feat: add experimental folder for beta features

src/crewai/crew.py

@@ -1337,7 +1337,7 @@ class Crew(FlowTrackable, BaseModel):
             evaluator = CrewEvaluator(test_crew, llm_instance)
 
             if include_agent_eval:
-                from crewai.evaluation import create_default_evaluator
+                from crewai.experimental.evaluation import create_default_evaluator
                 agent_evaluator = create_default_evaluator(crew=test_crew)
 
             for i in range(1, n_iterations + 1):

src/crewai/evaluation/experiment/__init__.py

@@ -1,8 +0,0 @@
-from crewai.evaluation.experiment.runner import ExperimentRunner
-from crewai.evaluation.experiment.result import ExperimentResults, ExperimentResult
-
-__all__ = [
-    "ExperimentRunner",
-    "ExperimentResults",
-    "ExperimentResult"
-]

src/crewai/experimental/__init__.py

@@ -0,0 +1,40 @@
+from crewai.experimental.evaluation import (
+    BaseEvaluator,
+    EvaluationScore,
+    MetricCategory,
+    AgentEvaluationResult,
+    SemanticQualityEvaluator,
+    GoalAlignmentEvaluator,
+    ReasoningEfficiencyEvaluator,
+    ToolSelectionEvaluator,
+    ParameterExtractionEvaluator,
+    ToolInvocationEvaluator,
+    EvaluationTraceCallback,
+    create_evaluation_callbacks,
+    AgentEvaluator,
+    create_default_evaluator,
+    ExperimentRunner,
+    ExperimentResults,
+    ExperimentResult,
+)
+
+
+__all__ = [
+    "BaseEvaluator",
+    "EvaluationScore",
+    "MetricCategory",
+    "AgentEvaluationResult",
+    "SemanticQualityEvaluator",
+    "GoalAlignmentEvaluator",
+    "ReasoningEfficiencyEvaluator",
+    "ToolSelectionEvaluator",
+    "ParameterExtractionEvaluator",
+    "ToolInvocationEvaluator",
+    "EvaluationTraceCallback",
+    "create_evaluation_callbacks",
+    "AgentEvaluator",
+    "create_default_evaluator",
+    "ExperimentRunner",
+    "ExperimentResults",
+    "ExperimentResult"
+]

src/crewai/experimental/evaluation/__init__.py

@@ -1,41 +1,30 @@
-from crewai.evaluation.base_evaluator import (
+from crewai.experimental.evaluation.base_evaluator import (
     BaseEvaluator,
     EvaluationScore,
     MetricCategory,
     AgentEvaluationResult
 )
 
-from crewai.evaluation.metrics.semantic_quality_metrics import (
+from crewai.experimental.evaluation.metrics import (
-    SemanticQualityEvaluator
+    SemanticQualityEvaluator,
-)
+    GoalAlignmentEvaluator,
-
+    ReasoningEfficiencyEvaluator,
-from crewai.evaluation.metrics.goal_metrics import (
-    GoalAlignmentEvaluator
-)
-
-from crewai.evaluation.metrics.reasoning_metrics import (
-    ReasoningEfficiencyEvaluator
-)
-
-
-from crewai.evaluation.metrics.tools_metrics import (
     ToolSelectionEvaluator,
     ParameterExtractionEvaluator,
     ToolInvocationEvaluator
 )
 
-from crewai.evaluation.evaluation_listener import (
+from crewai.experimental.evaluation.evaluation_listener import (
     EvaluationTraceCallback,
     create_evaluation_callbacks
 )
 
-
+from crewai.experimental.evaluation.agent_evaluator import (
-from crewai.evaluation.agent_evaluator import (
     AgentEvaluator,
     create_default_evaluator
 )
 
-from crewai.evaluation.experiment import (
+from crewai.experimental.evaluation.experiment import (
     ExperimentRunner,
     ExperimentResults,
     ExperimentResult
@@ -59,4 +48,4 @@ __all__ = [
     "ExperimentRunner",
     "ExperimentResults",
     "ExperimentResult"
-]
+]

src/crewai/experimental/evaluation/agent_evaluator.py

@@ -1,16 +1,16 @@
-from crewai.evaluation.base_evaluator import AgentEvaluationResult, AggregationStrategy
+from crewai.experimental.evaluation.base_evaluator import AgentEvaluationResult, AggregationStrategy
 from crewai.agent import Agent
 from crewai.task import Task
-from crewai.evaluation.evaluation_display import EvaluationDisplayFormatter
+from crewai.experimental.evaluation.evaluation_display import EvaluationDisplayFormatter
 
 from typing import Any, Dict
 from collections import defaultdict
-from crewai.evaluation import BaseEvaluator, create_evaluation_callbacks
+from crewai.experimental.evaluation import BaseEvaluator, create_evaluation_callbacks
 from collections.abc import Sequence
 from crewai.crew import Crew
 from crewai.utilities.events.crewai_event_bus import crewai_event_bus
 from crewai.utilities.events.utils.console_formatter import ConsoleFormatter
-from crewai.evaluation.evaluation_display import AgentAggregatedEvaluationResult
+from crewai.experimental.evaluation.evaluation_display import AgentAggregatedEvaluationResult
 
 class AgentEvaluator:
     def __init__(
@@ -161,7 +161,7 @@ class AgentEvaluator:
         return result
 
 def create_default_evaluator(crew, llm=None):
-    from crewai.evaluation import (
+    from crewai.experimental.evaluation import (
         GoalAlignmentEvaluator,
         SemanticQualityEvaluator,
         ToolSelectionEvaluator,

src/crewai/experimental/evaluation/evaluation_display.py

@@ -3,8 +3,8 @@ from typing import Dict, Any, List
 from rich.table import Table
 from rich.box import HEAVY_EDGE, ROUNDED
 from collections.abc import Sequence
-from crewai.evaluation.base_evaluator import AgentAggregatedEvaluationResult, AggregationStrategy, AgentEvaluationResult, MetricCategory
+from crewai.experimental.evaluation.base_evaluator import AgentAggregatedEvaluationResult, AggregationStrategy, AgentEvaluationResult, MetricCategory
-from crewai.evaluation import EvaluationScore
+from crewai.experimental.evaluation import EvaluationScore
 from crewai.utilities.events.utils.console_formatter import ConsoleFormatter
 from crewai.utilities.llm_utils import create_llm
 

src/crewai/experimental/evaluation/experiment/__init__.py

@@ -0,0 +1,8 @@
+from crewai.experimental.evaluation.experiment.runner import ExperimentRunner
+from crewai.experimental.evaluation.experiment.result import ExperimentResults, ExperimentResult
+
+__all__ = [
+    "ExperimentRunner",
+    "ExperimentResults",
+    "ExperimentResult"
+]

src/crewai/experimental/evaluation/experiment/result.py

@@ -18,7 +18,7 @@ class ExperimentResults:
         self.metadata = metadata or {}
         self.timestamp = datetime.now(timezone.utc)
 
-        from crewai.evaluation.experiment.result_display import ExperimentResultsDisplay
+        from crewai.experimental.evaluation.experiment.result_display import ExperimentResultsDisplay
         self.display = ExperimentResultsDisplay()
 
     def to_json(self, filepath: str | None = None) -> dict[str, Any]:

src/crewai/experimental/evaluation/experiment/result_display.py

@@ -2,7 +2,7 @@ from typing import Dict, Any
 from rich.console import Console
 from rich.table import Table
 from rich.panel import Panel
-from crewai.evaluation.experiment.result import ExperimentResults
+from crewai.experimental.evaluation.experiment.result import ExperimentResults
 
 class ExperimentResultsDisplay:
     def __init__(self):

src/crewai/experimental/evaluation/experiment/runner.py

@@ -3,10 +3,10 @@ from hashlib import md5
 from typing import Any
 
 from crewai import Crew
-from crewai.evaluation import AgentEvaluator, create_default_evaluator
+from crewai.experimental.evaluation import AgentEvaluator, create_default_evaluator
-from crewai.evaluation.experiment.result_display import ExperimentResultsDisplay
+from crewai.experimental.evaluation.experiment.result_display import ExperimentResultsDisplay
-from crewai.evaluation.experiment.result import ExperimentResults, ExperimentResult
+from crewai.experimental.evaluation.experiment.result import ExperimentResults, ExperimentResult
-from crewai.evaluation.evaluation_display import AgentAggregatedEvaluationResult
+from crewai.experimental.evaluation.evaluation_display import AgentAggregatedEvaluationResult
 
 class ExperimentRunner:
     def __init__(self, dataset: list[dict[str, Any]]):

src/crewai/experimental/evaluation/metrics/__init__.py

@@ -0,0 +1,26 @@
+from crewai.experimental.evaluation.metrics.reasoning_metrics import (
+    ReasoningEfficiencyEvaluator
+)
+
+from crewai.experimental.evaluation.metrics.tools_metrics import (
+    ToolSelectionEvaluator,
+    ParameterExtractionEvaluator,
+    ToolInvocationEvaluator
+)
+
+from crewai.experimental.evaluation.metrics.goal_metrics import (
+    GoalAlignmentEvaluator
+)
+
+from crewai.experimental.evaluation.metrics.semantic_quality_metrics import (
+    SemanticQualityEvaluator
+)
+
+__all__ = [
+    "ReasoningEfficiencyEvaluator",
+    "ToolSelectionEvaluator",
+    "ParameterExtractionEvaluator",
+    "ToolInvocationEvaluator",
+    "GoalAlignmentEvaluator",
+    "SemanticQualityEvaluator"
+]

src/crewai/experimental/evaluation/metrics/goal_metrics.py

@@ -3,8 +3,8 @@ from typing import Any, Dict
 from crewai.agent import Agent
 from crewai.task import Task
 
-from crewai.evaluation.base_evaluator import BaseEvaluator, EvaluationScore, MetricCategory
+from crewai.experimental.evaluation.base_evaluator import BaseEvaluator, EvaluationScore, MetricCategory
-from crewai.evaluation.json_parser import extract_json_from_llm_response
+from crewai.experimental.evaluation.json_parser import extract_json_from_llm_response
 
 class GoalAlignmentEvaluator(BaseEvaluator):
     @property

src/crewai/experimental/evaluation/metrics/reasoning_metrics.py

@@ -16,8 +16,8 @@ from collections.abc import Sequence
 from crewai.agent import Agent
 from crewai.task import Task
 
-from crewai.evaluation.base_evaluator import BaseEvaluator, EvaluationScore, MetricCategory
+from crewai.experimental.evaluation.base_evaluator import BaseEvaluator, EvaluationScore, MetricCategory
-from crewai.evaluation.json_parser import extract_json_from_llm_response
+from crewai.experimental.evaluation.json_parser import extract_json_from_llm_response
 from crewai.tasks.task_output import TaskOutput
 
 class ReasoningPatternType(Enum):

src/crewai/experimental/evaluation/metrics/semantic_quality_metrics.py

@@ -3,8 +3,8 @@ from typing import Any, Dict
 from crewai.agent import Agent
 from crewai.task import Task
 
-from crewai.evaluation.base_evaluator import BaseEvaluator, EvaluationScore, MetricCategory
+from crewai.experimental.evaluation.base_evaluator import BaseEvaluator, EvaluationScore, MetricCategory
-from crewai.evaluation.json_parser import extract_json_from_llm_response
+from crewai.experimental.evaluation.json_parser import extract_json_from_llm_response
 
 class SemanticQualityEvaluator(BaseEvaluator):
     @property

src/crewai/experimental/evaluation/metrics/tools_metrics.py

@@ -1,8 +1,8 @@
 import json
 from typing import Dict, Any
 
-from crewai.evaluation.base_evaluator import BaseEvaluator, EvaluationScore, MetricCategory
+from crewai.experimental.evaluation.base_evaluator import BaseEvaluator, EvaluationScore, MetricCategory
-from crewai.evaluation.json_parser import extract_json_from_llm_response
+from crewai.experimental.evaluation.json_parser import extract_json_from_llm_response
 from crewai.agent import Agent
 from crewai.task import Task
 

tests/evaluation/metrics/test_goal_metrics.py

@@ -1,8 +1,8 @@
 from unittest.mock import patch, MagicMock
 from tests.evaluation.metrics.base_evaluation_metrics_test import BaseEvaluationMetricsTest
 
-from crewai.evaluation.base_evaluator import EvaluationScore
+from crewai.experimental.evaluation.base_evaluator import EvaluationScore
-from crewai.evaluation.metrics.goal_metrics import GoalAlignmentEvaluator
+from crewai.experimental.evaluation.metrics.goal_metrics import GoalAlignmentEvaluator
 from crewai.utilities.llm_utils import LLM
 
 

tests/evaluation/metrics/test_reasoning_metrics.py

@@ -3,12 +3,12 @@ from unittest.mock import patch, MagicMock
 from typing import List, Dict, Any
 
 from crewai.tasks.task_output import TaskOutput
-from crewai.evaluation.metrics.reasoning_metrics import (
+from crewai.experimental.evaluation.metrics.reasoning_metrics import (
     ReasoningEfficiencyEvaluator,
 )
 from tests.evaluation.metrics.base_evaluation_metrics_test import BaseEvaluationMetricsTest
 from crewai.utilities.llm_utils import LLM
-from crewai.evaluation.base_evaluator import EvaluationScore
+from crewai.experimental.evaluation.base_evaluator import EvaluationScore
 
 class TestReasoningEfficiencyEvaluator(BaseEvaluationMetricsTest):
     @pytest.fixture

tests/evaluation/metrics/test_semantic_quality_metrics.py

@@ -1,7 +1,7 @@
 from unittest.mock import patch, MagicMock
 
-from crewai.evaluation.base_evaluator import EvaluationScore
+from crewai.experimental.evaluation.base_evaluator import EvaluationScore
-from crewai.evaluation.metrics.semantic_quality_metrics import SemanticQualityEvaluator
+from crewai.experimental.evaluation.metrics.semantic_quality_metrics import SemanticQualityEvaluator
 from tests.evaluation.metrics.base_evaluation_metrics_test import BaseEvaluationMetricsTest
 from crewai.utilities.llm_utils import LLM
 

tests/evaluation/metrics/test_tools_metrics.py

@@ -1,6 +1,6 @@
 from unittest.mock import patch, MagicMock
 
-from crewai.evaluation.metrics.tools_metrics import (
+from crewai.experimental.evaluation.metrics.tools_metrics import (
     ToolSelectionEvaluator,
     ParameterExtractionEvaluator,
     ToolInvocationEvaluator

tests/evaluation/test_agent_evaluator.py

@@ -3,9 +3,9 @@ import pytest
 from crewai.agent import Agent
 from crewai.task import Task
 from crewai.crew import Crew
-from crewai.evaluation.agent_evaluator import AgentEvaluator
+from crewai.experimental.evaluation.agent_evaluator import AgentEvaluator
-from crewai.evaluation.base_evaluator import AgentEvaluationResult
+from crewai.experimental.evaluation.base_evaluator import AgentEvaluationResult
-from crewai.evaluation import (
+from crewai.experimental.evaluation import (
     GoalAlignmentEvaluator,
     SemanticQualityEvaluator,
     ToolSelectionEvaluator,
@@ -14,7 +14,7 @@ from crewai.evaluation import (
     ReasoningEfficiencyEvaluator
 )
 
-from crewai.evaluation import create_default_evaluator
+from crewai.experimental.evaluation import create_default_evaluator
 class TestAgentEvaluator:
     @pytest.fixture
     def mock_crew(self):

tests/evaluation/test_experiment_result.py

@@ -1,7 +1,7 @@
 import pytest
 from unittest.mock import MagicMock, patch
 
-from crewai.evaluation.experiment.result import ExperimentResult, ExperimentResults
+from crewai.experimental.evaluation.experiment.result import ExperimentResult, ExperimentResults
 
 
 class TestExperimentResult:

tests/evaluation/test_experiment_runner.py

@@ -2,10 +2,10 @@ import pytest
 from unittest.mock import MagicMock, patch
 
 from crewai.crew import Crew
-from crewai.evaluation.experiment.runner import ExperimentRunner
+from crewai.experimental.evaluation.experiment.runner import ExperimentRunner
-from crewai.evaluation.experiment.result import ExperimentResults
+from crewai.experimental.evaluation.experiment.result import ExperimentResults
-from crewai.evaluation.evaluation_display import AgentAggregatedEvaluationResult
+from crewai.experimental.evaluation.evaluation_display import AgentAggregatedEvaluationResult
-from crewai.evaluation.base_evaluator import MetricCategory, EvaluationScore
+from crewai.experimental.evaluation.base_evaluator import MetricCategory, EvaluationScore
 
 
 class TestExperimentRunner:
@@ -44,7 +44,7 @@ class TestExperimentRunner:
 
         return {"Test Agent": agent_evaluation}
 
-    @patch('crewai.evaluation.experiment.runner.create_default_evaluator')
+    @patch('crewai.experimental.evaluation.experiment.runner.create_default_evaluator')
     def test_run_success(self, mock_create_evaluator, mock_crew, mock_evaluator_results):
         dataset = [
             {
@@ -102,7 +102,7 @@ class TestExperimentRunner:
         assert mock_evaluator.get_agent_evaluation.call_count == 3
 
 
-    @patch('crewai.evaluation.experiment.runner.create_default_evaluator')
+    @patch('crewai.experimental.evaluation.experiment.runner.create_default_evaluator')
     def test_run_success_with_unknown_metric(self, mock_create_evaluator, mock_crew, mock_evaluator_results):
         dataset = [
             {
@@ -130,7 +130,7 @@ class TestExperimentRunner:
         assert "unknown_metric" in result.expected_score.keys()
         assert result.passed is True
 
-    @patch('crewai.evaluation.experiment.runner.create_default_evaluator')
+    @patch('crewai.experimental.evaluation.experiment.runner.create_default_evaluator')
     def test_run_success_with_single_metric_evaluator_and_expected_specific_metric(self, mock_create_evaluator, mock_crew, mock_evaluator_results):
         dataset = [
             {
@@ -163,7 +163,7 @@ class TestExperimentRunner:
         assert "goal_alignment" in result.expected_score.keys()
         assert result.passed is True
 
-    @patch('crewai.evaluation.experiment.runner.create_default_evaluator')
+    @patch('crewai.experimental.evaluation.experiment.runner.create_default_evaluator')
     def test_run_success_when_expected_metric_is_not_available(self, mock_create_evaluator, mock_crew, mock_evaluator_results):
         dataset = [
             {