How to use Vertex AI for AutoML?

Get your data ready: Make sure your data is prepared and labeled appropriately for the job and data type you intend to utilize. To add inline data or reference data to Cloud Storage or BigQuery, utilize CSV or JSON files. The Data Labeling Service also allows you to obtain human annotations for your data.

Now after the preparation of data, you need to follow the below steps to train and deploy your model with Vertex AI using AutoML

Create a dataset

To start training and deployment of an AutoML model in vertexAI, first, we need to create a dataset, as example, we will discuss tabular dataset and classification autonomy model deployment in GCP(VertexAI):

Step 1 – Dashboard

Tabular data consists of many rows of information. In every row, the columns or qualities are the same. Every feature has a unique source data type that varies depending on the data source (BigQuery or a CSV file in Cloud Storage). When you use the data to train a model, Vertex AI examines the source data type and feature values to determine how the feature will be used. This refers to the modification of that trait. If necessary, a specific supported transformation can be defined for every feature.

For example, here I am using a sample dataset named “bank-marketing.csv” available online at:

gs://cloud-ml-tables-data/bank-marketing.csv

• In the Google Cloud Console, scroll through the left navigation panel and then click on Vertex AI, then choose Dashboard.

Step 2 – Create a Dataset Instance

• On the navigation menu on the left, scroll down a bit and click on Datasets.
• with the Dataset panel open, click on Create.

Step 3 – Create a Tabular form for the Dataset

• Enter the necessary details like name. I am using Structured_AutoML_Tutorial for the dataset name.
• Select the Tabular tab.
• Select the Regression/Classification objective.
• Click Create to create the dataset.

Step 4 – Import the dataset from Cloud Storage

• Click Select CSV files from Cloud Storage to choose a data source, or you may select any option where your training dataset is stored.

Since I had uploaded my prepared data to cloud storage in GCP using a bucket as we mentioned in the very first step, I am utilizing a CSV file from cloud storage in this example. after that Enter your data path (for me, it is cloud-ml-tables-data/bank-marketing.csv) in the Import file path field, then click Continue.

Vertex AI determines how it will use a feature in model training by looking at the source data type and feature values. You should check the data type of each column to ensure that it has been understood appropriately. Any feature can have a distinct supported transformation specified if necessary. Find out more about changes.

Analyze the dataset (Optional)

You may access more dataset details, such as missing or NULL values, under the analyze section. This part is unnecessary because the dataset is set up properly for this article.

Step 5 – Generating Statistics for the Dataset

• You can view the number of missing values in the dataset by clicking on Generate Statistics to view . The whole process would take roughly 10 minutes.
• To view the result and learn more about the data values of a feature, click on the feature columns.

The whole machine-learning process, from the preparation of the data through the model deployment, is automated using the AutoML technique. For users with various degrees of expertise and resources, it aims to make machine learning simpler and more efficient. Using diverse methods, such as AutoML or custom code training, and a variety of data types, such as photos, texts, or tables, you may develop and compare models using AutoML. AutoML may also assist you in tracking and explaining the behaviour and performance of your models.