Disabling request logging
Models log all request data by default. Disable logging if you provide personal or confidential data or any manner of sensitive information in your requests. To do this, add the x-data-logging-enabled: false option to the REST request header or gRPC call metadata. The requests you submit without logging will not be saved on Yandex Cloud servers.
To disable request logging:
When initializing a YCloudML class object, set the enable_server_data_logging parameter to False. In which case Yandex Cloud ML SDK will add x-data-logging-enabled: false to the meta information of each gRPC call.
Here is an example:
...
sdk = YCloudML(
folder_id="<folder_ID>",
auth="<API_key>",
enable_server_data_logging=False,
)
...
Where:
<folder_ID>: ID of the folder in which the service account was created.<API_key>: Service account API key required for authentication in the API. You can also use other authentication options. For more information, see Authentication in Yandex Cloud ML SDK.
Note
To use the examples, install cURL.
The example below is intended to be run in MacOS and Linux. To run it in Windows, see how to work with Bash in Microsoft Windows.
Add the x-data-logging-enabled: false header to your REST request. Here is an example:
export FOLDER_ID=<folder_ID>
export IAM_TOKEN=<IAM_token>
curl \
--request POST \
--header "Content-Type: application/json" \
--header "Authorization: Bearer ${IAM_TOKEN}" \
--header "x-data-logging-enabled: false" \
--header "x-folder-id: ${FOLDER_ID}" \
--data "@<path_to_JSON_file>" \
"<model_endpoint>"
Where:
-
FOLDER_ID: ID of the folder for which your account has the required role. -
IAM_TOKEN: IAM token used for authentication. -
@<path_to_JSON_file>: Path to the JSON file containing a request to the model. -
<model_endpoint>: Endpoint to access the model. Examples:https://llm.api.cloud.yandex.net/foundationModels/v1/completion: For synchronous requests to YandexGPT.https://llm.api.cloud.yandex.net/foundationModels/v1/completionAsync: For asynchronous requests to YandexGPT.https://llm.api.cloud.yandex.net:443/foundationModels/v1/textClassification: For fine-tuned classifiers.https://llm.api.cloud.yandex.net/foundationModels/v1/fewShotTextClassification: For zero-shot and few-shot classifiers.https://llm.api.cloud.yandex.net/foundationModels/v1/imageGenerationAsync: For YandexART.
For a full list of available endpoints, see the relevant Yandex Foundation Models API references.