deepeval 3.4.8__py3-none-any.whl → 3.5.0__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (47) hide show
  1. deepeval/__init__.py +8 -5
  2. deepeval/_version.py +1 -1
  3. deepeval/benchmarks/drop/drop.py +2 -3
  4. deepeval/benchmarks/hellaswag/hellaswag.py +2 -2
  5. deepeval/benchmarks/logi_qa/logi_qa.py +2 -2
  6. deepeval/benchmarks/math_qa/math_qa.py +2 -2
  7. deepeval/benchmarks/mmlu/mmlu.py +2 -2
  8. deepeval/benchmarks/truthful_qa/truthful_qa.py +2 -2
  9. deepeval/cli/main.py +561 -727
  10. deepeval/confident/api.py +30 -14
  11. deepeval/config/__init__.py +0 -0
  12. deepeval/config/settings.py +565 -0
  13. deepeval/config/settings_manager.py +133 -0
  14. deepeval/config/utils.py +86 -0
  15. deepeval/dataset/__init__.py +1 -0
  16. deepeval/dataset/dataset.py +70 -10
  17. deepeval/dataset/test_run_tracer.py +82 -0
  18. deepeval/dataset/utils.py +23 -0
  19. deepeval/integrations/pydantic_ai/__init__.py +2 -4
  20. deepeval/integrations/pydantic_ai/{setup.py → otel.py} +0 -8
  21. deepeval/integrations/pydantic_ai/patcher.py +376 -0
  22. deepeval/key_handler.py +1 -0
  23. deepeval/metrics/answer_relevancy/template.py +7 -2
  24. deepeval/metrics/faithfulness/template.py +11 -8
  25. deepeval/metrics/multimodal_metrics/multimodal_answer_relevancy/template.py +6 -4
  26. deepeval/metrics/multimodal_metrics/multimodal_faithfulness/template.py +6 -4
  27. deepeval/metrics/tool_correctness/tool_correctness.py +7 -3
  28. deepeval/models/llms/amazon_bedrock_model.py +24 -3
  29. deepeval/models/llms/grok_model.py +1 -1
  30. deepeval/models/llms/kimi_model.py +1 -1
  31. deepeval/models/llms/openai_model.py +37 -41
  32. deepeval/models/retry_policy.py +280 -0
  33. deepeval/openai_agents/agent.py +4 -2
  34. deepeval/test_run/api.py +1 -0
  35. deepeval/tracing/otel/exporter.py +20 -8
  36. deepeval/tracing/otel/utils.py +57 -0
  37. deepeval/tracing/perf_epoch_bridge.py +4 -4
  38. deepeval/tracing/tracing.py +37 -16
  39. deepeval/tracing/utils.py +98 -1
  40. deepeval/utils.py +111 -70
  41. {deepeval-3.4.8.dist-info → deepeval-3.5.0.dist-info}/METADATA +16 -13
  42. {deepeval-3.4.8.dist-info → deepeval-3.5.0.dist-info}/RECORD +45 -40
  43. deepeval/env.py +0 -35
  44. deepeval/integrations/pydantic_ai/agent.py +0 -364
  45. {deepeval-3.4.8.dist-info → deepeval-3.5.0.dist-info}/LICENSE.md +0 -0
  46. {deepeval-3.4.8.dist-info → deepeval-3.5.0.dist-info}/WHEEL +0 -0
  47. {deepeval-3.4.8.dist-info → deepeval-3.5.0.dist-info}/entry_points.txt +0 -0
deepeval/__init__.py CHANGED
@@ -3,9 +3,9 @@ import warnings
3
3
  import re
4
4
 
5
5
  # load environment variables before other imports
6
- from .env import autoload_dotenv as _autoload_dotenv
6
+ from deepeval.config.settings import autoload_dotenv, get_settings
7
7
 
8
- _autoload_dotenv()
8
+ autoload_dotenv()
9
9
 
10
10
  from ._version import __version__
11
11
  from deepeval.evaluate import evaluate, assert_test
@@ -14,9 +14,12 @@ from deepeval.test_run import on_test_run_end, log_hyperparameters
14
14
  from deepeval.utils import login
15
15
  from deepeval.telemetry import *
16
16
 
17
- if os.getenv("DEEPEVAL_GRPC_LOGGING") != "1":
18
- os.environ["GRPC_VERBOSITY"] = "ERROR"
19
- os.environ["GRPC_TRACE"] = ""
17
+
18
+ settings = get_settings()
19
+ if not settings.DEEPEVAL_GRPC_LOGGING:
20
+ os.environ.setdefault("GRPC_VERBOSITY", "ERROR")
21
+ os.environ.setdefault("GRPC_TRACE", "")
22
+
20
23
 
21
24
  __all__ = [
22
25
  "login",
deepeval/_version.py CHANGED
@@ -1 +1 @@
1
- __version__: str = "3.4.8"
1
+ __version__: str = "3.5.0"
deepeval/benchmarks/drop/drop.py CHANGED
@@ -1,6 +1,5 @@
1
- from typing import List, Optional, Dict
1
+ from typing import List, Optional, Dict, Union
2
2
  from tqdm import tqdm
3
- from typing import Union
4
3
 
5
4
  from deepeval.dataset import Golden
6
5
  from deepeval.benchmarks.base_benchmark import (
@@ -50,7 +49,7 @@ class DROP(DeepEvalBaseBenchmark):
50
49
  self,
51
50
  model: DeepEvalBaseLLM,
52
51
  *args,
53
- batch_size: int | None = None,
52
+ batch_size: Union[int, None] = None,
54
53
  **kwargs,
55
54
  ) -> DeepEvalBaseBenchmarkResult:
56
55
  import pandas as pd
deepeval/benchmarks/hellaswag/hellaswag.py CHANGED
@@ -1,4 +1,4 @@
1
- from typing import List, Dict, Optional
1
+ from typing import List, Dict, Optional, Union
2
2
  from tqdm import tqdm
3
3
 
4
4
  from deepeval.dataset import Golden
@@ -51,7 +51,7 @@ class HellaSwag(DeepEvalBaseBenchmark):
51
51
  self,
52
52
  model: DeepEvalBaseLLM,
53
53
  *args,
54
- batch_size: int | None = None,
54
+ batch_size: Union[int, None] = None,
55
55
  **kwargs,
56
56
  ) -> DeepEvalBaseBenchmarkResult:
57
57
  import pandas as pd
deepeval/benchmarks/logi_qa/logi_qa.py CHANGED
@@ -1,4 +1,4 @@
1
- from typing import List, Optional, Dict
1
+ from typing import List, Optional, Dict, Union
2
2
  from tqdm import tqdm
3
3
  import requests
4
4
  import json
@@ -52,7 +52,7 @@ class LogiQA(DeepEvalBaseBenchmark):
52
52
  self,
53
53
  model: DeepEvalBaseLLM,
54
54
  *args,
55
- batch_size: int | None = None,
55
+ batch_size: Union[int, None] = None,
56
56
  **kwargs,
57
57
  ) -> DeepEvalBaseBenchmarkResult:
58
58
  import pandas as pd
deepeval/benchmarks/math_qa/math_qa.py CHANGED
@@ -1,4 +1,4 @@
1
- from typing import List, Optional, Dict
1
+ from typing import List, Optional, Dict, Union
2
2
  from tqdm import tqdm
3
3
 
4
4
  from deepeval.dataset import Golden
@@ -50,7 +50,7 @@ class MathQA(DeepEvalBaseBenchmark):
50
50
  self,
51
51
  model: DeepEvalBaseLLM,
52
52
  *args,
53
- batch_size: int | None = None,
53
+ batch_size: Union[int, None] = None,
54
54
  **kwargs,
55
55
  ) -> DeepEvalBaseBenchmarkResult:
56
56
  import pandas as pd
deepeval/benchmarks/mmlu/mmlu.py CHANGED
@@ -1,4 +1,4 @@
1
- from typing import List, Optional, Dict
1
+ from typing import List, Optional, Dict, Union
2
2
  from tqdm import tqdm
3
3
 
4
4
  from deepeval.dataset import Golden
@@ -49,7 +49,7 @@ class MMLU(DeepEvalBaseBenchmark):
49
49
  self,
50
50
  model: DeepEvalBaseLLM,
51
51
  *args,
52
- batch_size: int | None = None,
52
+ batch_size: Union[int, None] = None,
53
53
  **kwargs,
54
54
  ) -> DeepEvalBaseBenchmarkResult:
55
55
  import pandas as pd
deepeval/benchmarks/truthful_qa/truthful_qa.py CHANGED
@@ -1,4 +1,4 @@
1
- from typing import List, Dict, Optional
1
+ from typing import List, Dict, Optional, Union
2
2
  from tqdm import tqdm
3
3
 
4
4
  from deepeval.dataset import Golden
@@ -59,7 +59,7 @@ class TruthfulQA(DeepEvalBaseBenchmark):
59
59
  self,
60
60
  model: DeepEvalBaseLLM,
61
61
  *args,
62
- batch_size: int | None = None,
62
+ batch_size: Union[int, None] = None,
63
63
  **kwargs,
64
64
  ) -> DeepEvalBaseBenchmarkResult:
65
65
  import pandas as pd