Closed pmcjr closed 1 month ago
Since the detectors are called in parallel, some detectors may return detections first before others (potentially also depending on how loaded each detector service is) leading to this "inconsistency"- we should consider if there should be sorting after all detections, such as based on start
Describe the bug
When calling
/api/v1/task/classification-with-text-generation
with multiple models present inguardrail_config.input
objects in the payload, response'stoken_classification_results.input
ordering by span are not consistent.Platform
Please provide details about the environment you are using, including the following:
Observed behavior
Request body for both calls
Response
Response 1
Response 2