Skip to content

feat(bedrock): Record inferenceConfig parameters #8232

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

Draft
alexander-alderman-webb wants to merge 1 commit into
base: master
Choose a base branch
from
+92 −1
Draft

feat(bedrock): Record inferenceConfig parameters #8232

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
21 changes: 20 additions & 1 deletion sentry_sdk/integrations/boto3/_services/bedrock.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,12 +28,31 @@ def get_request_attributes(self, ctx: "AwsCallContext") -> "Attributes":
if ctx.operation_name != "Converse" and ctx.operation_name != "ConverseStream":
return {}

return {
attributes: "Attributes" = {
SPANDATA.GEN_AI_PROVIDER_NAME: "aws.bedrock",
SPANDATA.GEN_AI_OPERATION_NAME: "chat",
SPANDATA.GEN_AI_RESPONSE_STREAMING: ctx.operation_name == "ConverseStream",
}

inference_config = ctx.params.get("inferenceConfig")

if not isinstance(inference_config, dict):
return attributes

max_tokens = inference_config.get("maxTokens")
if max_tokens is not None:
attributes[SPANDATA.GEN_AI_REQUEST_MAX_TOKENS] = max_tokens

temperature = inference_config.get("temperature")
if temperature is not None:
attributes[SPANDATA.GEN_AI_REQUEST_TEMPERATURE] = temperature

top_p = inference_config.get("topP")
if top_p is not None:
attributes[SPANDATA.GEN_AI_REQUEST_TOP_P] = top_p

return attributes

def get_response_attributes(
self,
ctx: "AwsCallContext",
Expand Down
72 changes: 72 additions & 0 deletions tests/integrations/boto3/test_bedrock.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -55,6 +55,11 @@ def test_converse(
client.converse(
modelId="anthropic.claude-3-sonnet-20240229-v1:0",
messages=[{"role": "user", "content": [{"text": "Hi"}]}],
inferenceConfig={
"maxTokens": 1024,
"temperature": 0.7,
"topP": 0.9,
},
)

sentry_sdk.flush()
Expand All @@ -68,13 +73,22 @@ def test_converse(
assert span["attributes"][SPANDATA.GEN_AI_PROVIDER_NAME] == "aws.bedrock"
assert span["attributes"][SPANDATA.GEN_AI_OPERATION_NAME] == "chat"
assert span["attributes"][SPANDATA.GEN_AI_RESPONSE_STREAMING] is False

assert span["attributes"][SPANDATA.GEN_AI_REQUEST_MAX_TOKENS] == 1024
assert span["attributes"][SPANDATA.GEN_AI_REQUEST_TEMPERATURE] == 0.7
assert span["attributes"][SPANDATA.GEN_AI_REQUEST_TOP_P] == 0.9
else:
events = capture_events()

with stubber, sentry_sdk.start_transaction(name="parent"):
client.converse(
modelId="anthropic.claude-3-sonnet-20240229-v1:0",
messages=[{"role": "user", "content": [{"text": "Hi"}]}],
inferenceConfig={
"maxTokens": 1024,
"temperature": 0.7,
"topP": 0.9,
},
)

(transaction,) = events
Expand All @@ -84,6 +98,10 @@ def test_converse(
assert span["data"][SPANDATA.GEN_AI_OPERATION_NAME] == "chat"
assert span["data"][SPANDATA.GEN_AI_RESPONSE_STREAMING] is False

assert span["data"][SPANDATA.GEN_AI_REQUEST_MAX_TOKENS] == 1024
assert span["data"][SPANDATA.GEN_AI_REQUEST_TEMPERATURE] == 0.7
assert span["data"][SPANDATA.GEN_AI_REQUEST_TOP_P] == 0.9


@pytest.mark.parametrize("span_streaming", [True, False])
@pytest.mark.parametrize("stream_gen_ai_spans", [True, False])
Expand Down Expand Up @@ -127,6 +145,11 @@ def raise_client_error(**kwargs):
client.converse(
modelId="anthropic.claude-3-sonnet-20240229-v1:0",
messages=[{"role": "user", "content": [{"text": "Hi"}]}],
inferenceConfig={
"maxTokens": 1024,
"temperature": 0.7,
"topP": 0.9,
},
)

sentry_sdk.flush()
Expand All @@ -140,6 +163,10 @@ def raise_client_error(**kwargs):
assert span["attributes"][SPANDATA.GEN_AI_PROVIDER_NAME] == "aws.bedrock"
assert span["attributes"][SPANDATA.GEN_AI_OPERATION_NAME] == "chat"

assert span["attributes"][SPANDATA.GEN_AI_REQUEST_MAX_TOKENS] == 1024
assert span["attributes"][SPANDATA.GEN_AI_REQUEST_TEMPERATURE] == 0.7
assert span["attributes"][SPANDATA.GEN_AI_REQUEST_TOP_P] == 0.9

assert span["attributes"][SPANDATA.ERROR_TYPE] == "AccessDeniedException"
assert span["attributes"][SPANDATA.HTTP_STATUS_CODE] == 403
assert span["attributes"][SPANDATA.AWS_REQUEST_ID] == "test-id"
Expand All @@ -152,6 +179,11 @@ def raise_client_error(**kwargs):
client.converse(
modelId="anthropic.claude-3-sonnet-20240229-v1:0",
messages=[{"role": "user", "content": [{"text": "Hi"}]}],
inferenceConfig={
"maxTokens": 1024,
"temperature": 0.7,
"topP": 0.9,
},
)

(transaction,) = events
Expand All @@ -160,6 +192,10 @@ def raise_client_error(**kwargs):
assert span["data"][SPANDATA.GEN_AI_PROVIDER_NAME] == "aws.bedrock"
assert span["data"][SPANDATA.GEN_AI_OPERATION_NAME] == "chat"

assert span["data"][SPANDATA.GEN_AI_REQUEST_MAX_TOKENS] == 1024
assert span["data"][SPANDATA.GEN_AI_REQUEST_TEMPERATURE] == 0.7
assert span["data"][SPANDATA.GEN_AI_REQUEST_TOP_P] == 0.9

assert span["data"][SPANDATA.ERROR_TYPE] == "AccessDeniedException"
assert span["data"][SPANDATA.HTTP_STATUS_CODE] == 403
assert span["data"][SPANDATA.AWS_REQUEST_ID] == "test-id"
Expand Down Expand Up @@ -193,6 +229,11 @@ def test_converse_stream(
client.converse_stream(
modelId="anthropic.claude-3-sonnet-20240229-v1:0",
messages=[{"role": "user", "content": [{"text": "Hi"}]}],
inferenceConfig={
"maxTokens": 1024,
"temperature": 0.7,
"topP": 0.9,
},
)

sentry_sdk.flush()
Expand All @@ -206,13 +247,22 @@ def test_converse_stream(
assert span["attributes"][SPANDATA.GEN_AI_PROVIDER_NAME] == "aws.bedrock"
assert span["attributes"][SPANDATA.GEN_AI_OPERATION_NAME] == "chat"
assert span["attributes"][SPANDATA.GEN_AI_RESPONSE_STREAMING] is True

assert span["attributes"][SPANDATA.GEN_AI_REQUEST_MAX_TOKENS] == 1024
assert span["attributes"][SPANDATA.GEN_AI_REQUEST_TEMPERATURE] == 0.7
assert span["attributes"][SPANDATA.GEN_AI_REQUEST_TOP_P] == 0.9
else:
events = capture_events()

with stubber, sentry_sdk.start_transaction(name="parent"):
client.converse_stream(
modelId="anthropic.claude-3-sonnet-20240229-v1:0",
messages=[{"role": "user", "content": [{"text": "Hi"}]}],
inferenceConfig={
"maxTokens": 1024,
"temperature": 0.7,
"topP": 0.9,
},
)

(transaction,) = events
Expand All @@ -222,6 +272,10 @@ def test_converse_stream(
assert span["data"][SPANDATA.GEN_AI_OPERATION_NAME] == "chat"
assert span["data"][SPANDATA.GEN_AI_RESPONSE_STREAMING] is True

assert span["data"][SPANDATA.GEN_AI_REQUEST_MAX_TOKENS] == 1024
assert span["data"][SPANDATA.GEN_AI_REQUEST_TEMPERATURE] == 0.7
assert span["data"][SPANDATA.GEN_AI_REQUEST_TOP_P] == 0.9


@pytest.mark.parametrize("span_streaming", [True, False])
@pytest.mark.parametrize("stream_gen_ai_spans", [True, False])
Expand Down Expand Up @@ -265,6 +319,11 @@ def raise_client_error(**kwargs):
client.converse_stream(
modelId="anthropic.claude-3-sonnet-20240229-v1:0",
messages=[{"role": "user", "content": [{"text": "Hi"}]}],
inferenceConfig={
"maxTokens": 1024,
"temperature": 0.7,
"topP": 0.9,
},
)

sentry_sdk.flush()
Expand All @@ -279,6 +338,10 @@ def raise_client_error(**kwargs):
assert span["attributes"][SPANDATA.GEN_AI_OPERATION_NAME] == "chat"
assert span["attributes"][SPANDATA.GEN_AI_RESPONSE_STREAMING] is True

assert span["attributes"][SPANDATA.GEN_AI_REQUEST_MAX_TOKENS] == 1024
assert span["attributes"][SPANDATA.GEN_AI_REQUEST_TEMPERATURE] == 0.7
assert span["attributes"][SPANDATA.GEN_AI_REQUEST_TOP_P] == 0.9

assert span["attributes"][SPANDATA.ERROR_TYPE] == "AccessDeniedException"
assert span["attributes"][SPANDATA.HTTP_STATUS_CODE] == 403
assert span["attributes"][SPANDATA.AWS_REQUEST_ID] == "test-id"
Expand All @@ -291,6 +354,11 @@ def raise_client_error(**kwargs):
client.converse_stream(
modelId="anthropic.claude-3-sonnet-20240229-v1:0",
messages=[{"role": "user", "content": [{"text": "Hi"}]}],
inferenceConfig={
"maxTokens": 1024,
"temperature": 0.7,
"topP": 0.9,
},
)

(transaction,) = events
Expand All @@ -300,6 +368,10 @@ def raise_client_error(**kwargs):
assert span["data"][SPANDATA.GEN_AI_OPERATION_NAME] == "chat"
assert span["data"][SPANDATA.GEN_AI_RESPONSE_STREAMING] is True

assert span["data"][SPANDATA.GEN_AI_REQUEST_MAX_TOKENS] == 1024
assert span["data"][SPANDATA.GEN_AI_REQUEST_TEMPERATURE] == 0.7
assert span["data"][SPANDATA.GEN_AI_REQUEST_TOP_P] == 0.9

assert span["data"][SPANDATA.ERROR_TYPE] == "AccessDeniedException"
assert span["data"][SPANDATA.HTTP_STATUS_CODE] == 403
assert span["data"][SPANDATA.AWS_REQUEST_ID] == "test-id"
Expand Down
Loading