fix circular imports

src/intelligence_layer/evaluation/argilla.py

@@ -14,20 +14,34 @@
     RecordDataSequence,
 )
 from intelligence_layer.core import Input, InstructInput, Output, PromptOutput
-from intelligence_layer.evaluation.data_storage.run_repository import RunRepository
-from intelligence_layer.evaluation.data_storage.dataset_repository import DatasetRepository
-from intelligence_layer.evaluation.evaluator import Evaluator
-
+from intelligence_layer.evaluation.accumulator import MeanAccumulator
 from intelligence_layer.evaluation.base_logic import AggregationLogic, EvaluationLogic
+from intelligence_layer.evaluation.data_storage.aggregation_repository import (
+    AggregationRepository,
+)
+from intelligence_layer.evaluation.data_storage.dataset_repository import (
+    DatasetRepository,
+)
+from intelligence_layer.evaluation.data_storage.evaluation_repository import (
+    ArgillaEvaluationRepository,
+    EvaluationRepository,
+)
+from intelligence_layer.evaluation.data_storage.run_repository import RunRepository
 from intelligence_layer.evaluation.domain import (
     AggregatedEvaluation,
-    SuccessfulExampleOutput, ExpectedOutput, Example,
+    Example,
+    ExpectedOutput,
+    SuccessfulExampleOutput,
 )
 from intelligence_layer.evaluation.elo import (
     AutomatedEloComparison,
+    EloCalculator,
     EloComparison,
-    build_tournaments, PlayerScore,
+    PlayerScore,
+    WinRateCalculator,
+    build_tournaments,
 )
+from intelligence_layer.evaluation.evaluator import Evaluator
 
 
 class ArgillaEvaluationLogic(

src/intelligence_layer/evaluation/base_logic.py

@@ -2,9 +2,11 @@
 from typing import Generic, Iterable
 
 from intelligence_layer.core import Input, Output
-from intelligence_layer.evaluation import Evaluation, Example, ExpectedOutput
 from intelligence_layer.evaluation.domain import (
     AggregatedEvaluation,
+    Evaluation,
+    Example,
+    ExpectedOutput,
     SuccessfulExampleOutput,
 )
 

tests/use_cases/summarize/test_summarize.py

@@ -199,7 +199,6 @@ def test_long_context_summarize_evaluator(
     long_context_summarize_runner: Runner[str, str],
     in_memory_dataset_repository: InMemoryDatasetRepository,
     long_text: str,
-    no_op_tracer: NoOpTracer,
 ) -> None:
     input = LongContextSummarizeInput(text=long_text, language=Language("en"))
     bad_example = Example(