Use tuning and evaluation to improve model performance

This document shows you how to create a BigQuery ML remote model that references a Vertex AI gemini-1.5-flash-002 model. You then use supervised tuning to tune the model with new training data, followed by evaluating the model with the ML.EVALUATE function.

Tuning can help you address scenarios where you need to customize the hosted Vertex AI model, such as when the expected behavior of the model is hard to concisely define in a prompt, or when prompts don't produce expected results consistently enough. Supervised tuning also influences the model in the following ways:

Guides the model to return specific response styles—for example being more concise or more verbose.
Teaches the model new behaviors—for example responding to prompts as a specific persona.
Causes the model to update itself with new information.

In this tutorial, the goal is to have the model generate text whose style and content conforms as closely as possible to provided ground truth content.

Required permissions

To create a connection, you need the following Identity and Access Management (IAM) role:
- roles/bigquery.connectionAdmin
To grant permissions to the connection's service account, you need the following permission:
- resourcemanager.projects.setIamPolicy
To create the model using BigQuery ML, you need the following IAM permissions:
- bigquery.jobs.create
- bigquery.models.create
- bigquery.models.getData
- bigquery.models.updateData
- bigquery.models.updateMetadata
To run inference, you need the following permissions:
- bigquery.tables.getData on the table
- bigquery.models.getData on the model
- bigquery.jobs.create

Before you begin

In the Google Cloud console, on the project selector page, select or create a Google Cloud project.

Note: If you don't plan to keep the resources that you create in this procedure, create a project instead of selecting an existing project. After you finish these steps, you can delete the project, removing all resources associated with the project.

Go to project selector
Make sure that billing is enabled for your Google Cloud project.
Enable the BigQuery, BigQuery Connection, Vertex AI, and Compute Engine APIs.
Enable the APIs

Costs

In this document, you use the following billable components of Google Cloud:

BigQuery: You incur costs for the queries that you run in BigQuery.
BigQuery ML: You incur costs for the model that you create and the processing that you perform in BigQuery ML.
Vertex AI: You incur costs for calls to and supervised tuning of the gemini-1.0-flash-002 model.

To generate a cost estimate based on your projected usage, use the pricing calculator. New Google Cloud users might be eligible for a free trial.

For more information, see the following resources:

Create a dataset

Create a BigQuery dataset to store your ML model:

In the Google Cloud console, go to the BigQuery page.

Go to the BigQuery page
In the Explorer pane, click your project name.
Click View actions > Create dataset.
On the Create dataset page, do the following:
- For Dataset ID, enter bqml_tutorial.
- For Location type, select Multi-region, and then select US (multiple regions in United States).
  
  The public datasets are stored in the US multi-region. For simplicity, store your dataset in the same location.
- Leave the remaining default settings as they are, and click Create dataset.

Create a connection

Create a Cloud resource connection and get the connection's service account ID. Create the connection in the same location as the dataset that you created in the previous step.

Select one of the following options:

Console

Go to the BigQuery page.

Go to BigQuery
To create a connection, click Add, and then click Connections to external data sources.
In the Connection type list, select Vertex AI remote models, remote functions and BigLake (Cloud Resource).
In the Connection ID field, enter a name for your connection.
Click Create connection.
Click Go to connection.
In the Connection info pane, copy the service account ID for use in a later step.

bq

In a command-line environment, create a connection:
```
bq mk --connection --location=REGION --project_id=PROJECT_ID \
    --connection_type=CLOUD_RESOURCE CONNECTION_ID
```
The --project_id parameter overrides the default project.

Replace the following:
- REGION: your connection region
- PROJECT_ID: your Google Cloud project ID
- CONNECTION_ID: an ID for your connection
When you create a connection resource, BigQuery creates a unique system service account and associates it with the connection.

Troubleshooting: If you get the following connection error, update the Google Cloud SDK:
```
Flags parsing error: flag --connection_type=CLOUD_RESOURCE: value should be one of...
```

Retrieve and copy the service account ID for use in a later step:

bq show --connection PROJECT_ID.REGION.CONNECTION_ID

The output is similar to the following:

name                          properties
1234.REGION.CONNECTION_ID     {"serviceAccountId": "connection-1234-9u56h9@gcp-sa-bigquery-condel.iam.gserviceaccount.com"}

Terraform

Use the google_bigquery_connection resource.

To authenticate to BigQuery, set up Application Default Credentials. For more information, see Set up authentication for client libraries.

The following example creates a Cloud resource connection named my_cloud_resource_connection in the US region:


# This queries the provider for project information.
data "google_project" "default" {}

# This creates a cloud resource connection in the US region named my_cloud_resource_connection.
# Note: The cloud resource nested object has only one output field - serviceAccountId.
resource "google_bigquery_connection" "default" {
  connection_id = "my_cloud_resource_connection"
  project       = data.google_project.default.project_id
  location      = "US"
  cloud_resource {}
}

To apply your Terraform configuration in a Google Cloud project, complete the steps in the following sections.

Prepare Cloud Shell

Launch Cloud Shell.
Set the default Google Cloud project where you want to apply your Terraform configurations.

You only need to run this command once per project, and you can run it in any directory.
```
export GOOGLE_CLOUD_PROJECT=PROJECT_ID
```
Environment variables are overridden if you set explicit values in the Terraform configuration file.

Prepare the directory

Each Terraform configuration file must have its own directory (also called a root module).

In Cloud Shell, create a directory and a new file within that directory. The filename must have the .tf extension—for example main.tf. In this tutorial, the file is referred to as main.tf.
```
mkdir DIRECTORY && cd DIRECTORY && touch main.tf
```
If you are following a tutorial, you can copy the sample code in each section or step.

Copy the sample code into the newly created main.tf.

Optionally, copy the code from GitHub. This is recommended when the Terraform snippet is part of an end-to-end solution.
Review and modify the sample parameters to apply to your environment.
Save your changes.
Initialize Terraform. You only need to do this once per directory.
```
terraform init
```
Optionally, to use the latest Google provider version, include the -upgrade option:
```
terraform init -upgrade
```

Apply the changes

Review the configuration and verify that the resources that Terraform is going to create or update match your expectations:
```
terraform plan
```
Make corrections to the configuration as necessary.
Apply the Terraform configuration by running the following command and entering yes at the prompt:
```
terraform apply
```
Wait until Terraform displays the "Apply complete!" message.
Open your Google Cloud project to view the results. In the Google Cloud console, navigate to your resources in the UI to make sure that Terraform has created or updated them.

Give the connection's service account access

Grant your service account the Vertex AI Service Agent role so that the service account can access Vertex AI. Failure to grant this role results in an error. Select one of the following options:

Console

Go to the IAM & Admin page.

Go to IAM & Admin
Click Grant access.

The Add principals dialog opens.
In the New principals field, enter the service account ID that you copied earlier.
Click Select a role.
In Filter, type Vertex AI Service Agent and then select that role.
Click Save.

gcloud

Use the gcloud projects add-iam-policy-binding command:

gcloud projects add-iam-policy-binding 'PROJECT_NUMBER' --member='serviceAccount:MEMBER' --role='roles/aiplatform.serviceAgent' --condition=None

Replace the following:

PROJECT_NUMBER: your project number
MEMBER: the service account ID that you copied earlier

The service account associated with your connection is an instance of the BigQuery Connection Delegation Service Agent, so it is acceptable to assign a service agent role to it.

Create test tables

Create tables of training and evaluation data based on the public task955_wiki_auto_style_transfer dataset from Hugging Face.

Open the Cloud Shell.

In the Cloud Shell, run the following commands to create tables of test and evaluation data:

python3 -m pip install pandas pyarrow fsspec huggingface_hub

python3 -c "import pandas as pd; df_train = pd.read_parquet('hf://datasets/Lots-of-LoRAs/task955_wiki_auto_style_transfer/data/train-00000-of-00001.parquet').drop('id', axis=1); df_train['output'] = [x[0] for x in df_train['output']]; df_train.to_json('wiki_auto_style_transfer_train.jsonl', orient='records', lines=True);"

python3 -c "import pandas as pd; df_valid = pd.read_parquet('hf://datasets/Lots-of-LoRAs/task955_wiki_auto_style_transfer/data/valid-00000-of-00001.parquet').drop('id', axis=1); df_valid['output'] = [x[0] for x in df_valid['output']]; df_valid.to_json('wiki_auto_style_transfer_valid.jsonl', orient='records', lines=True);"

bq rm -t bqml_tutorial.wiki_auto_style_transfer_train

bq rm -t bqml_tutorial.wiki_auto_style_transfer_valid

bq load --source_format=NEWLINE_DELIMITED_JSON bqml_tutorial.wiki_auto_style_transfer_train wiki_auto_style_transfer_train.jsonl input:STRING,output:STRING

bq load --source_format=NEWLINE_DELIMITED_JSON bqml_tutorial.wiki_auto_style_transfer_valid wiki_auto_style_transfer_valid.jsonl input:STRING,output:STRING

Create a baseline model

Create a remote model over the Vertex AI gemini-1.0-flash-002 model.

In the Google Cloud console, go to the BigQuery page.

Go to BigQuery
In the query editor, run the following statement to create a remote model:
```
CREATE OR REPLACE MODEL `bqml_tutorial.gemini_baseline`
REMOTE WITH CONNECTION `LOCATION.CONNECTION_ID`
OPTIONS (ENDPOINT ='gemini-1.5-flash-002');
```
Replace the following:
- LOCATION: the connection location.
- CONNECTION_ID: the ID of your BigQuery connection.
  When you view the connection details in the Google Cloud console, the CONNECTION_ID is the value in the last section of the fully qualified connection ID that is shown in Connection ID, for example projects/myproject/locations/connection_location/connections/myconnection.
The query takes several seconds to complete, after which the gemini_baseline model appears in the bqml_tutorial dataset in the Explorer pane. Because the query uses a CREATE MODEL statement to create a model, there are no query results.

Check baseline model performance

Run the ML.GENERATE_TEXT function with the remote model to see how it performs on the evaluation data without any tuning.

In the Google Cloud console, go to the BigQuery page.

Go to BigQuery

In the query editor, run the following statement:

SELECT ml_generate_text_llm_result, ground_truth
FROM
  ML.GENERATE_TEXT(
    MODEL `bqml_tutorial.gemini_baseline`,
    (
      SELECT
        input AS prompt, output AS ground_truth
      FROM `bqml_tutorial.wiki_auto_style_transfer_valid`
      LIMIT 10
    ),
    STRUCT(TRUE AS flatten_json_output));

If you examine the output data and compare the ml_generate_text_llm_result and ground_truth values, you see that while the baseline model generates text that accurately reflects the facts provided in the ground truth content, the style of the text is fairly different.

Evaluate the baseline model

To perform a more detailed evaluation of the model performance, use the ML.EVALUATE function. This function computes model metrics that measure the accuracy and quality of the generated text, in order to see how the model's responses compare to ideal esponses.

In the Google Cloud console, go to the BigQuery page.

Go to BigQuery

In the query editor, run the following statement:

SELECT *
FROM
  ML.EVALUATE(
    MODEL `bqml_tutorial.gemini_baseline`,
    (
      SELECT
        input AS input_text, output AS output_text
      FROM `bqml_tutorial.wiki_auto_style_transfer_valid`
    ),
    STRUCT('text_generation' AS task_type));

The output looks similar to the following:

   +---------------------+---------------------+-------------------------------------------+--------------------------------------------+
   | bleu4_score         | rouge-l_precision   | rouge-l_recall      | rouge-l_f1_score    | evaluation_status                          |
   +---------------------+---------------------+---------------------+---------------------+--------------------------------------------+
   | 0.15289758194680161 | 0.24925921915413246 | 0.44622484204944518 | 0.30851122211104348 | {                                          |
   |                     |                     |                     |                     |  "num_successful_rows": 176,               |
   |                     |                     |                     |                     |  "num_total_rows": 176                     |
   |                     |                     |                     |                     | }                                          |
   +---------------------+---------------------+ --------------------+---------------------+--------------------------------------------+

You can see that the baseline model perforance isn't bad, but the similarity of the generated text to the ground truth is low, based on the evaluation metrics. This indicates that it is worth performing supervised tuning to see if you can improve model performance for this use case.

Create a tuned model

Create a remote model very similar to the one you created in Create a model, but this time specifying the AS SELECT clause to provide the training data in order to tune the model.

In the Google Cloud console, go to the BigQuery page.

Go to BigQuery
In the query editor, run the following statement to create a remote model:
```
CREATE OR REPLACE MODEL `bqml_tutorial.gemini_tuned`
  REMOTE
    WITH CONNECTION `LOCATION.CONNECTION_ID`
  OPTIONS (
    endpoint = 'gemini-1.5-flash-002',
    max_iterations = 500,
    data_split_method = 'no_split')
AS
SELECT
  input AS prompt, output AS label
FROM `bqml_tutorial.wiki_auto_style_transfer_train`;
```
Replace the following:
- LOCATION: the connection location.
- CONNECTION_ID: the ID of your BigQuery connection.
  When you view the connection details in the Google Cloud console, the CONNECTION_ID is the value in the last section of the fully qualified connection ID that is shown in Connection ID, for example projects/myproject/locations/connection_location/connections/myconnection.
The query takes a few minutes to complete, after which the gemini_tuned model appears in the bqml_tutorial dataset in the Explorer pane. Because the query uses a CREATE MODEL statement to create a model, there are no query results.

Check tuned model performance

Run the ML.GENERATE_TEXT function to see how the tuned model performs on the evaluation data.

In the Google Cloud console, go to the BigQuery page.

Go to BigQuery

In the query editor, run the following statement:

SELECT ml_generate_text_llm_result, ground_truth
FROM
  ML.GENERATE_TEXT(
    MODEL `bqml_tutorial.gemini_tuned`,
    (
      SELECT
        input AS prompt, output AS ground_truth
      FROM `bqml_tutorial.wiki_auto_style_transfer_valid`
      LIMIT 10
    ),
    STRUCT(TRUE AS flatten_json_output));

If you examine the output data, you see that the tuned model produces text that is much more similar in style to the ground truth content.

Evaluate the tuned model

Use the ML.EVALUATE function to see how the tuned model's responses compare to ideal responses.

In the Google Cloud console, go to the BigQuery page.

Go to BigQuery

In the query editor, run the following statement:

SELECT *
FROM
  ML.EVALUATE(
    MODEL `bqml_tutorial.gemini_tuned`,
    (
      SELECT
        input AS prompt, output AS label
      FROM `bqml_tutorial.wiki_auto_style_transfer_valid`
    ),
    STRUCT('text_generation' AS task_type));

The output looks similar to the following:

   +---------------------+---------------------+-------------------------------------------+--------------------------------------------+
   | bleu4_score         | rouge-l_precision   | rouge-l_recall      | rouge-l_f1_score    | evaluation_status                          |
   +---------------------+---------------------+---------------------+---------------------+--------------------------------------------+
   | 0.19391708685890585 | 0.34170970869469058 | 0.46793189219384496 | 0.368190192211538   | {                                          |
   |                     |                     |                     |                     |  "num_successful_rows": 176,               |
   |                     |                     |                     |                     |  "num_total_rows": 176                     |
   |                     |                     |                     |                     | }                                          |
   +---------------------+---------------------+ --------------------+---------------------+--------------------------------------------+

You can see that even though the training dataset used only 1,408 examples, there is a marked improvement in performance as indicated by the higher evaluation metrics.

Clean up

In the Google Cloud console, go to the Manage resources page.
Go to Manage resources
In the project list, select the project that you want to delete, and then click Delete.
In the dialog, type the project ID, and then click Shut down to delete the project.