diff --git a/.scripts/bk-render-jinja2 b/.scripts/bk-render-jinja2 index 9fe8cc1..197d86a 100755 --- a/.scripts/bk-render-jinja2 +++ b/.scripts/bk-render-jinja2 @@ -74,6 +74,7 @@ def render_mdbook(content): rendered = template.render( BK_REPO=BK_REPO, MDBOOK_VIEW=True, + ON_ARGOLIS=False, PROJECT_ID="", PROJECT_NUMBER="", REGION="us-central1", @@ -90,6 +91,7 @@ def render_tutorial(content): rendered = template.render( MDBOOK_VIEW=False, + ON_ARGOLIS=(os.getenv("GCP_USERNAME") or "").endswith("altostrat.com"), PROJECT_ID=os.getenv("PROJECT_ID"), PROJECT_NUMBER=os.getenv("PROJECT_NUMBER"), REGION=os.getenv("REGION"), diff --git a/docs/labs/2_data_ingestion.md b/docs/labs/2_data_ingestion.md index 2fb8dd9..1c7c5d0 100644 --- a/docs/labs/2_data_ingestion.md +++ b/docs/labs/2_data_ingestion.md @@ -233,6 +233,11 @@ gsutil cp spark-3.3-bigquery-0.37.0.jar gs://${PROJECT_ID}-bucket/jar/spark-3.3- Open import_parquet_to_bigquery.py in the Cloud Shell editor and replace the project id with your project id `{{ PROJECT_ID }}`. Don't forget to save. +{% if ON_ARGOLIS %} +❗ You are on Argolis. The next command requires to disable the +`constraints/compute.requireOsLogin` constraint for your project in IAM. +{% endif %} + Execute it: ```bash gcloud dataproc batches submit pyspark src/data_ingestion/import_parquet_to_bigquery.py \ diff --git a/src/dataform/definitions/models/llm_model_connection.sqlx b/src/dataform/definitions/models/llm_model_connection.sqlx index 950d341..5634901 100644 --- a/src/dataform/definitions/models/llm_model_connection.sqlx +++ b/src/dataform/definitions/models/llm_model_connection.sqlx @@ -14,4 +14,4 @@ config { CREATE OR REPLACE MODEL ulb_fraud_detection_llm.ulb_fraud_detection_llm_model REMOTE WITH CONNECTION `us.llm-connection` - OPTIONS (ENDPOINT = 'gemini-pro'); + OPTIONS (ENDPOINT = 'gemini-2.0-flash');