Skip to main content

JSON Evaluators

Evaluating extraction and function calling applications often comes down to validation that the LLM's string output can be parsed correctly and how it compares to a reference object. The following JSON validators provide functionality to check your model's output consistently.

JsonValidityEvaluator

The JsonValidityEvaluator is designed to check the validity of a JSON string prediction.

Overview:

  • Requires Input?: No
  • Requires Reference?: No
from langchain.evaluation import JsonValidityEvaluator

evaluator = JsonValidityEvaluator()
# Equivalently
# evaluator = load_evaluator("json_validity")
prediction = '{"name": "John", "age": 30, "city": "New York"}'

result = evaluator.evaluate_strings(prediction=prediction)
print(result)

API Reference:

{'score': 1}
prediction = '{"name": "John", "age": 30, "city": "New York",}'
result = evaluator.evaluate_strings(prediction=prediction)
print(result)
{'score': 0, 'reasoning': 'Expecting property name enclosed in double quotes: line 1 column 48 (char 47)'}

JsonEqualityEvaluator

The JsonEqualityEvaluator assesses whether a JSON prediction matches a given reference after both are parsed.

Overview:

  • Requires Input?: No
  • Requires Reference?: Yes
from langchain.evaluation import JsonEqualityEvaluator

evaluator = JsonEqualityEvaluator()
# Equivalently
# evaluator = load_evaluator("json_equality")
result = evaluator.evaluate_strings(prediction='{"a": 1}', reference='{"a": 1}')
print(result)

API Reference:

{'score': True}
result = evaluator.evaluate_strings(prediction='{"a": 1}', reference='{"a": 2}')
print(result)
{'score': False}

The evaluator also by default lets you provide a dictionary directly

result = evaluator.evaluate_strings(prediction={"a": 1}, reference={"a": 2})
print(result)
{'score': False}

JsonEditDistanceEvaluator

The JsonEditDistanceEvaluator computes a normalized Damerau-Levenshtein distance between two "canonicalized" JSON strings.

Overview:

  • Requires Input?: No
  • Requires Reference?: Yes
  • Distance Function: Damerau-Levenshtein (by default)

Note: Ensure that rapidfuzz is installed or provide an alternative string_distance function to avoid an ImportError.

from langchain.evaluation import JsonEditDistanceEvaluator

evaluator = JsonEditDistanceEvaluator()
# Equivalently
# evaluator = load_evaluator("json_edit_distance")

result = evaluator.evaluate_strings(
prediction='{"a": 1, "b": 2}', reference='{"a": 1, "b": 3}'
)
print(result)
{'score': 0.07692307692307693}
# The values are canonicalized prior to comparison
result = evaluator.evaluate_strings(
prediction="""
{
"b": 3,
"a": 1
}""",
reference='{"a": 1, "b": 3}',
)
print(result)
{'score': 0.0}
# Lists maintain their order, however
result = evaluator.evaluate_strings(
prediction='{"a": [1, 2]}', reference='{"a": [2, 1]}'
)
print(result)
{'score': 0.18181818181818182}
# You can also pass in objects directly
result = evaluator.evaluate_strings(prediction={"a": 1}, reference={"a": 2})
print(result)
{'score': 0.14285714285714285}

JsonSchemaEvaluator

The JsonSchemaEvaluator validates a JSON prediction against a provided JSON schema. If the prediction conforms to the schema, it returns a score of True (indicating no errors). Otherwise, it returns a score of 0 (indicating an error).

Overview:

  • Requires Input?: Yes
  • Requires Reference?: Yes (A JSON schema)
  • Score: True (No errors) or False (Error occurred)
from langchain.evaluation import JsonSchemaEvaluator

evaluator = JsonSchemaEvaluator()
# Equivalently
# evaluator = load_evaluator("json_schema_validation")

result = evaluator.evaluate_strings(
prediction='{"name": "John", "age": 30}',
reference={
"type": "object",
"properties": {"name": {"type": "string"}, "age": {"type": "integer"}},
},
)
print(result)

API Reference:

{'score': True}
result = evaluator.evaluate_strings(
prediction='{"name": "John", "age": 30}',
reference='{"type": "object", "properties": {"name": {"type": "string"}, "age": {"type": "integer"}}}',
)
print(result)
{'score': True}
result = evaluator.evaluate_strings(
prediction='{"name": "John", "age": 30}',
reference='{"type": "object", "properties": {"name": {"type": "string"},'
'"age": {"type": "integer", "minimum": 66}}}',
)
print(result)
{'score': False, 'reasoning': "<ValidationError: '30 is less than the minimum of 66'>"}

Help us out by providing feedback on this documentation page: