judgeval 0.0.14__py3-none-any.whl → 0.0.15__py3-none-any.whl

judgeval/run_evaluation.py

@@ -20,6 +20,7 @@ from judgeval.constants import (
     ROOT_API,
     JUDGMENT_EVAL_API_URL,
     JUDGMENT_EVAL_LOG_API_URL,
+    MAX_CONCURRENT_EVALUATIONS
 )
 from judgeval.common.exceptions import JudgmentAPIError
 from judgeval.evaluation_run import EvaluationRun
@@ -30,6 +31,7 @@ from judgeval.common.logger import (
     error, 
     example_logging_context
 )
+from judgeval.rules import RulesEngine, Rule, AlertResult, AlertStatus
 
 
 def execute_api_eval(evaluation_run: EvaluationRun) -> List[Dict]:
@@ -53,7 +55,8 @@ def execute_api_eval(evaluation_run: EvaluationRun) -> List[Dict]:
             "Authorization": f"Bearer {evaluation_run.judgment_api_key}",
             "X-Organization-Id": evaluation_run.organization_id
         }, 
-        json=payload)
+        json=payload,
+        verify=False)
         response_data = response.json()
     except Exception as e:
         error(f"Error: {e}")
@@ -166,7 +169,8 @@ def check_eval_run_name_exists(eval_name: str, project_name: str, judgment_api_k
                 "eval_name": eval_name,
                 "project_name": project_name,
                 "judgment_api_key": judgment_api_key,
-            }
+            },
+            verify=False
         )
         
         if response.status_code == 409:
@@ -208,7 +212,8 @@ def log_evaluation_results(merged_results: List[ScoringResult], evaluation_run:
                 "results": [result.to_dict() for result in merged_results],
                 "project_name": evaluation_run.project_name,
                 "eval_name": evaluation_run.eval_name,
-            }
+            },
+            verify=False
         )
         
         if not res.ok:
@@ -228,6 +233,7 @@ def log_evaluation_results(merged_results: List[ScoringResult], evaluation_run:
         raise ValueError(f"Failed to save evaluation results to DB: {str(e)}")
 
 
+
 def run_eval(evaluation_run: EvaluationRun, override: bool = False) -> List[ScoringResult]:
     """
     Executes an evaluation of `Example`s using one or more `Scorer`s
@@ -245,7 +251,7 @@ def run_eval(evaluation_run: EvaluationRun, override: bool = False) -> List[Scor
             metadata (Optional[Dict[str, Any]]): Additional metadata to include for this evaluation run, e.g. comments, dataset name, purpose, etc.
             judgment_api_key (Optional[str]): The API key for running evaluations on the Judgment API
             log_results (bool): Whether to log the results to the Judgment API
-
+            rules (Optional[List[Rule]]): Rules to evaluate against scoring results
 
     Returns:
         List[ScoringResult]: The results of the evaluation. Each result is a dictionary containing the fields of a `ScoringResult` object.
@@ -316,7 +322,8 @@ def run_eval(evaluation_run: EvaluationRun, override: bool = False) -> List[Scor
                 metadata=evaluation_run.metadata,
                 judgment_api_key=evaluation_run.judgment_api_key,
                 organization_id=evaluation_run.organization_id,
-                log_results=evaluation_run.log_results
+                log_results=evaluation_run.log_results,
+                rules=evaluation_run.rules
             )
             debug("Sending request to Judgment API")    
             response_data: List[Dict] = execute_api_eval(api_evaluation_run)  # Dicts are `ScoringResult` objs
@@ -346,7 +353,6 @@ def run_eval(evaluation_run: EvaluationRun, override: bool = False) -> List[Scor
                     ]
                 
                 api_results.append(ScoringResult(**filtered_result))
-
     # Run local evals
     if local_scorers:  # List[JudgevalScorer]
         info("Starting local evaluation")
@@ -364,12 +370,11 @@ def run_eval(evaluation_run: EvaluationRun, override: bool = False) -> List[Scor
                 show_indicator=True,
                 _use_bar_indicator=True,
                 throttle_value=0,
-                max_concurrent=100,
+                max_concurrent=MAX_CONCURRENT_EVALUATIONS,
             )
         )
         local_results = results
         info(f"Local evaluation complete with {len(local_results)} results")
-
     # Aggregate the ScorerData from the API and local evaluations
     debug("Merging API and local results")
     merged_results: List[ScoringResult] = merge_results(api_results, local_results)
@@ -377,6 +382,15 @@ def run_eval(evaluation_run: EvaluationRun, override: bool = False) -> List[Scor
 
     info(f"Successfully merged {len(merged_results)} results")
 
+    # Evaluate rules against local scoring results if rules exist (this cant be done just yet)
+    # if evaluation_run.rules and merged_results:
+    #     run_rules(
+    #         local_results=merged_results, 
+    #         rules=evaluation_run.rules, 
+    #         judgment_api_key=evaluation_run.judgment_api_key,
+    #         organization_id=evaluation_run.organization_id
+    #     )
+    
     if evaluation_run.log_results:
         log_evaluation_results(merged_results, evaluation_run)
 

judgeval/scorers/api_scorer.py

@@ -34,22 +34,22 @@ class APIJudgmentScorer(BaseModel):
     @field_validator('score_type')
     def convert_to_enum_value(cls, v):
         """
-        Validates that the `score_type` is a valid `JudgmentMetric` enum value.
-        Converts string values to `JudgmentMetric` enum values.
+        Validates that the `score_type` is a valid `APIScorer` enum value.
+        Converts string values to `APIScorer` enum values.
         """
         debug(f"Attempting to convert score_type value: {v}")
         if isinstance(v, APIScorer):
-            info(f"Using existing JudgmentMetric: {v.value}")
-            return v.value
+            info(f"Using existing APIScorer: {v}")
+            return v
         elif isinstance(v, str):
-            debug(f"Converting string value to JudgmentMetric enum: {v}")
-            return APIScorer[v.upper()].value
+            debug(f"Converting string value to APIScorer enum: {v}")
+            return APIScorer[v.upper()]
         error(f"Invalid score_type value: {v}")
         raise ValueError(f"Invalid value for score_type: {v}")
-    
+
     def __str__(self):
-        return f"JudgmentScorer(score_type={self.score_type}, threshold={self.threshold})"
-    
+        return f"JudgmentScorer(score_type={self.score_type.value}, threshold={self.threshold})"
+
     def to_dict(self) -> dict:
         """
         Converts the scorer configuration to a dictionary format.
@@ -58,7 +58,6 @@ class APIJudgmentScorer(BaseModel):
             dict: A dictionary containing the scorer's configuration
         """
         return {
-            "score_type": self.score_type,
+            "score_type": str(self.score_type.value),  # Convert enum to string for serialization
             "threshold": self.threshold
-        }
-    
+        }

judgeval/scorers/base_scorer.py

@@ -48,5 +48,5 @@ class APIJudgmentScorer(BaseModel):
         raise ValueError(f"Invalid value for score_type: {v}")
     
     def __str__(self):
-        return f"JudgmentScorer(score_type={self.score_type}, threshold={self.threshold})"
+        return f"JudgmentScorer(score_type={self.score_type.value}, threshold={self.threshold})"
     

judgeval/scorers/judgeval_scorers/local_implementations/answer_relevancy/answer_relevancy_scorer.py

@@ -46,7 +46,6 @@ class AnswerRelevancyScorer(JudgevalScorer):
         )
         self.model, self.using_native_model = create_judge(model)
         self.evaluation_model = self.model.get_model_name()
-        print(self.model)
 
     def score_example(
         self,

judgeval/utils/alerts.py

@@ -0,0 +1,43 @@
+"""
+Handling alerts in Judgeval.
+"""
+from enum import Enum
+from typing import Dict, Any, List, Optional
+from pydantic import BaseModel
+
+class AlertStatus(str, Enum):
+    """Status of an alert evaluation."""
+    TRIGGERED = "triggered"
+    NOT_TRIGGERED = "not_triggered"
+
+class AlertResult(BaseModel):
+    """
+    Result of a rule evaluation.
+    
+    Attributes:
+        rule_name: Name of the rule that was evaluated
+        rule_id: Unique identifier of the rule
+        status: Status of the alert (triggered or not)
+        conditions_result: List of condition evaluation results
+        metadata: Dictionary containing example_id, timestamp, and other metadata
+    """
+    rule_name: str
+    rule_id: Optional[str] = None  # The unique identifier of the rule
+    status: AlertStatus
+    conditions_result: List[Dict[str, Any]] = []
+    metadata: Dict[str, Any] = {}
+    
+    @property
+    def example_id(self) -> Optional[str]:
+        """Get example_id from metadata for backward compatibility"""
+        return self.metadata.get("example_id")
+        
+    @property
+    def timestamp(self) -> Optional[str]:
+        """Get timestamp from metadata for backward compatibility"""
+        return self.metadata.get("timestamp")
+    
+    @property
+    def conditions_results(self) -> List[Dict[str, Any]]:
+        """Backwards compatibility property for the conditions_result field"""
+        return self.conditions_result

{judgeval-0.0.14.dist-info → judgeval-0.0.15.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: judgeval
-Version: 0.0.14
+Version: 0.0.15
 Summary: Judgeval Package
 Project-URL: Homepage, https://github.com/JudgmentLabs/judgeval
 Project-URL: Issues, https://github.com/JudgmentLabs/judgeval/issues

{judgeval-0.0.14.dist-info → judgeval-0.0.15.dist-info}/RECORD

@@ -1,33 +1,34 @@
 judgeval/__init__.py,sha256=xiiG4CkeaOtey4fusCd9CBz0BVqzTIbV-K2EFIU0rUM,283
 judgeval/clients.py,sha256=Ns5ljrgPPXUMo7fSPJxO12H64lcPyKeQPIVG_RMi2cM,1162
-judgeval/constants.py,sha256=43hGesvBbX1uzc4KXvjLCVdd6cyZRMSnEJp11oA7h74,2794
-judgeval/evaluation_run.py,sha256=59lG8AUFTKqbY_JVEEA0I093-Pmiy0ERYDK5BuXuEGg,5965
-judgeval/judgment_client.py,sha256=ryGT3A9-Him6oco3WvuHbjB-FVvAR3wCiiGz03eO_Q4,15409
-judgeval/run_evaluation.py,sha256=Cc7BS07WyqsNpQ38HdMdRI782N3DANjM8UcIq9AwaGA,20769
+judgeval/constants.py,sha256=9ndMvpjXEUPF3RlZZXyTXGDVWE3roH9MH6CZ9h38lrc,4700
+judgeval/evaluation_run.py,sha256=qGNafAPLW9tq6KV3eDolIASodLxcSBWQO0mQ0Aq9Cig,6285
+judgeval/judgment_client.py,sha256=HfNcpppOQH2QPHMsGpt5E2YuEgGfp31dTT4ZexhtN7o,23377
+judgeval/rules.py,sha256=6RT3DdMvu-LpjqiSDMAd4d2T8S8pOfc_IBEvTgf3GS4,15665
+judgeval/run_evaluation.py,sha256=4GnezPm_gGjPppJ_9thJqMJ56uHBIuNkZYc8JoYN3Fw,21490
 judgeval/common/__init__.py,sha256=7d24BRxtncpMj3AAJCj8RS7TqgjXmW777HVZH6-3sBs,289
 judgeval/common/exceptions.py,sha256=U-TxHLn7oVMezsMuoYouNDb2XuS8RCggfntYf5_6u4E,565
 judgeval/common/logger.py,sha256=QXN3UMymmKu2iMEMEgATLBnMDjGr_pE2iOSEFoICgg8,6092
-judgeval/common/tracer.py,sha256=qam2suh-0_Cu_B7AWg3AMfEo2TisRZVY1SnAfqhiFQo,33211
-judgeval/common/utils.py,sha256=3WRyyX0tvnnj_VAVlEdtZrfzyWj6zfX04xdpCtE1m5Y,33736
+judgeval/common/tracer.py,sha256=cq-8welgUQSKxHZ1z3rbnSnEir_j19z4bK9XYrVmvEo,36529
+judgeval/common/utils.py,sha256=T1lpObopcH868NIgOTzNViTB33OGadcVWxWcfh2pm3E,33439
 judgeval/data/__init__.py,sha256=YferxwmUqoBi18hrdgro0BD0h4pt20LAqISeUzGMcVU,474
 judgeval/data/api_example.py,sha256=vwWFbI6eJr5VgURCRbuSiMtEXLUbTCih_BcaqEBy-pg,4108
 judgeval/data/example.py,sha256=Rd-eDEM-giYfkfsGh_PBS2wwl15QlQPzbMV-J64Yj5E,2991
 judgeval/data/result.py,sha256=8FIO-bFKPegZuByKRjA2_sumjb8oGWQ5ZeQ1RVz5z2w,4393
 judgeval/data/scorer_data.py,sha256=pYljblCPZrlMIv5Eg7R-clnmsqzUBAwokKjZpwa0DXE,3280
 judgeval/data/datasets/__init__.py,sha256=eO6ayeM_bTGwIt0eDSlTBIIBvXvIWRWWSfYZrZROPiQ,265
-judgeval/data/datasets/dataset.py,sha256=KdAY0KRUB2jxcGmc1XXXheFFcPsGFOIGY-kTwBNQS_Y,12080
-judgeval/data/datasets/eval_dataset_client.py,sha256=DzxWQIiHlbpg6FpmWY6brcSP_h_rGcztk2A_6tQNFys,11411
+judgeval/data/datasets/dataset.py,sha256=FRl2efBQZEpyK_ZTM7FMQQ7wjmtvcHCMFBq8L7O2Wus,12080
+judgeval/data/datasets/eval_dataset_client.py,sha256=azdv6rnyoX5O-d8g0AjNW620yS-7zm4tV1kmyQ63TlE,11531
 judgeval/data/datasets/ground_truth.py,sha256=OTBs3VZe-Wp0vEXEsq14GPZHYtpWT16bhGQTycIvkKc,2057
 judgeval/data/datasets/utils.py,sha256=lQxyl7mevct7JcDSyIrU_8QOzT-EYPWEvoUiAeOdeek,2502
 judgeval/judges/__init__.py,sha256=tyQ5KY88Kp1Ctfw2IJxnVEpy8DnFCtmy04JdPOpp-As,339
 judgeval/judges/base_judge.py,sha256=qhYSFxE21WajYNaT4X-qwWGtpo_tqzBzdqbszSheSD8,1000
 judgeval/judges/litellm_judge.py,sha256=EIL58Teptv8DzZUO3yP2RDQCDq-aoBB6HPZzPdK6KTg,2424
 judgeval/judges/mixture_of_judges.py,sha256=OuGWCuXyqe7s_Y74ij90TJFRfHU-VAFyJVVrwBM0RO0,15532
-judgeval/judges/together_judge.py,sha256=x3jf-tq77QPXHeeoF739f69hE_0VceXD9FHLrVFdGVA,2275
-judgeval/judges/utils.py,sha256=sYxSJq5cI9LtyJaxurcW9IwngALC9Ty8F_Mb8gz81nE,2732
+judgeval/judges/together_judge.py,sha256=l00hhPerAZXg3oYBd8cyMtWsOTNt_0FIqoxhKJKQe3k,2302
+judgeval/judges/utils.py,sha256=9lvUxziGV86ISvVFxYBWc09TWFyAQgUTyPf_a9mD5Rs,2686
 judgeval/scorers/__init__.py,sha256=XcDdLn_s16rSQob0896oj4JXTA8-Xfl271TUEBj6Oew,998
-judgeval/scorers/api_scorer.py,sha256=88kCWr6IetLFn3ziTPG-lwDWvMhFUC6xfINU1MJBoho,2125
-judgeval/scorers/base_scorer.py,sha256=mbOReG88fWaqCnC8F0u5QepRlzgVkuOz89KEKYxrmMc,1794
+judgeval/scorers/api_scorer.py,sha256=PPpropMg_vFyUZULWqRPhtz_h2-NVydBMNnGtRpGk4E,2135
+judgeval/scorers/base_scorer.py,sha256=lz3QWPQQIbtsA-TWUjXYYRfQ96uCaAzqxt7Dn4TJa4s,1800
 judgeval/scorers/exceptions.py,sha256=eGW5CuJgZ5YJBFrE4FHDSF651PO1dKAZ379mJ8gOsfo,178
 judgeval/scorers/judgeval_scorer.py,sha256=T9fkJwFVYMzW88TFr-RWg-Fqmp-cdrA8bLFymqMzOa8,6291
 judgeval/scorers/prompt_scorer.py,sha256=UHkOUts1aIQCoYFcr-sKyucmvv_8ONFE5LZO01aObd0,17825
@@ -53,7 +54,7 @@ judgeval/scorers/judgeval_scorers/local_implementations/answer_correctness/__ini
 judgeval/scorers/judgeval_scorers/local_implementations/answer_correctness/answer_correctness_scorer.py,sha256=PDThn6SzqxgMXT7BpQs2TEBOsgfD5fi6fnKk31qaCTo,10227
 judgeval/scorers/judgeval_scorers/local_implementations/answer_correctness/prompts.py,sha256=5B_G7PPEsfLq6cwWkKWcLuy2k_5RgoOzsW3wOZLIeMk,6703
 judgeval/scorers/judgeval_scorers/local_implementations/answer_relevancy/__init__.py,sha256=r6yae5iaWtlBL_cP8I-1SuhS9dulsy1e7W9Rcz82v6E,169
-judgeval/scorers/judgeval_scorers/local_implementations/answer_relevancy/answer_relevancy_scorer.py,sha256=QG-oxa6-c74VzTuni17RQ9aeT0t1lCuxQXDMznqX8rc,10714
+judgeval/scorers/judgeval_scorers/local_implementations/answer_relevancy/answer_relevancy_scorer.py,sha256=ipER9zyJLq0NqcmxYwfDhavp4rUDYIaDbghR1R0YpaU,10688
 judgeval/scorers/judgeval_scorers/local_implementations/answer_relevancy/prompts.py,sha256=GfbKv595s1a0dB1No_kDsap6gfcr6dYRGiXx0PDb89k,6557
 judgeval/scorers/judgeval_scorers/local_implementations/contextual_precision/__init__.py,sha256=J6tc-T60AVOEaNVuoVU0XIG6dvQri99Q0tnX_Tm-0vc,108
 judgeval/scorers/judgeval_scorers/local_implementations/contextual_precision/contextual_precision_scorer.py,sha256=tRgRyjGpc4Pe3nQ1c-5NeNYFvbulL7YEnoRa9zLp1gc,9649
@@ -78,7 +79,8 @@ judgeval/scorers/judgeval_scorers/local_implementations/summarization/summarizat
 judgeval/scorers/judgeval_scorers/local_implementations/tool_correctness/__init__.py,sha256=JUB3TMqS1OHr6PqpIGqkyiBNbyfUaw7lZuUATjU3_ek,168
 judgeval/scorers/judgeval_scorers/local_implementations/tool_correctness/tool_correctness_scorer.py,sha256=CYGRJY5EuyICYzHrmFdLykwXakX8AC7G3Bhj7p6szfY,5493
 judgeval/tracer/__init__.py,sha256=wy3DYpH8U_z0GO_K_gOSkK0tTTD-u5eLDo0T5xIBoAc,147
-judgeval-0.0.14.dist-info/METADATA,sha256=ZmCAECDNWwzpuES1slYKWcY_U-SMOsjaOdtSoj6wu0I,1283
-judgeval-0.0.14.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-judgeval-0.0.14.dist-info/licenses/LICENSE.md,sha256=tKmCg7k5QOmxPK19XMfzim04QiQJPmgIm0pAn55IJwk,11352
-judgeval-0.0.14.dist-info/RECORD,,
+judgeval/utils/alerts.py,sha256=RgW5R9Dn3Jtim0OyAYDbNzjoX2s6SA4Mw16GyyaikjI,1424
+judgeval-0.0.15.dist-info/METADATA,sha256=ZYYDsxPClqUtpV8_l-9lUSHPnx9eWCj06oR6jeen17g,1283
+judgeval-0.0.15.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+judgeval-0.0.15.dist-info/licenses/LICENSE.md,sha256=tKmCg7k5QOmxPK19XMfzim04QiQJPmgIm0pAn55IJwk,11352
+judgeval-0.0.15.dist-info/RECORD,,