fix rule for black & pydocstyle

README.md

@@ -34,6 +34,7 @@ export OPENAI_API_KEY="your-key"
 
 # Run evaluation
 lightspeed-eval --system-config config/system.yaml --eval-data config/evaluation_data.yaml
+```
 
 ## 📊 Supported Metrics
 

archive/lightspeed_core_evaluation/taxonomy_eval.py

@@ -187,8 +187,8 @@ def _get_score(
                 retrieved_contexts=[data.context],
             )
             return scorer.single_turn_score(
-                data  # pyright: ignore [reportArgumentType]
-            )
+                data
+            )  # pyright: ignore [reportArgumentType]
 
         df = self._taxonomy_df.copy()
         if self._args.eval_type in ("all", "context"):

lsc_agent_eval/src/lsc_agent_eval/core/agent_goal_eval/agent_goal_eval.py

@@ -75,8 +75,7 @@ def run_evaluation(self) -> None:
             conversations = self.data_manager.get_conversations()
 
             logger.info(
-                "Starting Agent Goal Evaluation\n"
-                "Total: %d evaluations across %d conversations",
+                "Starting Agent Goal Evaluation\nTotal: %d evaluations across %d conversations",
                 self.data_manager.get_eval_count(),
                 len(conversations),
             )

lsc_agent_eval/src/lsc_agent_eval/core/utils/judge.py

@@ -38,8 +38,7 @@ def _setup_litellm(self) -> None:
             api_key = os.environ.get("OPENAI_API_KEY")
             if not api_key:
                 raise JudgeModelError(
-                    "OPENAI_API_KEY environment variable is required "
-                    "for OpenAI provider"
+                    "OPENAI_API_KEY environment variable is required for OpenAI provider"
                 )
             self.model_name = self.judge_model
 

pyproject.toml

@@ -48,6 +48,12 @@ generate_answers = "generate_answers.generate_answers:main"
 # Note: torch[cpu] variant configuration removed for uv compatibility
 # Modern PyTorch versions are available on PyPI directly
 
+[tool.black]
+line-length = 88
+
+[tool.pydocstyle]
+convention = "google"
+
 [tool.mypy]
 disable_error_code = ["union-attr", "return-value", "arg-type", "import-untyped"]
 ignore_missing_imports = true

src/lightspeed_evaluation/__init__.py

@@ -1,5 +1,4 @@
-"""
-LightSpeed Evaluation Framework.
+"""LightSpeed Evaluation Framework.
 
 Main components:
 - EvaluationDriver: Runs complete evaluation

src/lightspeed_evaluation/core/llm/deepeval.py

@@ -6,8 +6,7 @@
 
 
 class DeepEvalLLMManager:
-    """
-    DeepEval LLM Manager - Takes LLM parameters directly.
+    """DeepEval LLM Manager - Takes LLM parameters directly.
 
     This manager focuses solely on DeepEval-specific LLM integration.
     """

src/lightspeed_evaluation/core/llm/manager.py

@@ -11,8 +11,7 @@ class LLMError(Exception):
 
 
 class LLMManager:
-    """
-    Generic LLM Manager for all use cases (Ragas, DeepEval, Custom metrics).
+    """Generic LLM Manager for all use cases (Ragas, DeepEval, Custom metrics).
 
     Responsibilities:
     - Environment validation for multiple providers

src/lightspeed_evaluation/core/llm/ragas.py

@@ -82,8 +82,7 @@ def is_finished(self, response: LLMResult) -> bool:
 
 
 class RagasLLMManager:
-    """
-    Ragas LLM Manager - Takes LLM parameters directly.
+    """Ragas LLM Manager - Takes LLM parameters directly.
 
     This manager focuses solely on Ragas-specific LLM integration.
     """

src/lightspeed_evaluation/core/metrics/custom.py

@@ -30,8 +30,7 @@ class CustomMetrics:
     """Handles custom metrics using LLMManager for direct LiteLLM calls."""
 
     def __init__(self, llm_manager: LLMManager):
-        """
-        Initialize with LLM Manager.
+        """Initialize with LLM Manager.
 
         Args:
             llm_manager: Pre-configured LLMManager with validated parameters
@@ -89,8 +88,7 @@ def _call_llm(self, prompt: str, system_prompt: Optional[str] = None) -> str:
             raise RuntimeError(f"LiteLLM call failed: {str(e)}") from e
 
     def _parse_score_response(self, response: str) -> Tuple[Optional[float], str]:
-        r"""
-        Parse LLM response to extract score and reason.
+        r"""Parse LLM response to extract score and reason.
 
         Expected formats:
         - "Score: 0.85\nReason: The answer is accurate..."

src/lightspeed_evaluation/core/metrics/deepeval.py

@@ -20,8 +20,7 @@ class DeepEvalMetrics:
     """Handles DeepEval metrics evaluation using LLM Manager."""
 
     def __init__(self, llm_manager: LLMManager):
-        """
-        Initialize with LLM Manager.
+        """Initialize with LLM Manager.
 
         Args:
             llm_manager: Pre-configured LLMManager with validated parameters

src/lightspeed_evaluation/core/metrics/ragas.py

@@ -24,8 +24,7 @@ class RagasMetrics:
     """Handles Ragas metrics evaluation using LLM Manager."""
 
     def __init__(self, llm_manager: LLMManager):
-        """
-        Initialize with LLM Manager.
+        """Initialize with LLM Manager.
 
         Args:
             llm_manager: Pre-configured LLMManager with validated parameters

src/lightspeed_evaluation/drivers/evaluation.py

@@ -1,7 +1,6 @@
-"""
-Evaluation Driver - Main evaluation controller.
+"""Evaluation Driver - Main evaluation controller.
 
-Controls the evaluation flow through conversations & turns
+Controls the evaluation flow through conversations & turns.
 """
 
 import time
@@ -116,14 +115,13 @@ def get_supported_frameworks(self) -> List[str]:
 
 
 class EvaluationDriver:
-    """
-    Main evaluation driver - orchestrates the evaluation process.
+    """Main evaluation driver - orchestrates the evaluation process.
 
     Responsibilities:
     - Data validation
     - Metric routing and evaluation
     - Result collection
-    - Status determination (PASS/FAIL/ERROR)
+    - Status determination (PASS/FAIL/ERROR).
     """
 
     def __init__(self, config_loader: ConfigLoader):
@@ -146,14 +144,13 @@ def validate_data(self, evaluation_data: List[EvaluationData]) -> bool:
     def run_evaluation(
         self, evaluation_data: List[EvaluationData]
     ) -> List[EvaluationResult]:
-        """
-        Run complete evaluation pipeline.
+        """Run complete evaluation pipeline.
 
         Args:
             evaluation_data: List of conversation data to evaluate
 
         Returns:
-            List of evaluation results
+            List of evaluation results.
         """
         print("🚀 Starting evaluation...")
         self.results = []
@@ -214,11 +211,10 @@ def _evaluate_conversation(self, conv_data: EvaluationData) -> None:
     def _evaluate_metric(
         self, request: EvaluationRequest
     ) -> Optional[EvaluationResult]:
-        """
-        Evaluate single metric using context.
+        """Evaluate single metric using context.
 
         Returns:
-            EvaluationResult or None if evaluation fails
+            EvaluationResult or None if evaluation fails.
         """
         start_time = time.time()
 

src/lightspeed_evaluation/runner/evaluation.py

@@ -15,8 +15,7 @@
 def run_evaluation(
     system_config_path: str, evaluation_data_path: str, output_dir: Optional[str] = None
 ) -> Optional[Dict[str, int]]:
-    """
-    Run the complete evaluation pipeline using EvaluationDriver.
+    """Run the complete evaluation pipeline using EvaluationDriver.
 
     Args:
         system_config_path: Path to system.yaml