upgini 1.2.111__py3-none-any.whl → 1.2.113a1__py3-none-any.whl

upgini/__about__.py

@@ -1 +1 @@
-__version__ = "1.2.111"
+__version__ = "1.2.113a1"

upgini/dataset.py

@@ -50,7 +50,7 @@ except Exception:
 
 class Dataset:
     MIN_ROWS_COUNT = 100
-    MAX_ROWS = 200_000
+    MAX_ROWS = 100_000
     IMBALANCE_THESHOLD = 0.6
     MIN_TARGET_CLASS_ROWS = 100
     MAX_MULTICLASS_CLASS_COUNT = 100

upgini/features_enricher.py

@@ -415,6 +415,7 @@ class FeaturesEnricher(TransformerMixin):
         y: Union[pd.Series, np.ndarray, List],
         eval_set: Optional[Union[List[tuple], tuple]] = None,
         *args,
+        oot: Union[pd.DataFrame, pd.Series, np.ndarray, None] = None,
         exclude_features_sources: Optional[List[str]] = None,
         calculate_metrics: Optional[bool] = None,
         estimator: Optional[Any] = None,
@@ -443,6 +444,9 @@ class FeaturesEnricher(TransformerMixin):
         eval_set: List[tuple], optional (default=None)
             List of pairs (X, y) for validation.
 
+        oot: pandas.DataFrame of shape (n_samples, n_features)
+            Out of time data.
+
         importance_threshold: float, optional (default=None)
             Minimum SHAP value to select a feature. Default value is 0.0.
 
@@ -508,7 +512,8 @@ class FeaturesEnricher(TransformerMixin):
                     X,
                     y,
                     self.eval_set,
-                    progress_bar,
+                    oot=oot,
+                    progress_bar=progress_bar,
                     start_time=start_time,
                     exclude_features_sources=exclude_features_sources,
                     calculate_metrics=calculate_metrics,
@@ -563,6 +568,7 @@ class FeaturesEnricher(TransformerMixin):
         y: Union[pd.DataFrame, pd.Series, np.ndarray, List],
         eval_set: Optional[Union[List[tuple], tuple]] = None,
         *args,
+        oot: Union[pd.DataFrame, pd.Series, np.ndarray, None] = None,
         exclude_features_sources: Optional[List[str]] = None,
         keep_input: bool = True,
         importance_threshold: Optional[float] = None,
@@ -667,7 +673,8 @@ class FeaturesEnricher(TransformerMixin):
                     X,
                     y,
                     self.eval_set,
-                    progress_bar,
+                    oot=oot,
+                    progress_bar=progress_bar,
                     start_time=start_time,
                     exclude_features_sources=exclude_features_sources,
                     calculate_metrics=calculate_metrics,
@@ -940,7 +947,7 @@ class FeaturesEnricher(TransformerMixin):
             ):
                 raise ValidationError(self.bundle.get("metrics_unfitted_enricher"))
 
-            validated_X, validated_y, validated_eval_set = self._validate_train_eval(
+            validated_X, validated_y, validated_eval_set, _ = self._validate_train_eval(
                 effective_X, effective_y, effective_eval_set
             )
 
@@ -1535,7 +1542,7 @@ class FeaturesEnricher(TransformerMixin):
         is_input_same_as_fit, X, y, eval_set = self._is_input_same_as_fit(X, y, eval_set)
         is_demo_dataset = hash_input(X, y, eval_set) in DEMO_DATASET_HASHES
         checked_eval_set = self._check_eval_set(eval_set, X, self.bundle)
-        validated_X, validated_y, validated_eval_set = self._validate_train_eval(X, y, checked_eval_set)
+        validated_X, validated_y, validated_eval_set, _ = self._validate_train_eval(X, y, checked_eval_set)
 
         sampled_data = self._get_enriched_for_metrics(
             trace_id,
@@ -1931,11 +1938,8 @@ class FeaturesEnricher(TransformerMixin):
         )
 
         # Handle eval sets extraction based on EVAL_SET_INDEX
-        if EVAL_SET_INDEX in enriched_Xy.columns:
-            eval_set_indices = list(enriched_Xy[EVAL_SET_INDEX].unique())
-            if 0 in eval_set_indices:
-                eval_set_indices.remove(0)
-            for eval_set_index in eval_set_indices:
+        if EVAL_SET_INDEX in enriched_Xy.columns and eval_set is not None:
+            for eval_set_index in range(1, len(eval_set) + 1):
                 enriched_eval_sets[eval_set_index] = enriched_Xy.loc[
                     enriched_Xy[EVAL_SET_INDEX] == eval_set_index
                 ].copy()
@@ -2047,7 +2051,11 @@ class FeaturesEnricher(TransformerMixin):
         )
 
     def __combine_train_and_eval_sets(
-        self, X: pd.DataFrame, y: Optional[pd.Series] = None, eval_set: Optional[List[tuple]] = None
+        self,
+        X: pd.DataFrame,
+        y: Optional[pd.Series] = None,
+        eval_set: Optional[List[tuple]] = None,
+        oot: Optional[pd.DataFrame] = None,
     ) -> pd.DataFrame:
         df = X.copy()
         if y is not None:
@@ -2063,6 +2071,11 @@ class FeaturesEnricher(TransformerMixin):
             eval_df_with_index[TARGET] = eval_y
             eval_df_with_index[EVAL_SET_INDEX] = idx + 1
             df = pd.concat([df, eval_df_with_index])
+        
+        if oot is not None:
+            oot_df_with_index = oot.copy()
+            oot_df_with_index[EVAL_SET_INDEX] = -1
+            df = pd.concat([df, oot_df_with_index])
 
         return df
 
@@ -2115,12 +2128,12 @@ class FeaturesEnricher(TransformerMixin):
     ) -> Dict[int, Tuple]:
         eval_set_sampled_dict = {}
 
-        for idx in range(eval_set_len):
-            enriched_eval_xy = enriched_df.query(f"{EVAL_SET_INDEX} == {idx + 1}")
+        for idx in range(1, eval_set_len + 1):
+            enriched_eval_xy = enriched_df.query(f"{EVAL_SET_INDEX} == {idx}")
             eval_x_sampled = enriched_eval_xy[x_columns].copy()
             eval_y_sampled = enriched_eval_xy[TARGET].copy()
             enriched_eval_x = enriched_eval_xy[enriched_X_columns].copy()
-            eval_set_sampled_dict[idx] = (eval_x_sampled, enriched_eval_x, eval_y_sampled)
+            eval_set_sampled_dict[idx - 1] = (eval_x_sampled, enriched_eval_x, eval_y_sampled)
 
         return eval_set_sampled_dict
 
@@ -2312,10 +2325,10 @@ if response.status_code == 200:
         with MDC(trace_id=trace_id, search_id=search_id):
             self.logger.info("Start transform")
 
-            validated_X, validated_y, validated_eval_set = self._validate_train_eval(
-                X, y, eval_set=None, is_transform=True
+            validated_X, validated_y, _, _ = self._validate_train_eval(
+                X, y, is_transform=True
             )
-            df = self.__combine_train_and_eval_sets(validated_X, validated_y, validated_eval_set)
+            df = self.__combine_train_and_eval_sets(validated_X, validated_y)
 
             validated_Xy = df.copy()
 
@@ -2790,9 +2803,10 @@ if response.status_code == 200:
         X: Union[pd.DataFrame, pd.Series, np.ndarray],
         y: Union[pd.DataFrame, pd.Series, np.ndarray, List, None],
         eval_set: Optional[List[tuple]],
+        *,
+        oot: Union[pd.DataFrame, pd.Series, np.ndarray, None] = None,
         progress_bar: Optional[ProgressBar],
         start_time: int,
-        *,
         exclude_features_sources: Optional[List[str]] = None,
         calculate_metrics: Optional[bool],
         scoring: Union[Callable, str, None],
@@ -2813,7 +2827,7 @@ if response.status_code == 200:
         self.fit_dropped_features = set()
         self.fit_generated_features = []
 
-        validated_X, validated_y, validated_eval_set = self._validate_train_eval(X, y, eval_set)
+        validated_X, validated_y, validated_eval_set, validated_oot = self._validate_train_eval(X, y, eval_set, oot)
 
         is_demo_dataset = hash_input(validated_X, validated_y, validated_eval_set) in DEMO_DATASET_HASHES
         if is_demo_dataset:
@@ -2854,6 +2868,7 @@ if response.status_code == 200:
             validated_X,
             validated_y,
             validated_eval_set,
+            validated_oot,
             exclude_features_sources=exclude_features_sources,
             calculate_metrics=calculate_metrics,
             scoring=scoring,
@@ -2861,7 +2876,7 @@ if response.status_code == 200:
             remove_outliers_calc_metrics=remove_outliers_calc_metrics,
         )
 
-        df = self.__combine_train_and_eval_sets(validated_X, validated_y, validated_eval_set)
+        df = self.__combine_train_and_eval_sets(validated_X, validated_y, validated_eval_set, validated_oot)
         self.id_columns_encoder = OrdinalEncoder().fit(df[self.id_columns or []])
 
         self.fit_search_keys = self.search_keys.copy()
@@ -3288,12 +3303,14 @@ if response.status_code == 200:
         X: pd.DataFrame,
         y: Optional[pd.Series] = None,
         eval_set: Optional[List[Tuple[pd.DataFrame, pd.Series]]] = None,
+        oot: Union[pd.DataFrame, pd.Series, np.ndarray, None] = None,
         is_transform: bool = False,
     ) -> Tuple[pd.DataFrame, pd.Series, Optional[List[Tuple[pd.DataFrame, pd.Series]]]]:
         validated_X = self._validate_X(X, is_transform)
         validated_y = self._validate_y(validated_X, y, enforce_y=not is_transform)
         validated_eval_set = self._validate_eval_set(validated_X, eval_set)
-        return validated_X, validated_y, validated_eval_set
+        validated_oot = self._validate_oot(validated_X, oot)
+        return validated_X, validated_y, validated_eval_set, validated_oot
 
     def _encode_id_columns(
         self,
@@ -3429,6 +3446,49 @@ if response.status_code == 200:
             return None
         return [self._validate_eval_set_pair(X, eval_pair) for eval_pair in eval_set]
 
+    def _validate_oot(self, X: pd.DataFrame, oot: Optional[pd.DataFrame]):
+        if oot is None:
+            return None
+
+        if _num_samples(oot) == 0:
+            raise ValidationError(self.bundle.get("oot_is_empty"))
+        if isinstance(oot, pd.DataFrame):
+            if isinstance(oot.columns, pd.MultiIndex) or isinstance(oot.index, pd.MultiIndex):
+                raise ValidationError(self.bundle.get("oot_multiindex_unsupported"))
+            validated_oot = oot.copy()
+        elif isinstance(oot, pd.Series):
+            validated_oot = oot.to_frame()
+        elif isinstance(oot, (list, np.ndarray)):
+            validated_oot = pd.DataFrame(oot)
+            renaming = {c: str(c) for c in validated_oot.columns}
+            validated_oot = validated_oot.rename(columns=renaming)
+        else:
+            raise ValidationError(self.bundle.get("unsupported_type_oot").format(type(oot)))
+
+        if not validated_oot.index.is_unique:
+            raise ValidationError(self.bundle.get("non_unique_index_oot"))
+
+        if self.exclude_columns is not None:
+            validated_oot = validated_oot.drop(columns=self.exclude_columns, errors="ignore")
+
+        if self.baseline_score_column:
+            validated_oot[self.baseline_score_column] = validated_oot[self.baseline_score_column].astype(
+                "float64", errors="ignore"
+            )
+
+        if validated_oot.columns.to_list() != X.columns.to_list():
+            if set(validated_oot.columns.to_list()) == set(X.columns.to_list()):
+                validated_oot = validated_oot[X.columns.to_list()]
+            else:
+                raise ValidationError(self.bundle.get("oot_and_x_diff_shape"))
+
+        # Check for duplicates between train and eval sets by comparing all values
+        train_eval_intersection = pd.merge(X, validated_oot, how="inner")
+        if len(train_eval_intersection) > 0:
+            raise ValidationError(self.bundle.get("oot_has_train_samples"))
+
+        return validated_oot
+
     def _validate_eval_set_pair(self, X: pd.DataFrame, eval_pair: Tuple) -> Tuple[pd.DataFrame, pd.Series]:
         if len(eval_pair) != 2:
             raise ValidationError(self.bundle.get("eval_set_invalid_tuple_size").format(len(eval_pair)))
@@ -3600,6 +3660,7 @@ if response.status_code == 200:
         X: pd.DataFrame,
         y: Union[pd.Series, np.ndarray, list, None] = None,
         eval_set: Optional[List[tuple]] = None,
+        oot: Optional[pd.DataFrame] = None,
         exclude_features_sources: Optional[List[str]] = None,
         calculate_metrics: Optional[bool] = None,
         cv: Optional[Any] = None,
@@ -3668,6 +3729,8 @@ if response.status_code == 200:
                         self.logger.info(
                             f"First 10 rows of the eval_y_{idx} with shape {_num_samples(eval_y)}:\n{sample(eval_y)}"
                         )
+                if oot is not None:
+                    self.logger.info(f"First 10 rows of the oot with shape {oot.shape}:\n{sample(oot)}")
 
             do_without_pandas_limits(print_datasets_sample)
 
@@ -4577,7 +4640,7 @@ if response.status_code == 200:
             print(msg)
 
     def _validate_PSI(self, df: pd.DataFrame):
-        if EVAL_SET_INDEX in df.columns:
+        if EVAL_SET_INDEX in df.columns and (df[EVAL_SET_INDEX] == 1).any():
             train = df.query(f"{EVAL_SET_INDEX} == 0")
             eval1 = df.query(f"{EVAL_SET_INDEX} == 1")
         else:

upgini/resource_bundle/strings.properties

@@ -139,6 +139,13 @@ eval_x_is_empty=X in eval_set is empty.
 eval_y_is_empty=y in eval_set is empty.
 x_and_eval_x_diff_types=X and eval_set X has different types: {} and {}
 eval_x_has_train_samples=Eval set X has rows that are present in train set X
+# OOT
+oot_is_empty=Out of time data is empty
+oot_multiindex_unsupported=Multi index in out of time data is not supported
+unsupported_type_oot=Unsupported type of out of time data: {}. Use pandas.DataFrame, pandas.Series or numpy.ndarray
+non_unique_index_oot=Out of time data has non unique index. Use reset_index
+oot_and_x_diff_shape=Out of time data has different columns than train set X
+oot_has_train_samples=Out of time data has rows that are present in train set X
 
 baseline_score_column_not_exists=baseline_score_column {} doesn't exist in input dataframe
 baseline_score_column_has_na=baseline_score_column contains NaN. Clear it and and retry

upgini/utils/sample_utils.py

@@ -22,10 +22,10 @@ MULTICLASS_MIN_SAMPLE_THRESHOLD = 25_000
 BINARY_BOOTSTRAP_LOOPS = 5
 MULTICLASS_BOOTSTRAP_LOOPS = 2
 
-FIT_SAMPLE_THRESHOLD = 200_000
-FIT_SAMPLE_ROWS = 200_000
-FIT_SAMPLE_ROWS_WITH_EVAL_SET = 200_000
-FIT_SAMPLE_THRESHOLD_WITH_EVAL_SET = 200_000
+FIT_SAMPLE_THRESHOLD = 100_000
+FIT_SAMPLE_ROWS = 100_000
+FIT_SAMPLE_ROWS_WITH_EVAL_SET = 100_000
+FIT_SAMPLE_THRESHOLD_WITH_EVAL_SET = 100_000
 
 
 @dataclass

{upgini-1.2.111.dist-info → upgini-1.2.113a1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: upgini
-Version: 1.2.111
+Version: 1.2.113a1
 Summary: Intelligent data search & enrichment for Machine Learning
 Project-URL: Bug Reports, https://github.com/upgini/upgini/issues
 Project-URL: Homepage, https://upgini.com/

{upgini-1.2.111.dist-info → upgini-1.2.113a1.dist-info}/RECORD

@@ -1,9 +1,9 @@
-upgini/__about__.py,sha256=4KgOTmI-DrbwzTYsi8NTo4T_nHot9MIavj4EH29sgYA,24
+upgini/__about__.py,sha256=H2pDgAfR-AAZibgkBF4HysX4COuPa8QPX6H6srUsYKU,26
 upgini/__init__.py,sha256=LXSfTNU0HnlOkE69VCxkgIKDhWP-JFo_eBQ71OxTr5Y,261
 upgini/ads.py,sha256=nvuRxRx5MHDMgPr9SiU-fsqRdFaBv8p4_v1oqiysKpc,2714
-upgini/dataset.py,sha256=e6JDYTZ2AwC5aF-dqclKZKkiKrHo2f6cFmMQO2ZZmjM,32724
+upgini/dataset.py,sha256=xFi0a-A3uvtxVwFM6JOyitkEPd1I2slIBj5SWfys3hQ,32724
 upgini/errors.py,sha256=2b_Wbo0OYhLUbrZqdLIx5jBnAsiD1Mcenh-VjR4HCTw,950
-upgini/features_enricher.py,sha256=rfVdHgUYEq9saqhWcI04jUmNQcAAn5Kto4w3WpxlOpA,221762
+upgini/features_enricher.py,sha256=lWcxE606_hSGXcFG3lXKzWCwNaAVHN0_j9AUWi1CTkc,224671
 upgini/http.py,sha256=zeAZvT6IAzOs9jQ3WG8mJBANLajgvv2LZePFzKz004w,45482
 upgini/metadata.py,sha256=9_0lFEWPpIHRBW-xWYSEcwPzICTC6_bQ6dUUlE75Xns,12773
 upgini/metrics.py,sha256=V2SP6NS5bfFHzRqufeKVsCXME1yG4t_8Dmk2E3zKdYk,45715
@@ -38,7 +38,7 @@ upgini/normalizer/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU
 upgini/normalizer/normalize_utils.py,sha256=mDh2mBW3aQMB4EFP2aHbf2dGMVkOcWnp4sKKvKDBh8w,8511
 upgini/resource_bundle/__init__.py,sha256=S5F2G47pnJd2LDpmFsjDqEwiKkP8Hm-hcseDbMka6Ko,8345
 upgini/resource_bundle/exceptions.py,sha256=5fRvx0_vWdE1-7HcSgF0tckB4A9AKyf5RiinZkInTsI,621
-upgini/resource_bundle/strings.properties,sha256=NyxRwzehkrL5LMoVyjkhN811MvalepavNfjlC9ubE0Q,28677
+upgini/resource_bundle/strings.properties,sha256=2ohGjki3qn1fkRKDCFm8Hy1DU-3HEP_b3AVHBJka4vg,29147
 upgini/resource_bundle/strings_widget.properties,sha256=gOdqvZWntP2LCza_tyVk1_yRYcG4c04K9sQOAVhF_gw,1577
 upgini/sampler/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 upgini/sampler/base.py,sha256=7GpjYqjOp58vYcJLiX__1R5wjUlyQbxvHJ2klFnup_M,6389
@@ -64,14 +64,14 @@ upgini/utils/mstats.py,sha256=u3gQVUtDRbyrOQK6V1UJ2Rx1QbkSNYGjXa6m3Z_dPVs,6286
 upgini/utils/phone_utils.py,sha256=IrbztLuOJBiePqqxllfABWfYlfAjYevPhXKipl95wUI,10432
 upgini/utils/postal_code_utils.py,sha256=5M0sUqH2DAr33kARWCTXR-ACyzWbjDq_-0mmEml6ZcU,1716
 upgini/utils/progress_bar.py,sha256=N-Sfdah2Hg8lXP_fV9EfUTXz_PyRt4lo9fAHoUDOoLc,1550
-upgini/utils/sample_utils.py,sha256=ETLPKQU_YngiYbdlnEoF2h7QS-3oN8et54q3Qs2ZAbA,15417
+upgini/utils/sample_utils.py,sha256=lZJ4yf9Jiq9Em2Ny9m3RIiF7WSxBPrc4E3xxn_8sQk8,15417
 upgini/utils/sklearn_ext.py,sha256=jLJWAKkqQinV15Z4y1ZnsN3c-fKFwXTsprs00COnyVU,49315
 upgini/utils/sort.py,sha256=8uuHs2nfSMVnz8GgvbOmgMB1PgEIZP1uhmeRFxcwnYw,7039
 upgini/utils/target_utils.py,sha256=i3Xt5l9ybB2_nF_ma5cfPuL3OeFTs2dY2xDI0p4Azpg,9049
 upgini/utils/track_info.py,sha256=G5Lu1xxakg2_TQjKZk4b5SvrHsATTXNVV3NbvWtT8k8,5663
 upgini/utils/ts_utils.py,sha256=26vhC0pN7vLXK6R09EEkMK3Lwb9IVPH7LRdqFIQ3kPs,1383
 upgini/utils/warning_counter.py,sha256=-GRY8EUggEBKODPSuXAkHn9KnEQwAORC0mmz_tim-PM,254
-upgini-1.2.111.dist-info/METADATA,sha256=FFU_9hZGBiuvJFZpxnlMJVkWlZ4ykBOnA0KcGqOBSyQ,49529
-upgini-1.2.111.dist-info/WHEEL,sha256=1yFddiXMmvYK7QYTqtRNtX66WJ0Mz8PYEiEUoOUUxRY,87
-upgini-1.2.111.dist-info/licenses/LICENSE,sha256=5RRzgvdJUu3BUDfv4bzVU6FqKgwHlIay63pPCSmSgzw,1514
-upgini-1.2.111.dist-info/RECORD,,
+upgini-1.2.113a1.dist-info/METADATA,sha256=An6LMif5xPvDV-osNxUvgAaz3xsvTnsQ4CnsNF5b6AQ,49531
+upgini-1.2.113a1.dist-info/WHEEL,sha256=1yFddiXMmvYK7QYTqtRNtX66WJ0Mz8PYEiEUoOUUxRY,87
+upgini-1.2.113a1.dist-info/licenses/LICENSE,sha256=5RRzgvdJUu3BUDfv4bzVU6FqKgwHlIay63pPCSmSgzw,1514
+upgini-1.2.113a1.dist-info/RECORD,,