You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: clients/google-api-services-ml/v1/1.28.0/com/google/api/services/ml/v1/model/GoogleCloudMlV1RequestLoggingConfig.java
+48-16Lines changed: 48 additions & 16 deletions
Original file line number
Diff line number
Diff line change
@@ -17,10 +17,15 @@
17
17
packagecom.google.api.services.ml.v1.model;
18
18
19
19
/**
20
-
* Configurations for logging request-response pairs. Currently only BigQuery logging is supported.
21
-
* The request and response will be converted to raw string and stored within the specified BigQuery
22
-
* table. The schema is: model: STRING version: STRING time: Timestamp raw_data: STRING
23
-
* raw_prediction: STRING ground_truth: STRING
20
+
* Configuration for logging request-response pairs to a BigQuery table. Online prediction requests
21
+
* to a model version and the responses to these requests are converted to raw strings and saved to
22
+
* the specified BigQuery table. Logging is constrained by [BigQuery quotas and
23
+
* limits](/bigquery/quotas). If your project exceeds BigQuery quotas or limits, AI Platform
24
+
* Prediction does not log request-response pairs, but it continues to serve predictions.
25
+
*
26
+
* If you are using [continuous evaluation](/ml-engine/docs/continuous-evaluation/), you do not need
27
+
* to specify this configuration manually. Setting up continuous evaluation automatically enables
28
+
* logging of request-response pairs.
24
29
*
25
30
* <p> This is the Java data model class that specifies how to parse/serialize into the JSON that is
26
31
* transmitted over HTTP when working with the Cloud Machine Learning Engine. For a detailed
0 commit comments