Skip to content

feat: Limit metric registry support to only custom code execution metric and llm based metric in SDK #6519

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
copybara-service merged 1 commit into from
Apr 1, 2026
+5 −23
Merged

feat: Limit metric registry support to only custom code execution metric and llm based metric in SDK #6519

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 3 additions & 7 deletions tests/unit/vertexai/genai/replays/test_evaluation_metric.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,13 +24,12 @@

def test_create_and_get_evaluation_metric(client):
client._api_client._http_options.api_version = "v1beta1"
client._api_client._http_options.base_url = (
"https://us-central1-staging-aiplatform.sandbox.googleapis.com/"
)
result = client.evals.create_evaluation_metric(
display_name="test_metric",
description="test_description",
metric=types.RubricMetric.GENERAL_QUALITY,
metric=types.LLMMetric(
name="custom_llm_metric", prompt_template="test_prompt_template"
),
)
assert isinstance(result, str)
assert re.match(
Expand All @@ -44,9 +43,6 @@ def test_create_and_get_evaluation_metric(client):

def test_list_evaluation_metrics(client):
client._api_client._http_options.api_version = "v1beta1"
client._api_client._http_options.base_url = (
"https://us-central1-staging-aiplatform.sandbox.googleapis.com/"
)
response = client.evals.list_evaluation_metrics()
assert isinstance(response, types.ListEvaluationMetricsResponse)
assert len(response.evaluation_metrics) >= 0
Expand Down
18 changes: 2 additions & 16 deletions vertexai/_genai/_transformers.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -187,22 +187,8 @@ def t_metric_for_registry(
if metric_name:
metric_name = metric_name.lower()

# Handle standard computation metrics
if metric_name == "exact_match":
metric_payload_item["exact_match_spec"] = {}
elif metric_name == "bleu":
metric_payload_item["bleu_spec"] = {}
elif metric_name and metric_name.startswith("rouge"):
rouge_type = metric_name.replace("_", "")
metric_payload_item["rouge_spec"] = {"rouge_type": rouge_type}
# API Pre-defined metrics
elif metric_name and metric_name in _evals_constant.SUPPORTED_PREDEFINED_METRICS:
metric_payload_item["predefined_metric_spec"] = {
"metric_spec_name": metric_name,
"metric_spec_parameters": metric.metric_spec_parameters,
}
# Custom Code Execution Metric
elif hasattr(metric, "remote_custom_function") and metric.remote_custom_function:
if hasattr(metric, "remote_custom_function") and metric.remote_custom_function:
metric_payload_item["custom_code_execution_spec"] = {
"evaluation_function": metric.remote_custom_function
}
Expand All @@ -217,7 +203,7 @@ def t_metric_for_registry(
"evaluation_function": metric.custom_function
}

# Map LLM-based metrics to the new llm_based_metric_spec
# LLM-based metric
elif (hasattr(metric, "prompt_template") and metric.prompt_template) or (
hasattr(metric, "rubric_group_name") and metric.rubric_group_name
):
Expand Down
Loading