通过 Aident Loadout 使用 DataRobot Tools 的 Datarobot Create Batch Monitoring。执行前查看输入、输出、设置、价格和风险。
Tool to create a DataRobot Batch Monitoring job for tracking model performance and data drift on batch predictions. Use when you need to monitor predictions at scale. The job is created asynchronously and returns immediately with a job ID.
modelIdstringID of leaderboard model which is used for processing predictions dataset.
chunkSizeStrategy for determining chunk size. Can be "auto", "fixed", "dynamic", or a fixed size in bytes.
csvSettings必填objectCSV settings for intake and output files.
abortOnErrorbooleanShould this job abort if too many errors are encountered.
batchJobTypestringBatch job type enumeration
deploymentId必填stringID of deployment which is used in job for processing predictions dataset. Use LIST_DEPLOYMENTS to find available deployments.
thresholdLownumberCompute explanations for predictions below this threshold.
numConcurrentintegerNumber of simultaneous requests to run against the prediction instance.
pinnedModelIdstringSpecify a model ID used for scoring.
thresholdHighnumberCompute explanations for predictions above this threshold.
intakeSettings必填objectIntake option for the job. Must include "type" field. Example for HTTP: {"type": "http", "url": "https://example.com/test.csv"}. Example for local file: {"type": "localFile"}.
modelPackageIdstringID of model package from registry used for processing predictions dataset.
data必填Data from the action execution
errorstringError if any occurred during the execution of the action
successful必填booleanWhether or not the action execution was successful or not
API 密钥连接。风险等级 2/5。
免费操作
目前没有已发布技能明确引用此操作。