Added conversational messages

b.py

@@ -0,0 +1,19 @@
+from deepeval.test_case import ConversationalTestCase, LLMTestCase, Message
+
+from deepeval.metrics import AnswerRelevancyMetric, BiasMetric
+from deepeval import evaluate
+
+
+tc1 = LLMTestCase(input="hi again", actual_output="bye")
+tc2 = LLMTestCase(input="hi", actual_output="bye")
+
+c_test_case = ConversationalTestCase(
+    messages=[Message(llm_test_case=tc2), Message(llm_test_case=tc1)]
+)
+
+
+evaluate(
+    test_cases=[c_test_case, tc1],
+    metrics=[AnswerRelevancyMetric(), BiasMetric()],
+    run_async=False,
+)

deepeval/benchmarks/big_bench_hard/big_bench_hard.py

@@ -10,7 +10,7 @@
 from deepeval.benchmarks.big_bench_hard.template import BigBenchHardTemplate
 from deepeval.benchmarks.utils import should_use_batch
 from deepeval.scorer import Scorer
-from deepeval.benchmarks.models import *
+from deepeval.benchmarks.schema import *
 
 
 class BigBenchHard(DeepEvalBaseBenchmark):

deepeval/dataset/utils.py

@@ -1,6 +1,6 @@
 from typing import List, Optional
 from deepeval.dataset.api import Golden, ConversationalGolden
-from deepeval.test_case import LLMTestCase, ConversationalTestCase
+from deepeval.test_case import LLMTestCase, ConversationalTestCase, Message
 
 
 def convert_test_cases_to_goldens(
@@ -42,6 +42,29 @@ def convert_goldens_to_test_cases(
     return test_cases
 
 
+def convert_goldens_to_messages(
+    goldens: List[Golden],
+    _alias: Optional[str] = None,
+    _id: Optional[str] = None,
+) -> List[Message]:
+    messages = []
+    for index, golden in enumerate(goldens):
+        test_case = LLMTestCase(
+            input=golden.input,
+            actual_output=golden.actual_output,
+            expected_output=golden.expected_output,
+            context=golden.context,
+            retrieval_context=golden.retrieval_context,
+            additional_metadata=golden.additional_metadata,
+            comments=golden.comments,
+            _dataset_alias=_alias,
+            _dataset_id=_id,
+            _dataset_rank=index,
+        )
+        messages.append(Message(test_case=test_case))
+    return messages
+
+
 def convert_convo_goldens_to_convo_test_cases(
     convo_goldens: List[ConversationalGolden],
     _alias: Optional[str] = None,