validmind 2.13.12__tar.gz → 2.13.14__tar.gz

{validmind-2.13.12 → validmind-2.13.14}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: validmind
-Version: 2.13.12
+Version: 2.13.14
 Summary: ValidMind Library
 Author-email: Andres Rodriguez <andres@validmind.ai>, Juan Martinez <juan@validmind.ai>, Anil Sorathiya <anil@validmind.ai>, Luis Pallares <luis@validmind.ai>, John Walz <john@validmind.ai>
 License:                          DUAL LICENSE NOTICE
@@ -708,7 +708,7 @@ Requires-Dist: plotly>=6.0.0
 Requires-Dist: polars
 Requires-Dist: python-dotenv
 Requires-Dist: requests<3.0.0,>=2.28.0
-Requires-Dist: scikit-learn
+Requires-Dist: scikit-learn>=1.4
 Requires-Dist: seaborn
 Requires-Dist: segno<2.0.0,>=1.6.6
 Requires-Dist: tabulate<0.10.0,>=0.9.0
@@ -735,7 +735,7 @@ Requires-Dist: pyarrow<23,>=22; (python_version >= '3.14') and extra == 'all'
 Requires-Dist: pycocoevalcap; extra == 'all'
 Requires-Dist: ragas<=0.2.7,>=0.2.3; extra == 'all'
 Requires-Dist: rouge>=1; extra == 'all'
-Requires-Dist: scikit-learn<1.8; extra == 'all'
+Requires-Dist: scikit-learn<1.8,>=1.4; extra == 'all'
 Requires-Dist: scipy; extra == 'all'
 Requires-Dist: scorecardpy==0.1.9.6; extra == 'all'
 Requires-Dist: sentencepiece<0.3.0,>=0.2.0; extra == 'all'
@@ -803,7 +803,7 @@ Requires-Dist: arch>=7.0.0; extra == 'stats'
 Requires-Dist: scipy; extra == 'stats'
 Requires-Dist: statsmodels<0.15.0,>=0.14.2; extra == 'stats'
 Provides-Extra: xgboost
-Requires-Dist: scikit-learn<1.8; extra == 'xgboost'
+Requires-Dist: scikit-learn<1.8,>=1.4; extra == 'xgboost'
 Requires-Dist: xgboost<3.1,>=1.5.2; extra == 'xgboost'
 Description-Content-Type: text/markdown
 

{validmind-2.13.12 → validmind-2.13.14}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "validmind"
-version = "2.13.12"
+version = "2.13.14"
 description = "ValidMind Library"
 readme = "README.pypi.md"
 requires-python = ">=3.9,<3.15"
@@ -27,7 +27,7 @@ dependencies = [
   "plotly (>=6.0.0)",
   "polars",
   "python-dotenv",
-  "scikit-learn",
+  "scikit-learn (>=1.4)",
   "seaborn",
   "tabulate (>=0.9.0,<0.10.0)",
   "tiktoken",
@@ -41,7 +41,7 @@ dependencies = [
 all = [
   "torch (>=2.0.0)",
   "xgboost (>=1.5.2,<3.1)",
-  "scikit-learn (<1.8)",
+  "scikit-learn (>=1.4,<1.8)",
   "transformers (>=4.32.0,<5.0.0)",
   "pycocoevalcap",
   "ragas (>=0.2.3,<=0.2.7)",
@@ -111,7 +111,7 @@ pytorch = ["torch (>=2.0.0)"]
 stats = ["scipy", "statsmodels (>=0.14.2,<0.15.0)", "arch (>=7.0.0)"]
 xgboost = [
   "xgboost (>=1.5.2,<3.1)",
-  "scikit-learn (<1.8)",
+  "scikit-learn (>=1.4,<1.8)",
 ]
 explainability = [
   "shap (>=0.46.0)",

validmind-2.13.14/validmind/__version__.py

@@ -0,0 +1 @@
+__version__ = "2.13.14"

{validmind-2.13.12 → validmind-2.13.14}/validmind/ai/test_descriptions.py

@@ -171,7 +171,9 @@ def generate_description(
             "figures": [figure._get_b64_url() for figure in figures or []],
             "additional_context": additional_context,
             "instructions": instructions,
-            "params": params,
+            # params can hold anything the test accepts (DataFrames, arrays);
+            # round-trip so the request body is plain JSON
+            "params": json.loads(json.dumps(params, cls=NumpyEncoder)),
         }
     )["content"]
 

{validmind-2.13.12 → validmind-2.13.14}/validmind/datasets/credit_risk/lending_club.py

@@ -322,7 +322,9 @@ def woe_encoding(df: pd.DataFrame, verbose: bool = True) -> pd.DataFrame:
             print(f"Excluded {target_column} from WoE transformation.")
 
     # Apply the WoE transformation
-    df = sc.woebin_ply(df, bins=bins)
+    with warnings.catch_warnings():
+        warnings.simplefilter("ignore")  # scorecardpy uses deprecated pandas APIs
+        df = sc.woebin_ply(df, bins=bins)
 
     if verbose:
         print("Successfully converted features to WoE values.")
@@ -379,7 +381,9 @@ def _woebin(df: pd.DataFrame, verbose: bool = True) -> Dict[str, Any]:
             print(
                 f"Performing binning with breaks_adj: {breaks_adj}"
             )  # print the breaks_adj being used
-        bins = sc.woebin(df, target_column, breaks_list=breaks_adj)
+        with warnings.catch_warnings():
+            warnings.simplefilter("ignore")  # scorecardpy uses deprecated pandas APIs
+            bins = sc.woebin(df, target_column, breaks_list=breaks_adj)
     except Exception as e:
         print("Error during binning: ")
         print(e)

{validmind-2.13.12 → validmind-2.13.14}/validmind/tests/model_validation/sklearn/HyperParametersTuning.py

@@ -43,7 +43,7 @@ def _create_scoring_dict(scoring, metrics, threshold):
     for metric in metrics:
         if metric == "recall":
             scoring_dict[metric] = make_scorer(
-                custom_recall, needs_proba=True, threshold=threshold
+                custom_recall, response_method="predict_proba", threshold=threshold
             )
         elif metric == "roc_auc":
             scoring_dict[metric] = "roc_auc"

{validmind-2.13.12 → validmind-2.13.14}/validmind/tests/model_validation/sklearn/ScoreProbabilityAlignment.py

@@ -81,7 +81,7 @@ def ScoreProbabilityAlignment(
 
     # Calculate statistics per bin
     results = []
-    for bin_name, group in df.groupby("score_bin"):
+    for bin_name, group in df.groupby("score_bin", observed=True):
         bin_stats = {
             "Score Range": f"{bin_name.left:.0f}-{bin_name.right:.0f}",
             "Mean Score": group[score_column].mean(),

{validmind-2.13.12 → validmind-2.13.14}/validmind/tests/model_validation/sklearn/WeakspotsDiagnosis.py

@@ -285,14 +285,22 @@ def WeakspotsDiagnosis(
     figures = []
     passed = True
 
-    df_1 = datasets[0]._df[
-        feature_columns
-        + [datasets[0].target_column, datasets[0].prediction_column(model)]
-    ]
-    df_2 = datasets[1]._df[
-        feature_columns
-        + [datasets[1].target_column, datasets[1].prediction_column(model)]
-    ]
+    df_1 = (
+        datasets[0]
+        ._df[
+            feature_columns
+            + [datasets[0].target_column, datasets[0].prediction_column(model)]
+        ]
+        .copy()
+    )
+    df_2 = (
+        datasets[1]
+        ._df[
+            feature_columns
+            + [datasets[1].target_column, datasets[1].prediction_column(model)]
+        ]
+        .copy()
+    )
     results_1 = pd.DataFrame()
     results_2 = pd.DataFrame()
     for feature in feature_columns:

{validmind-2.13.12 → validmind-2.13.14}/validmind/tests/plots/BoxPlot.py

@@ -4,7 +4,6 @@
 
 from typing import List, Optional
 
-import pandas as pd
 import plotly.graph_objects as go
 from plotly.subplots import make_subplots
 
@@ -18,21 +17,12 @@ def _validate_inputs(
 ):
     """Validate inputs and return validated columns."""
 
-    # Get dtypes without loading data into memory
-    if not isinstance(columns, list):
-        columns = [columns]
-
-    columns_dtypes = dataset._df[columns].dtypes
-
-    columns_numeric = []
-    columns_numeric = columns_dtypes[
-        columns_dtypes.apply(lambda x: pd.api.types.is_numeric_dtype(x))
-    ].index.tolist()
-
     if columns is None:
-        columns = columns_numeric
+        columns = list(dataset.feature_columns_numeric)
     else:
-        available_columns = set(columns_numeric)
+        if not isinstance(columns, list):
+            columns = [columns]
+        available_columns = set(dataset.feature_columns_numeric)
         columns = [col for col in columns if col in available_columns]
 
     if not columns:

{validmind-2.13.12 → validmind-2.13.14}/validmind/vm_models/result/result.py

@@ -542,6 +542,18 @@ class TestResult(Result):
 
         return serialized
 
+    def _pre_serialize_figures(self):
+        """Render figure PNGs before any upload request starts.
+
+        Kaleido launches a Chromium per figure and blocks the event loop while it
+        runs. Rendering while uploads are in flight lets the request timeout
+        expire on them, and with older kaleido it conflicted with asyncio
+        outright (ZD-626). log() calls this outside the loop; log_async() calls
+        it too because the test-suite runner awaits log_async() directly.
+        """
+        for figure in self.figures or []:
+            figure.pre_serialize()
+
     async def log_async(
         self,
         section_id: str = None,
@@ -553,6 +565,8 @@ class TestResult(Result):
         if self._is_scorer_result:
             return
 
+        self._pre_serialize_figures()
+
         tasks = []  # collect tasks to run in parallel (async)
 
         # Default empty dict if None
@@ -680,10 +694,7 @@ class TestResult(Result):
         if section_id:
             self._validate_section_id_for_block(section_id, position)
 
-        # Pre-serialize figures before entering async context to avoid conflicts
-        # between Plotly's kaleido library and asyncio event loops (ZD-626)
-        for figure in self.figures or []:
-            figure.pre_serialize()
+        self._pre_serialize_figures()
 
         run_async(
             self.log_async,

validmind-2.13.12/validmind/__version__.py

@@ -1 +0,0 @@
-__version__ = "2.13.12"