judgeval 0.0.21__py3-none-any.whl → 0.0.23__py3-none-any.whl

judgeval/data/__init__.py

@@ -2,8 +2,7 @@ from judgeval.data.example import Example, ExampleParams
 from judgeval.data.api_example import ProcessExample, create_process_example
 from judgeval.data.scorer_data import ScorerData, create_scorer_data
 from judgeval.data.result import ScoringResult, generate_scoring_result
-from judgeval.data.ground_truth import GroundTruthExample
-
+from judgeval.data.custom_example import CustomExample
 __all__ = [
     "Example",
     "ExampleParams",
@@ -13,5 +12,5 @@ __all__ = [
     "create_scorer_data",
     "ScoringResult",
     "generate_scoring_result",
-    "GroundTruthExample",
+    "CustomExample",
 ]

judgeval/data/custom_example.py

@@ -0,0 +1,98 @@
+from pydantic import BaseModel, Field, field_validator
+from typing import Optional, Dict, Any
+from uuid import uuid4
+from datetime import datetime
+import json
+import warnings
+
+# Brainstorming what are the requirements for the fields?
+class CustomExample(BaseModel):
+    name: Optional[str] = None
+    additional_metadata: Optional[Dict[str, Any]] = None
+    example_id: str = Field(default_factory=lambda: str(uuid4()))
+    example_index: Optional[int] = None
+    timestamp: Optional[str] = None
+    trace_id: Optional[str] = None
+    
+    model_config = {
+        "extra": "allow",  # Allow extra fields with any types
+    }
+    
+    def __init__(self, **data):
+        if 'example_id' not in data:
+            data['example_id'] = str(uuid4())
+        # Set timestamp if not provided
+        if 'timestamp' not in data:
+            data['timestamp'] = datetime.now().isoformat()
+        super().__init__(**data)
+        
+    @field_validator('additional_metadata', mode='before')
+    @classmethod
+    def validate_additional_metadata(cls, v):
+        if v is not None and not isinstance(v, dict):
+            raise ValueError(f"Additional metadata must be a dictionary or None but got {v} of type {type(v)}")
+        return v
+    
+    @field_validator('example_index', mode='before')
+    @classmethod
+    def validate_example_index(cls, v):
+        if v is not None and not isinstance(v, int):
+            raise ValueError(f"Example index must be an integer or None but got {v} of type {type(v)}")
+        return v
+    
+    @field_validator('timestamp', mode='before')
+    @classmethod
+    def validate_timestamp(cls, v):
+        if v is not None and not isinstance(v, str):
+            raise ValueError(f"Timestamp must be a string or None but got {v} of type {type(v)}")
+        return v
+    
+    @field_validator('trace_id', mode='before')
+    @classmethod
+    def validate_trace_id(cls, v):
+        if v is not None and not isinstance(v, str):
+            raise ValueError(f"Trace ID must be a string or None but got {v} of type {type(v)}")
+        return v
+        
+    def to_dict(self):
+        return self.model_dump()
+    
+    def __str__(self):
+        return str(self.model_dump())
+    
+    def model_dump(self, **kwargs):
+        """
+        Custom serialization that handles special cases for fields that might fail standard serialization.
+        """
+        data = super().model_dump(**kwargs)
+        
+        # Get all fields including custom ones
+        all_fields = self.__dict__
+        
+        for field_name, value in all_fields.items():
+            try:
+                # Check if the field has its own serialization method
+                if hasattr(value, 'to_dict'):
+                    data[field_name] = value.to_dict()
+                elif hasattr(value, 'model_dump'):
+                    data[field_name] = value.model_dump()
+                # Field is already in data from super().model_dump()
+                elif field_name in data:
+                    continue
+                else:
+                    # Try standard JSON serialization
+                    json.dumps(value)
+                    data[field_name] = value
+            except (TypeError, OverflowError, ValueError):
+                # Handle non-serializable objects
+                try:
+                    # Try converting to string
+                    data[field_name] = str(value)
+                except Exception as _:
+                    # If all else fails, store as None and optionally warn
+                    warnings.warn(f"Could not serialize field {field_name}, setting to None")
+                    data[field_name] = None
+        
+        return data
+    
+    

judgeval/data/datasets/dataset.py

@@ -7,12 +7,11 @@ import yaml
 from dataclasses import dataclass, field
 from typing import List, Union, Literal
 
-from judgeval.data import Example, GroundTruthExample
+from judgeval.data import Example
 from judgeval.common.logger import debug, error, warning, info
 
 @dataclass
 class EvalDataset:
-    ground_truths: List[GroundTruthExample]
     examples: List[Example]
     _alias: Union[str, None] = field(default=None)
     _id: Union[str, None] = field(default=None)
@@ -21,13 +20,11 @@ class EvalDataset:
     def __init__(self, 
                  judgment_api_key: str = os.getenv("JUDGMENT_API_KEY"),  
                  organization_id: str = os.getenv("JUDGMENT_ORG_ID"),
-                 ground_truths: List[GroundTruthExample] = [], 
                  examples: List[Example] = [],
                  ):
-        debug(f"Initializing EvalDataset with {len(ground_truths)} ground truths and {len(examples)} examples")
+        debug(f"Initializing EvalDataset with {len(examples)} examples")
         if not judgment_api_key:
             warning("No judgment_api_key provided")
-        self.ground_truths = ground_truths
         self.examples = examples
         self._alias = None
         self._id = None
@@ -37,38 +34,13 @@ class EvalDataset:
     def add_from_json(self, file_path: str) -> None:
         debug(f"Loading dataset from JSON file: {file_path}")
         """
-        Adds examples and ground truths from a JSON file.
+        Adds examples from a JSON file.
 
-        The format of the JSON file is expected to be a dictionary with two keys: "examples" and "ground_truths". 
-        The value of each key is a list of dictionaries, where each dictionary represents an example or ground truth.
+        The format of the JSON file is expected to be a dictionary with one key: "examples". 
+        The value of the key is a list of dictionaries, where each dictionary represents an example.
 
         The JSON file is expected to have the following format:
         {
-            "ground_truths": [
-                {
-                    "input": "test input",
-                    "actual_output": null,
-                    "expected_output": "expected output",
-                    "context": [
-                    "context1"
-                ],
-                "retrieval_context": [
-                    "retrieval1"
-                ],
-                "additional_metadata": {
-                    "key": "value"
-                },
-                "comments": "test comment",
-                "tools_called": [
-                    "tool1"
-                ],
-                "expected_tools": [
-                    "tool1"
-                ],
-                "source_file": "test.py",
-                "trace_id": "094121"
-            }
-        ],
         "examples": [
             {
                 "input": "test input",
@@ -103,7 +75,6 @@ class EvalDataset:
             with open(file_path, "r") as file:
                 payload = json.load(file)
                 examples = payload.get("examples", [])
-                ground_truths = payload.get("ground_truths", [])
         except FileNotFoundError:
             error(f"JSON file not found: {file_path}")
             raise FileNotFoundError(f"The file {file_path} was not found.")
@@ -111,21 +82,17 @@ class EvalDataset:
             error(f"Invalid JSON file: {file_path}")
             raise ValueError(f"The file {file_path} is not a valid JSON file.")
 
-        info(f"Added {len(examples)} examples and {len(ground_truths)} ground truths from JSON")
+        info(f"Added {len(examples)} examples from JSON")
         new_examples = [Example(**e) for e in examples]
         for e in new_examples:
             self.add_example(e)
-
-        new_ground_truths = [GroundTruthExample(**g) for g in ground_truths]
-        for g in new_ground_truths:
-            self.add_ground_truth(g)
   
     def add_from_csv(
         self, 
         file_path: str,
         ) -> None:
         """
-        Add Examples and GroundTruthExamples from a CSV file.
+        Add Examples from a CSV file.
         """
         try:
             import pandas as pd
@@ -144,14 +111,14 @@ class EvalDataset:
         "expected_tools", "name", "comments", "source_file", "example", \
         "trace_id"
 
-        We want to collect the examples and ground truths separately which can
+        We want to collect the examples separately which can
         be determined by the "example" column. If the value is True, then it is an
-        example, otherwise it is a ground truth.
+        example
 
         We also assume that if there are multiple retrieval contexts or contexts, they are separated by semicolons.
         This can be adjusted using the `context_delimiter` and `retrieval_context_delimiter` parameters.
         """
-        examples, ground_truths = [], []
+        examples = []
 
         for _, row in df.iterrows():
             data = {
@@ -174,49 +141,20 @@ class EvalDataset:
                     examples.append(e)
                 else:
                     raise ValueError("Every example must have an 'input' and 'actual_output' field.")
-            else:
-                # GroundTruthExample has `comments` and `source_file` fields
-                data["comments"] = row["comments"] if pd.notna(row["comments"]) else None
-                data["source_file"] = row["source_file"] if pd.notna(row["source_file"]) else None
-                # every GroundTruthExample has `input` field
-                if data["input"] is not None:
-                    g = GroundTruthExample(**data)
-                    ground_truths.append(g)
-                else:
-                    raise ValueError("Every ground truth must have an 'input' field.")
+           
 
         for e in examples:
             self.add_example(e)
 
-        for g in ground_truths:
-            self.add_ground_truth(g)
-
     def add_from_yaml(self, file_path: str) -> None:
         debug(f"Loading dataset from YAML file: {file_path}")
         """
-        Adds examples and ground truths from a YAML file.
+        Adds examples from a YAML file.
 
-        The format of the YAML file is expected to be a dictionary with two keys: "examples" and "ground_truths". 
-        The value of each key is a list of dictionaries, where each dictionary represents an example or ground truth.
+        The format of the YAML file is expected to be a dictionary with one key: "examples". 
+        The value of the key is a list of dictionaries, where each dictionary represents an example.
 
         The YAML file is expected to have the following format:
-        ground_truths:
-          - input: "test input"
-            actual_output: null
-            expected_output: "expected output"
-            context:
-              - "context1"
-            retrieval_context:
-              - "retrieval1"
-            additional_metadata:
-              key: "value"
-            comments: "test comment"
-            tools_called:
-              - "tool1"
-            expected_tools:
-              - "tool1"
-            source_file: "test.py"
-            trace_id: "094121"
         examples:
           - input: "test input"
             actual_output: "test output"
@@ -244,7 +182,6 @@ class EvalDataset:
                 if payload is None:
                     raise ValueError("The YAML file is empty.")
                 examples = payload.get("examples", [])
-                ground_truths = payload.get("ground_truths", [])
         except FileNotFoundError:
             error(f"YAML file not found: {file_path}")
             raise FileNotFoundError(f"The file {file_path} was not found.")
@@ -252,25 +189,18 @@ class EvalDataset:
             error(f"Invalid YAML file: {file_path}")
             raise ValueError(f"The file {file_path} is not a valid YAML file.")
 
-        info(f"Added {len(examples)} examples and {len(ground_truths)} ground truths from YAML")
+        info(f"Added {len(examples)} examples from YAML")
         new_examples = [Example(**e) for e in examples]
         for e in new_examples:
             self.add_example(e)
 
-        new_ground_truths = [GroundTruthExample(**g) for g in ground_truths]
-        for g in new_ground_truths:
-            self.add_ground_truth(g)
-
     def add_example(self, e: Example) -> None:
         self.examples = self.examples + [e]
         # TODO if we need to add rank, then we need to do it here
-
-    def add_ground_truth(self, g: GroundTruthExample) -> None:
-        self.ground_truths = self.ground_truths + [g]
     
     def save_as(self, file_type: Literal["json", "csv", "yaml"], dir_path: str, save_name: str = None) -> None:
         """
-        Saves the dataset as a file. Save both the ground truths and examples.
+        Saves the dataset as a file. Save only the examples.
 
         Args:
             file_type (Literal["json", "csv"]): The file type to save the dataset as.
@@ -285,7 +215,6 @@ class EvalDataset:
             with open(complete_path, "w") as file:
                 json.dump(
                     {
-                        "ground_truths": [g.to_dict() for g in self.ground_truths],
                         "examples": [e.to_dict() for e in self.examples],
                     },
                     file,
@@ -319,24 +248,7 @@ class EvalDataset:
                         ]
                     )
                 
-                for g in self.ground_truths:
-                    writer.writerow(
-                        [
-                            g.input,
-                            g.actual_output,
-                            g.expected_output,
-                            ";".join(g.context),
-                            ";".join(g.retrieval_context),
-                            g.additional_metadata,
-                            ";".join(g.tools_called),
-                            ";".join(g.expected_tools),
-                            None,  # GroundTruthExample does not have name
-                            g.comments,
-                            g.source_file,
-                            False,  # Adding a GroundTruthExample, not an Example
-                            g.trace_id
-                        ]
-                    )
+                
         elif file_type == "yaml":
             with open(complete_path, "w") as file:
                 yaml_data = {
@@ -358,24 +270,6 @@ class EvalDataset:
                         }
                         for e in self.examples
                     ],
-                    "ground_truths": [
-                        {
-                            "input": g.input,
-                            "actual_output": g.actual_output,
-                            "expected_output": g.expected_output,
-                            "context": g.context,
-                            "retrieval_context": g.retrieval_context,
-                            "additional_metadata": g.additional_metadata,
-                            "tools_called": g.tools_called,
-                            "expected_tools": g.expected_tools,
-                            "name": None,  # GroundTruthExample does not have name
-                            "comments": g.comments,
-                            "source_file": g.source_file,
-                            "example": False,  # Adding a GroundTruthExample, not an Example
-                            "trace_id": g.trace_id
-                        }
-                        for g in self.ground_truths
-                    ]
                 }
                 yaml.dump(yaml_data, file, default_flow_style=False)
         else:
@@ -391,7 +285,6 @@ class EvalDataset:
     def __str__(self):
         return (
             f"{self.__class__.__name__}("
-            f"ground_truths={self.ground_truths}, "
             f"examples={self.examples}, "
             f"_alias={self._alias}, "
             f"_id={self._id}"

judgeval/data/datasets/eval_dataset_client.py

@@ -11,7 +11,7 @@ from judgeval.constants import (
     JUDGMENT_DATASETS_EDIT_API_URL,
     JUDGMENT_DATASETS_EXPORT_JSONL_API_URL
 )
-from judgeval.data import Example, GroundTruthExample
+from judgeval.data import Example
 from judgeval.data.datasets import EvalDataset
 
 
@@ -35,7 +35,6 @@ class EvalDatasetClient:
         Mock request:
         dataset = {
             "alias": alias,
-            "ground_truths": [...],
             "examples": [...],
             "overwrite": overwrite
         } ==>
@@ -55,7 +54,6 @@ class EvalDatasetClient:
             )
             content = {
                     "alias": alias,
-                    "ground_truths": [g.to_dict() for g in dataset.ground_truths],
                     "examples": [e.to_dict() for e in dataset.examples],
                     "overwrite": overwrite,
                 }
@@ -102,7 +100,6 @@ class EvalDatasetClient:
         } 
         ==>
         {
-            "ground_truths": [...],
             "examples": [...],
             "_alias": alias,
             "_id": "..."  # ID of the dataset
@@ -142,7 +139,6 @@ class EvalDatasetClient:
 
                 info(f"Successfully pulled dataset with alias '{alias}'")
                 payload = response.json()
-                dataset.ground_truths = [GroundTruthExample(**g) for g in payload.get("ground_truths", [])]
                 dataset.examples = [Example(**e) for e in payload.get("examples", [])]
                 dataset._alias = payload.get("_alias")
                 dataset._id = payload.get("_id")
@@ -164,8 +160,8 @@ class EvalDatasetClient:
         } 
         ==>
         {
-            "test_dataset_1": {"examples_count": len(dataset1.examples), "ground_truths_count": len(dataset1.ground_truths)},
-            "test_dataset_2": {"examples_count": len(dataset2.examples), "ground_truths_count": len(dataset2.ground_truths)},
+            "test_dataset_1": {"examples_count": len(dataset1.examples)},
+            "test_dataset_2": {"examples_count": len(dataset2.examples)},
             ...
         }
         """
@@ -209,15 +205,14 @@ class EvalDatasetClient:
 
                 return payload
 
-    def edit_dataset(self, alias: str, examples: List[Example], ground_truths: List[GroundTruthExample]) -> bool:
+    def edit_dataset(self, alias: str, examples: List[Example]) -> bool:
         """
-        Edits the dataset on Judgment platform by adding new examples and ground truths
+        Edits the dataset on Judgment platform by adding new examples
 
         Mock request:
         {
             "alias": alias,
             "examples": [...],
-            "ground_truths": [...],
             "judgment_api_key": self.judgment_api_key
         }
         """
@@ -234,7 +229,6 @@ class EvalDatasetClient:
             content = {
                 "alias": alias,
                 "examples": [e.to_dict() for e in examples],
-                "ground_truths": [g.to_dict() for g in ground_truths],
             }
 
             try:

judgeval/data/datasets/utils.py

@@ -1,73 +0,0 @@
-from typing import List, Optional
-
-from judgeval.data import Example, GroundTruthExample
-
-
-def examples_to_ground_truths(examples: List[Example]) -> List[GroundTruthExample]:
-    """
-    Convert a list of `Example` objects to a list of `GroundTruthExample` objects.
-
-    Args:
-        examples (List[Example]): A list of `Example` objects to convert.
-
-    Returns:
-        List[GroundTruthExample]: A list of `GroundTruthExample` objects.
-    """
-
-    if not isinstance(examples, list):
-        raise TypeError("Input should be a list of `Example` objects")
-
-    ground_truths = []
-    ground_truths = []
-    for e in examples:
-        g_truth = {
-            "input": e.input,
-            "actual_output": e.actual_output,
-            "expected_output": e.expected_output,
-            "context": e.context,
-            "retrieval_context": e.retrieval_context,
-            "tools_called": e.tools_called,
-            "expected_tools": e.expected_tools,
-        }
-        ground_truths.append(GroundTruthExample(**g_truth))
-    return ground_truths
-
-
-def ground_truths_to_examples(
-    ground_truths: List[GroundTruthExample], 
-    _alias: Optional[str] = None,
-    _id: Optional[str] = None,
-    ) -> List[Example]:
-    """
-    Converts a list of `GroundTruthExample` objects to a list of `Example` objects.
-
-    Args:
-        ground_truths (List[GroundTruthExample]): A list of `GroundTruthExample` objects to convert.
-        _alias (Optional[str]): The alias of the dataset.
-        _id (Optional[str]): The ID of the dataset.
-
-    Returns:
-        List[Example]: A list of `Example` objects.
-    """
-
-    if not isinstance(ground_truths, list):
-        raise TypeError("Input should be a list of `GroundTruthExample` objects")
-
-    examples = []
-    for index, ground_truth in enumerate(ground_truths):
-        e = Example(
-            input=ground_truth.input,
-            actual_output=ground_truth.actual_output,
-            expected_output=ground_truth.expected_output,
-            context=ground_truth.context,
-            retrieval_context=ground_truth.retrieval_context,
-            additional_metadata=ground_truth.additional_metadata,
-            tools_called=ground_truth.tools_called,
-            expected_tools=ground_truth.expected_tools,
-            comments=ground_truth.comments,
-            _dataset_alias=_alias,
-            _dataset_id=_id,
-            _dataset_rank=index,
-        )
-        examples.append(e)
-    return examples

judgeval/data/ground_truth.py

@@ -1,54 +0,0 @@
-from pydantic import BaseModel
-from typing import Optional, Dict, List
-
-
-class GroundTruthExample(BaseModel):
-    """
-    GroundTruthExample is the atomic unit of a `Dataset`. It is essentially the same
-    as an `Example`, but the `actual_output` field is optional to enable users to 
-    run their workflow on the `input` field at test-time to evaluate their current 
-    workflow's performance.
-    """
-    input: str
-    actual_output: Optional[str] = None
-    expected_output: Optional[str] = None
-    context: Optional[List[str]] = None
-    retrieval_context: Optional[List[str]] = None
-    additional_metadata: Optional[Dict] = None
-    comments: Optional[str] = None
-    tools_called: Optional[List[str]] = None
-    expected_tools: Optional[List[str]] = None
-    source_file: Optional[str] = None
-    trace_id: Optional[str] = None
-
-    def to_dict(self):
-        return {
-            "input": self.input,
-            "actual_output": self.actual_output,
-            "expected_output": self.expected_output,
-            "context": self.context,
-            "retrieval_context": self.retrieval_context,
-            "additional_metadata": self.additional_metadata,
-            "comments": self.comments,
-            "tools_called": self.tools_called,
-            "expected_tools": self.expected_tools,
-            "source_file": self.source_file,
-            "trace_id": self.trace_id,
-        }
-    
-    def __str__(self):
-        return (
-            f"{self.__class__.__name__}("
-            f"input={self.input}, "
-            f"actual_output={self.actual_output}, "
-            f"expected_output={self.expected_output}, "
-            f"context={self.context}, "
-            f"retrieval_context={self.retrieval_context}, "
-            f"additional_metadata={self.additional_metadata}, "
-            f"comments={self.comments}, "
-            f"tools_called={self.tools_called}, "
-            f"expected_tools={self.expected_tools}, "
-            f"source_file={self.source_file}, "
-            f"trace_id={self.trace_id}"
-            f")"
-        )

judgeval/judgment_client.py

@@ -10,7 +10,6 @@ from judgeval.data.datasets import EvalDataset, EvalDatasetClient
 from judgeval.data import (
     ScoringResult, 
     Example,
-    GroundTruthExample
 )
 from judgeval.scorers import (
     APIJudgmentScorer, 
@@ -283,11 +282,11 @@ class JudgmentClient:
         """
         return self.eval_dataset_client.pull_all_user_dataset_stats()
     
-    def edit_dataset(self, alias: str, examples: List[Example], ground_truths: List[GroundTruthExample]) -> bool:
+    def edit_dataset(self, alias: str, examples: List[Example]) -> bool:
         """
-        Edits the dataset on Judgment platform by adding new examples and ground truths
+        Edits the dataset on Judgment platform by adding new examples
         """
-        return self.eval_dataset_client.edit_dataset(alias, examples, ground_truths)
+        return self.eval_dataset_client.edit_dataset(alias, examples)
     
     # Maybe add option where you can pass in the EvaluationRun object and it will pull the eval results from the backend
     def pull_eval(self, project_name: str, eval_run_name: str) -> List[Dict[str, Union[str, List[ScoringResult]]]]:

judgeval/scorers/judgeval_scorer.py

@@ -116,8 +116,8 @@ class JudgevalScorer:
         For unit testing, determines whether the test case passes or fails
         """
         warning("Attempting to call unimplemented success_check method")
-        error("success_check method not implemented")
-        raise NotImplementedError("You must implement the `passes` method in your custom scorer")
+        error("_success_check method not implemented")
+        raise NotImplementedError("You must implement the `_success_check` method in your custom scorer")
 
     def __str__(self):
         debug("Converting JudgevalScorer instance to string representation")

{judgeval-0.0.21.dist-info → judgeval-0.0.23.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: judgeval
-Version: 0.0.21
+Version: 0.0.23
 Summary: Judgeval Package
 Project-URL: Homepage, https://github.com/JudgmentLabs/judgeval
 Project-URL: Issues, https://github.com/JudgmentLabs/judgeval/issues

{judgeval-0.0.21.dist-info → judgeval-0.0.23.dist-info}/RECORD

@@ -2,7 +2,7 @@ judgeval/__init__.py,sha256=dtXxsCmI4eEsZdGSUMy8P_pA0bc2-OSGAgb2C__yJoA,252
 judgeval/clients.py,sha256=6VQmEqmfCngUdS2MuPBIpHvtDFqOENm8-_BmMvjLyRQ,944
 judgeval/constants.py,sha256=VhJppAECTUDQwzC_FpzJw2wPlkYoogsadHxaJIY_J8U,5073
 judgeval/evaluation_run.py,sha256=RgJD60lJsunNQzObjo7iXnAzXWgubCLOAAuuamAAuoI,6354
-judgeval/judgment_client.py,sha256=5lqp9X67qPzBUu7kQYETslsc3L5JjxrDVgVLslF07A0,24173
+judgeval/judgment_client.py,sha256=e-2e4KK-xy8-WLgzg8H0D6pZC8By9IWdu2iK-lHe39A,24076
 judgeval/rules.py,sha256=ebsiDEBVAnYTQxwVNvh_RpmKeWBnjQXgHs8KofTjcAs,15526
 judgeval/run_evaluation.py,sha256=YOzkyeWl-r3vaz0jB5nM-1VULi7ALmJ9_f58ENqexXk,23827
 judgeval/common/__init__.py,sha256=7d24BRxtncpMj3AAJCj8RS7TqgjXmW777HVZH6-3sBs,289
@@ -10,16 +10,17 @@ judgeval/common/exceptions.py,sha256=U-TxHLn7oVMezsMuoYouNDb2XuS8RCggfntYf5_6u4E
 judgeval/common/logger.py,sha256=KO75wWXCxhUHUMvLaTU31ZzOk6tkZBa7heQ7y0f-zFE,6062
 judgeval/common/tracer.py,sha256=WFjFNf3NZ2BN8UAu2MG0F3Om9LgJNma3m_GrxyXgJqE,46655
 judgeval/common/utils.py,sha256=LUQV5JfDr6wj7xHAJoNq-gofNZ6mjXbeKrGKzBME1KM,33533
-judgeval/data/__init__.py,sha256=QykVE22Qf-I2f1g-jC9-iQyLNXgDmX1-vHbCgZg8Ra8,558
+judgeval/data/__init__.py,sha256=6ADbugtS3AporRv23Hxm67qcghU4tj0OScS8t3xLd6U,549
 judgeval/data/api_example.py,sha256=dzkrQ0xno08y6qNfqL2djXbapUyc2B2aQ5iANn0o4CY,3667
+judgeval/data/custom_example.py,sha256=C-j9iVenBy52dwnL6PIjJAdKsBO1ajKjsaRr4RJthUo,3676
 judgeval/data/example.py,sha256=BhGBhamFWgH6wtvrRYM8dGtDfXh-cDxDhtNL5Gbdz_M,5892
-judgeval/data/ground_truth.py,sha256=OTBs3VZe-Wp0vEXEsq14GPZHYtpWT16bhGQTycIvkKc,2057
+judgeval/data/ground_truth.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 judgeval/data/result.py,sha256=4fgjKtUmT3br7K6fkRiNIxTGKUuwMeGyRLqzkpxwXKE,4436
 judgeval/data/scorer_data.py,sha256=JVlaTx1EP2jw2gh3Vgx1CSEsvIFABAN26IquKyxwiJQ,3273
 judgeval/data/datasets/__init__.py,sha256=IdNKhQv9yYZ_op0rdBacrFaFVmiiYQ3JTzXzxOTsEVQ,176
-judgeval/data/datasets/dataset.py,sha256=LrBK8y3y1R9_BKmXxTzdXMMIQvXlq7tf7TM-u7jgSxE,16839
-judgeval/data/datasets/eval_dataset_client.py,sha256=QsfHyFC4WePV7uJGYUVjiIwtk1Ie_VpWUrnd2Q4kKdU,11479
-judgeval/data/datasets/utils.py,sha256=6DpGCPmGFNOKIGNcVCOSjTOdWemrpAuYnlo778sGG7g,2455
+judgeval/data/datasets/dataset.py,sha256=DjJNy-qvviXMGBl_JhiBzvgiJH1_3rYtAWeHP6Daw6E,11897
+judgeval/data/datasets/eval_dataset_client.py,sha256=B4bRy0Di2oFlaBbvp4_hRx2g_9e6Cs0y3ZUT9reMyhw,10926
+judgeval/data/datasets/utils.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 judgeval/judges/__init__.py,sha256=6X7VSwrwsdxGBNxCyapVRWGghhKOy3MVxFNMQ62kCXM,308
 judgeval/judges/base_judge.py,sha256=ch_S7uBB7lyv44Lf1d7mIGFpveOO58zOkkpImKgd9_4,994
 judgeval/judges/litellm_judge.py,sha256=EIL58Teptv8DzZUO3yP2RDQCDq-aoBB6HPZzPdK6KTg,2424
@@ -30,7 +31,7 @@ judgeval/scorers/__init__.py,sha256=gkeKJvjXhswCnkEyjijrVvGVM3Om86egrZ-PUOGvNvI,
 judgeval/scorers/api_scorer.py,sha256=NQ_CrrUPhSUk1k2Q8rKpCG_TU2FT32sFEqvb-Yi54B0,2688
 judgeval/scorers/base_scorer.py,sha256=xdUlY3CnLdCQ1Z5iUeY22Bim5v-OQruZmaVF_4Y1mC0,2183
 judgeval/scorers/exceptions.py,sha256=eGW5CuJgZ5YJBFrE4FHDSF651PO1dKAZ379mJ8gOsfo,178
-judgeval/scorers/judgeval_scorer.py,sha256=oIkfoGXA09wL_vcK1DRibzQSA-MFNa-hmw1IhGBErf8,6592
+judgeval/scorers/judgeval_scorer.py,sha256=jq_rzfTG0XBTuLCaa6TlaK4YcT-LlgsO1LEm6hpOYdg,6601
 judgeval/scorers/prompt_scorer.py,sha256=PaAs2qRolw1P3_I061Xvk9qzvF4O-JR8g_39RqXnHcM,17728
 judgeval/scorers/score.py,sha256=GALVmeApP1Cyih2vY93zRaU6RShtW4jJDG47Pm6yfnw,18657
 judgeval/scorers/utils.py,sha256=iHQVTlIANbmCTXz9kTeSdOytgUZ_T74Re61ajqsk_WQ,6827
@@ -88,7 +89,7 @@ judgeval/scorers/judgeval_scorers/local_implementations/summarization/prompts.py
 judgeval/scorers/judgeval_scorers/local_implementations/summarization/summarization_scorer.py,sha256=Qk7lwHgRPYeGoxTOyclAh1VfGItfvHJ6l1t7Nk3SWFM,20927
 judgeval/tracer/__init__.py,sha256=wy3DYpH8U_z0GO_K_gOSkK0tTTD-u5eLDo0T5xIBoAc,147
 judgeval/utils/alerts.py,sha256=RgW5R9Dn3Jtim0OyAYDbNzjoX2s6SA4Mw16GyyaikjI,1424
-judgeval-0.0.21.dist-info/METADATA,sha256=jQW4w6jGNaHvPWTcqX3ZGr_SKeCpNl7DsNr-cwrYHsA,1378
-judgeval-0.0.21.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-judgeval-0.0.21.dist-info/licenses/LICENSE.md,sha256=tKmCg7k5QOmxPK19XMfzim04QiQJPmgIm0pAn55IJwk,11352
-judgeval-0.0.21.dist-info/RECORD,,
+judgeval-0.0.23.dist-info/METADATA,sha256=EkRIGemm8UvM5J4RBR5KVzBfn0XTBBYvJjRM4-F0s0w,1378
+judgeval-0.0.23.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+judgeval-0.0.23.dist-info/licenses/LICENSE.md,sha256=tKmCg7k5QOmxPK19XMfzim04QiQJPmgIm0pAn55IJwk,11352
+judgeval-0.0.23.dist-info/RECORD,,