzit64FQ846J1qjm8PJtfI,426676
-validmind/datasets/nlp/datasets/cnn_dailymail_500_with_predictions.csv,sha256=bR8Z-m43lngaFW983ECJwSKea3GlfhZDYfPC0MDJMec,2246787
+validmind/datasets/nlp/datasets/cnn_dailymail_100_with_predictions.csv,sha256=Z-twNp-uRCA736YCQ2FK_95uZUc8nBtKd2ZJh_j6n2U,608859
+validmind/datasets/nlp/datasets/cnn_dailymail_500_with_predictions.csv,sha256=g-TKvZHmnysaE1jW8OR42x119Ilo1S-V6sD2nf1BHyw,3072969
 validmind/datasets/nlp/datasets/sentiments_with_predictions.csv,sha256=zwYI-XjUy72XqD0om-cMVYMrEUsL36u4rPAwsicoQtg,754686
 validmind/datasets/nlp/twitter_covid_19.py,sha256=PtBOh6pri6pVQKpR0ivQfXy4Wcom40G9-dyYgcv0jFI,878
 validmind/datasets/regression/__init__.py,sha256=wBcJyMyX4U6XjY6Y-aLozHdbjQlXbeFgB2YqFTk2kVU,1653
@@ -65,13 +65,13 @@ validmind/logging.py,sha256=J1Y1dYCH1dtkoYCHoXMOQH_B7EO4fJytWRDrDqZZz8U,5204
 validmind/models/__init__.py,sha256=lraTbNwoKckXNP3Dbyj-euI78UTkZ_w5wpUOb8l5nWs,729
 validmind/models/foundation.py,sha256=ZdVmwwRVbjgqMyfjguyf9Lka_KcgJnDD7ho8zv0gQok,1842
 validmind/models/function.py,sha256=xLNtgzRiCfF4jrIedHrX1lmCR-92fB3fVDzLS7el4SM,1785
-validmind/models/huggingface.py,sha256=oDB32iwP_FQ_ZtZgAC5iBPG4suPaSF-J6317TM8Ob2g,2304
+validmind/models/huggingface.py,sha256=DMHekLpWi6c4N0svh-3G0NYYwzxPXOvqiU95M4QffUs,2253
 validmind/models/metadata.py,sha256=PMcdYuACkSPvuG8io5BhZeMwclQr_q79mXbvd1SC-7I,1665
 validmind/models/pipeline.py,sha256=nSskKWxaS4SGmx_B0IAvS5ogDZyh6tdx_aUkyxSXt88,2051
 validmind/models/pytorch.py,sha256=aAEUWtISwLh-PMvHkcLwBEbBStAByt4J-NpK-Ndv38E,1826
 validmind/models/r_model.py,sha256=eYdpCREgBpYv-PxJDuG91I77OOAx3-43FoaYT560ziE,7172
 validmind/models/sklearn.py,sha256=lOCJlP2wvd5IJHtBS1XG9FXrtIvO_f8xm2Qp1UdsiBw,2406
-validmind/template.py,sha256=d5I8TjX8Sh5CJPWo_zezMfyvnh7OXo07VV8iKXxlk9E,7184
+validmind/template.py,sha256=Ja7PHeJsn7jNHop_DwU6tFrsi0bLy87lWZLAO130ZXE,7273
 validmind/test_suites/__init__.py,sha256=T7446YfTnxfBKNFwIXVcbMb4uIyRzHFAyUQLdDQCjVc,7039
 validmind/test_suites/classifier.py,sha256=0ZE3z5X_ZewTvmwQ3cVGJQh7dPgg0IlqcQshJJxCFWQ,4003
 validmind/test_suites/cluster.py,sha256=Wc2NViwivjiuiJMwrnGbOJYeZ3ApN8usWlOPYZYWAgE,2276
@@ -127,7 +127,7 @@ validmind/tests/data_validation/Skewness.py,sha256=MouMHv5wvpUTDueieqIceXG7JZji1
 validmind/tests/data_validation/SpreadPlot.py,sha256=yctCXm_hFVmhbpPywr9GGWf2W8jIbvmyxQIkr6w9sVo,4551
 validmind/tests/data_validation/TabularCategoricalBarPlots.py,sha256=UKGThQmKKxL2SL7siPlY7ODYgooNSrtUFdbLAalh9fU,4201
 validmind/tests/data_validation/TabularDateTimeHistograms.py,sha256=_1FU3kdAN7GI3208OGmlKrAJ8rIrTeRTa8bGNWSmBiI,3947
-validmind/tests/data_validation/TabularDescriptionTables.py,sha256=SEE_Ewxfja2HjJEZnN17f4xTHoKpDAJDU7vK9ZHLyRk,9241
+validmind/tests/data_validation/TabularDescriptionTables.py,sha256=nLv_9XB0d8_5KUUHFPUxzTx_usHxby7qhS3HyrS4nc0,7132
 validmind/tests/data_validation/TabularNumericalHistograms.py,sha256=-WqG6XI2hb5VEVOeNrfpUzyWzw6mcxJvn7wHCI7o4q4,4130
 validmind/tests/data_validation/TargetRateBarPlots.py,sha256=omgPJbA1WZ-5YEiatSdEvpEQ-QCJULU8kDOHAUE610E,5697
 validmind/tests/data_validation/TimeSeriesDescription.py,sha256=YIfet30KZSAOdYAkTQadKLNuY_SAM3a5Fn2z2AqQz7I,3130
@@ -175,21 +175,21 @@ validmind/tests/model_validation/TimeSeriesR2SquareBySegments.py,sha256=ec90bxP9
 validmind/tests/model_validation/TokenDisparity.py,sha256=EZlpFQH6qRWedjTQT5o4u-OIdgj1iKK-JB8GEQQlxoA,4394
 validmind/tests/model_validation/ToxicityScore.py,sha256=nFDHU1Z8mGpJrdKE6sWxo9nOqqzne1JsYIiNFyn_gYA,5299
 validmind/tests/model_validation/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-validmind/tests/model_validation/embeddings/ClusterDistribution.py,sha256=fs4Xrq18qQrxfmMWRlOGVSES48_Y81KDrNrwEArGs50,3512
+validmind/tests/model_validation/embeddings/ClusterDistribution.py,sha256=1Ywx0OTRZ13F0deQKoXzzuxDr31lYWjlC7HZqJY5Ys8,3507
 validmind/tests/model_validation/embeddings/CosineSimilarityComparison.py,sha256=xXM1pI-4XOLLSE-sZqQzdUryrAvql5maytFNjzMlrn4,4900
-validmind/tests/model_validation/embeddings/CosineSimilarityDistribution.py,sha256=zU4H9t2lQVvwhpHXJQKRbG1iGkMMZbw3_JiLFBSHy5E,3419
+validmind/tests/model_validation/embeddings/CosineSimilarityDistribution.py,sha256=uqWgh0Qbc9OQc0RArv6NCHbSNBA9Lko4jybJ_NVdTLY,3414
 validmind/tests/model_validation/embeddings/CosineSimilarityHeatmap.py,sha256=R16Jq2BQPG4xknAmDGEbRcX8RmDb879NWazSPmB0ARo,3495
-validmind/tests/model_validation/embeddings/DescriptiveAnalytics.py,sha256=t_nYQWSO8QpS15PZZmOFje6Ypj2wP2b9U_mX6-M85zU,4207
-validmind/tests/model_validation/embeddings/EmbeddingsVisualization2D.py,sha256=M2bnEKeiuD6cuTr1hDHEaXKUp-FIXP6Q9asioG8ZdCI,4302
+validmind/tests/model_validation/embeddings/DescriptiveAnalytics.py,sha256=urnF9XBMgJ7tmQzHkmYLKCSWrPjFM95PfhIRuO7-F7w,4202
+validmind/tests/model_validation/embeddings/EmbeddingsVisualization2D.py,sha256=vqFF_a4VcBL-iwtG_sgOiK04E00CzL-dmOLHGtFwSEw,4297
 validmind/tests/model_validation/embeddings/EuclideanDistanceComparison.py,sha256=UG5etgWn9Rnqts5X_1U6CTj-uDfz6VLqXdQVZ9Mu3y0,4502
 validmind/tests/model_validation/embeddings/EuclideanDistanceHeatmap.py,sha256=ruMY1LK2LwmzwdKMcgFLeaebhEi4_kbJNpEafVCTPfw,3375
 validmind/tests/model_validation/embeddings/PCAComponentsPairwisePlots.py,sha256=nUQU9e32RHa3Ytofi_dzId_7VSNH4bJG9I_r8VJNbFk,3893
-validmind/tests/model_validation/embeddings/StabilityAnalysis.py,sha256=jP-6oG0VmoWLqiWTJfeGrBtphocwlEYEAs40Xtq5gwU,4539
+validmind/tests/model_validation/embeddings/StabilityAnalysis.py,sha256=2pWHSV8weWX-KmdQ1SAKAFGQiDKff0qgwzH3oy4oVTE,4603
 validmind/tests/model_validation/embeddings/StabilityAnalysisKeyword.py,sha256=exnaM-XHiHHflflXfJQLNGQByTBDeKwCtxBoNPKNev8,3970
 validmind/tests/model_validation/embeddings/StabilityAnalysisRandomNoise.py,sha256=N1hYFnAQKLxSpjd6ZHwi57Zdx10ssEX2Ci73rrEXTGs,5796
 validmind/tests/model_validation/embeddings/StabilityAnalysisSynonyms.py,sha256=npnOPAoXb5FoiwQEwp_gDcbGa5xk4rYnXChTJnuGX64,4405
-validmind/tests/model_validation/embeddings/StabilityAnalysisTranslation.py,sha256=9dL06PRlzokEpaEQAj_12FuQnlzfQ2-__6alYfSajmU,4636
-validmind/tests/model_validation/embeddings/TSNEComponentsPairwisePlots.py,sha256=jfEJjLiBhrXGmafNPI0rk3sDugDmuQ6pbpek4rF8v1k,4512
+validmind/tests/model_validation/embeddings/StabilityAnalysisTranslation.py,sha256=qLydyTQ6mzHOYQzqysjPPe_ltiTsRfPEhZDEDm5XxX8,4825
+validmind/tests/model_validation/embeddings/TSNEComponentsPairwisePlots.py,sha256=ereo_dGf19xqvOGz7zcGwhDRU_UqvjFRi5n4KmGDKl8,4511
 validmind/tests/model_validation/ragas/AnswerCorrectness.py,sha256=PgSxFcyWx2WAI6DsgYJQW3N5EKQgm-uCAZgm8r9Cly4,5002
 validmind/tests/model_validation/ragas/AnswerRelevance.py,sha256=iIgN6CTAYs8V9sfJVhR_p6B5TQLUQ413rXYqwVP3770,4861
 validmind/tests/model_validation/ragas/AnswerSimilarity.py,sha256=csCdKnjKXwIAOcc6PdAWl0IO2ttou2ihoXtU-G37hz4,4321
@@ -211,7 +211,7 @@ validmind/tests/model_validation/sklearn/ConfusionMatrix.py,sha256=wdEr8KopFOgbY
 validmind/tests/model_validation/sklearn/FeatureImportanceComparison.py,sha256=zUXNy9SookZQQfLvwvI2W0yG6BlRBdWIK0IHgYZKNpg,3250
 validmind/tests/model_validation/sklearn/FowlkesMallowsScore.py,sha256=OxDyofYDNjPqhNrYg4_1NbDvE4J7W6QR-Mzn17NlIkY,2997
 validmind/tests/model_validation/sklearn/HomogeneityScore.py,sha256=VWRVfrLUptIxCzsZJRNe1p3lvE8Xv4sfC9r-dqyUmIc,2701
-validmind/tests/model_validation/sklearn/HyperParametersTuning.py,sha256=VbTgQ1NBfoHUAVvxfLkwu4cHy9_quSN4ywwD49cUH_U,4585
+validmind/tests/model_validation/sklearn/HyperParametersTuning.py,sha256=5a4K15UlecxFnKYVsgPksf29LjvdfHWPGT2t5-2JO9A,4575
 validmind/tests/model_validation/sklearn/KMeansClustersOptimization.py,sha256=lmSktn7usYZt-2v8Ykrig-9vEM910AfUdRkyOrsZd3U,5906
 validmind/tests/model_validation/sklearn/MinimumAccuracy.py,sha256=5KSAd29dbKs3nNJ734t1j806Td3H5LRIw950g3Xp2nc,5277
 validmind/tests/model_validation/sklearn/MinimumF1Score.py,sha256=TaLHk98CwQigyt17L1uBBLC25D5J_IKb6a_IFJFO7AE,4618
@@ -266,7 +266,7 @@ validmind/tests/prompt_validation/Robustness.py,sha256=fBdkYnO9yoBazz4wD-l62tT8D
 validmind/tests/prompt_validation/Specificity.py,sha256=h3gKRTTi2rfnGWmGC1YnSt2s_VbZU4KX0iY7LciZ3PU,6068
 validmind/tests/prompt_validation/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 validmind/tests/prompt_validation/ai_powered_test.py,sha256=7TTeIR5GotQosm7oVT8Y3KnwPB3XkVT1Fzhckpr-SgE,1963
-validmind/tests/run.py,sha256=H9LRkJ4mZFqKUje31Lz0J0VL-DVYAyztdvpOa3NaAoA,14319
+validmind/tests/run.py,sha256=OI0SxHvJGH0L2o7LyiK-723Cui1UFAu30YxG52ooU7M,14305
 validmind/tests/test_providers.py,sha256=47xe5eb5ufvj1jmhdRsbSvDQTXSDpFDFNeXg3xtXwhw,5320
 validmind/tests/utils.py,sha256=kNrxfUYbj4DwmkZtpp_1rG4GMUGxYEhvqnYR_A7qAKM,471
 validmind/unit_metrics/__init__.py,sha256=AlFnWA9pmzVf8xysPxYpQ3kBTQ81-YVxRTJpgC0Q41w,7344
@@ -290,7 +290,7 @@ validmind/unit_metrics/regression/sklearn/RootMeanSquaredError.py,sha256=_5IQIU9
 validmind/utils.py,sha256=MQDsW7YuwEJ50tA01n3xb8D_Ihmji_Mn22AlMnJJQT8,15819
 validmind/vm_models/__init__.py,sha256=lmWCD2u4tW6_AH39UnJ24sCcMUcsHbUttz7SaZfrh3s,1168
 validmind/vm_models/dataset/__init__.py,sha256=U4CxZjdoc0dd9u2AqBl5PJh1UVbzXWNrmundmjLF-qE,346
-validmind/vm_models/dataset/dataset.py,sha256=YP6l5sq7SJNExWK3RvkxeCBTLj4z2GkBsmv4KxfBS1I,22753
+validmind/vm_models/dataset/dataset.py,sha256=X0vKp1NuL0k5WLILAAlmnPs_WHR8Ji6ovlfWLjJs3qk,23305
 validmind/vm_models/dataset/utils.py,sha256=ygT6hUw0KklKCboo7tqLxh_hf-dEiaccVyCpR9DCPF8,5177
 validmind/vm_models/figure.py,sha256=iSrvPcCG5sQrMkX1Fh6c5utRzaroh3bc6IlnGDOK_Eg,6651
 validmind/vm_models/model.py,sha256=b-UL73EWOpj-X5aQbHQ3HLkONHCH9hYwUlKxVwPC6gI,6088
@@ -304,11 +304,11 @@ validmind/vm_models/test/threshold_test.py,sha256=xSadM5t9Z-XZjkxu7LKmeljy2bdwTw
 validmind/vm_models/test/threshold_test_result.py,sha256=EXP-g_e3NsnpkvNgYew030qVUoY6ZTHyuuFUXaq-BuM,1954
 validmind/vm_models/test_context.py,sha256=AN7-atBgOcD04MLVitCFJYooxF6_iNmvI2H4nkv32iw,9035
 validmind/vm_models/test_suite/runner.py,sha256=aewxadRfoOPH48jes2Gtb3Ju_FWFfVM_9ARIAJHD4wA,6982
-validmind/vm_models/test_suite/summary.py,sha256=co-xJJMUYGb7cOiVmw0i8vpZlfiMqrWjaCOmHKMAbcE,4686
+validmind/vm_models/test_suite/summary.py,sha256=GQRNe2ZvvqjQN0yKmaN7ohAUjRFQIN4YYUYxfOuWN6M,4682
 validmind/vm_models/test_suite/test.py,sha256=_GfbK36l98SjzgVcucmp0OKBJKqMW3neO7SqJ3EWeps,5049
 validmind/vm_models/test_suite/test_suite.py,sha256=Cns2wL54v0T5Mv5_HJb3kMeaa4rtycdqT8KxK9_rWEU,6279
-validmind-2.4.0.dist-info/LICENSE,sha256=XonPUfwjvrC5Ombl3y-ko0Wubb1xdG_7nzvIbkZRKHw,35772
-validmind-2.4.0.dist-info/METADATA,sha256=aN9HlbzIlfBx98XfsOSYr2Y0ob1UjLSJuBG2E9g3xVM,4133
-validmind-2.4.0.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
-validmind-2.4.0.dist-info/entry_points.txt,sha256=HuW7YyOv9u_OEWpViQXtv0nfoI67uieJHawKWA4Hv9A,76
-validmind-2.4.0.dist-info/RECORD,,
+validmind-2.4.4.dist-info/LICENSE,sha256=XonPUfwjvrC5Ombl3y-ko0Wubb1xdG_7nzvIbkZRKHw,35772
+validmind-2.4.4.dist-info/METADATA,sha256=keXWOaqBlRbo2FO8FDtwgd37iscJMcloL1A3qmY6NGk,4240
+validmind-2.4.4.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
+validmind-2.4.4.dist-info/entry_points.txt,sha256=HuW7YyOv9u_OEWpViQXtv0nfoI67uieJHawKWA4Hv9A,76
+validmind-2.4.4.dist-info/RECORD,,