run ruff

Author: nabinchha

src/data_designer/config/config_builder.py

@@ -6,7 +6,8 @@
 import json
 import logging
 from pathlib import Path
-from typing import Union, Optional
+from typing import Optional, Union
+
 from pygments import highlight
 from pygments.formatters import HtmlFormatter
 from pygments.lexers import PythonLexer
@@ -27,12 +28,12 @@
 )
 from .seed import (
     DatastoreSeedDatasetReference,
+    IndexRange,
     LocalSeedDatasetReference,
+    PartitionBlock,
     SamplingStrategy,
     SeedConfig,
     SeedDatasetReference,
-    IndexRange,
-    PartitionBlock,
 )
 from .utils.constants import DEFAULT_REPR_HTML_STYLE, REPR_HTML_TEMPLATE
 from .utils.info import DataDesignerInfo
@@ -115,7 +116,11 @@ def from_config(cls, config: dict | str | Path | BuilderConfig) -> Self:
                     datastore_settings=builder_config.datastore_settings,
                 )
                 builder.set_seed_datastore_settings(builder_config.datastore_settings)
-            builder.with_seed_dataset(seed_dataset_reference, sampling_strategy=config.seed_config.sampling_strategy, selection_strategy=config.seed_config.selection_strategy)
+            builder.with_seed_dataset(
+                seed_dataset_reference,
+                sampling_strategy=config.seed_config.sampling_strategy,
+                selection_strategy=config.seed_config.selection_strategy,
+            )
 
         return builder
 
@@ -511,7 +516,11 @@ def with_seed_dataset(
         Returns:
             The current Data Designer config builder instance.
         """
-        self._seed_config = SeedConfig(dataset=dataset_reference.dataset, sampling_strategy=sampling_strategy, selection_strategy=selection_strategy)
+        self._seed_config = SeedConfig(
+            dataset=dataset_reference.dataset,
+            sampling_strategy=sampling_strategy,
+            selection_strategy=selection_strategy,
+        )
         self.set_seed_datastore_settings(
             dataset_reference.datastore_settings if hasattr(dataset_reference, "datastore_settings") else None
         )

src/data_designer/engine/column_generators/generators/seed_dataset.py

@@ -60,7 +60,7 @@ def _initialize(self) -> None:
         self._dataset_uri = self.resource_provider.datastore.get_dataset_uri(self.config.dataset)
         self._seed_dataset_size = self.duckdb_conn.execute(f"SELECT COUNT(*) FROM '{self._dataset_uri}'").fetchone()[0]
         self._index_range = self._resolve_index_range()
-        
+
     def _validate_selection_strategy(self) -> None:
         err_msg = None
         if self.config.selection_strategy is not None:
@@ -115,7 +115,9 @@ def _sample_records(self, num_records: int) -> pd.DataFrame:
         logger.info(f"  |-- seed dataset size: {self._seed_dataset_size} records")
         logger.info(f"  |-- sampling strategy: {self.config.sampling_strategy}")
         if self._index_range is not None:
-            logger.info(f"  |-- selection strategy: {type(self.config.selection_strategy).__name__}\n{self.config.selection_strategy.model_dump_json(indent=4)}")
+            logger.info(
+                f"  |-- selection strategy: {type(self.config.selection_strategy).__name__}\n{self.config.selection_strategy.model_dump_json(indent=4)}"
+            )
             logger.info(f"  |-- seed dataset size after selection: {self._index_range.size} records")
         df_batch = pd.DataFrame()
         df_sample = pd.DataFrame() if self._df_remaining is None else self._df_remaining