You need to build classification workflows over several structured datasets currently stored in BigQuery. Because you will be performing the classification several times, you want to complete the foll

Training a TensorFlow model on Vertex AI requires significant coding expertise for model building, feature engineering, and pipeline orchestration, which goes against the requirement of "without writing code."

Vertex AutoML for Tabular Data is specifically designed for this scenario: it automates the entire ML workflow, including exploratory data analysis, feature engineering (implicitly), model architecture search, training, hyperparameter tuning, and deployment, all without requiring users to write any code for these steps. It directly integrates with BigQuery for data input.

While BigQuery ML allows for training models directly in BigQuery with SQL, it still requires SQL queries to define the model and features, and it doesn't provide comprehensive automated EDA, feature selection, or advanced hyperparameter tuning without manual intervention.

Using scikit-learn in Vertex AI Workbench notebooks with pandas requires extensive Python coding for every step of the ML workflow, directly contradicting the requirement of "without writing code."