Google Professional Machine Learning Engineer

• #1

  You work at a bank, and your task is to develop a credit risk model to support loan application decisions. You've chosen to implement this model using a neural network in TensorFlow. Regulatory requirements mandate that you should be able to explain the model's predictions based on its features. Additionally, when the model is deployed, you want to continuously monitor its performance over time. To achieve this, you have opted to utilize Vertex AI for both model development and deployment. What should be your course of action?

  • AA. Utilize Vertex Explainable AI with the sampled Shapley method, and enable Vertex AI Model Monitoring to check for feature distribution drift.
  • BB. Utilize Vertex Explainable AI with the sampled Shapley method, and enable Vertex AI Model Monitoring to check for feature distribution skew.
  • CC. Utilize Vertex Explainable AI with the XRAI method, and enable Vertex AI Model Monitoring to check for feature distribution drift.
  • DD. Utilize Vertex Explainable AI with the XRAI method, and enable Vertex AI Model Monitoring to check for feature distribution skew.
  Show answer
• #2

  You are developing a training pipeline for a new XGBoost classification model based on tabular data. The data is stored in a BigQuery table. You need to complete the following steps: - Randomly split the data into training and evaluation datasets in a 65/35 ratio - Conduct feature engineering - Obtain metrics for the evaluation dataset - Compare models trained in different pipeline executions How should you execute these steps?

  • AA. 1. Using Vertex AI Pipelines, add a component to divide the data into training and evaluation sets, and add another component for feature engineering. 2.Enable autologging of metrics in the training component. 3.Compare pipeline runs in Vertex AI Experiments.
  • BB. 1. Using Vertex AI Pipelines, add a component to divide the data into training and evaluation sets, and add another component for feature engineering. 2.Enable autologging of metrics in the training component. 3.Compare models using the artifacts’ lineage in Vertex ML Metadata.
  • CC. 1. In BigQuery ML, use the CREATE MODEL statement with BOOSTED_TREE_CLASSIFIER as the model type and use BigQuery to handle the data splits. 2.Use a SQL view to apply feature engineering and train the model using the data in that view. 3.Compare the evaluation metrics of the models by using a SQL query with the ML.TRAINING_INFO statement.
  • DD. 1. In BigQuery ML, use the CREATE MODEL statement with BOOSTED_TREE_CLASSIFIER as the model type and use BigQuery to handle the data splits. 2.Use ML TRANSFORM to specify the feature engineering transformations and train the model using the data in the table. 3.Compare the evaluation metrics of the models by using a SQL query with the ML.TRAINING_INFO statement.
  Show answer
• #3

  You are employed by a toy manufacturer that has witnessed a significant surge in demand. Your task is to create an ML model to expedite the inspection process for product defects, thereby achieving quicker defect detection. There is unreliable Wi-Fi connectivity within the factory, and the company is eager to implement the new ML model promptly. Which model should you select for this purpose?

  • AA. Vertex AI AutoML Vision Edge mobile-high-accuracy-1 model
  • BB. Vertex AI AutoML Vision Edge mobile-low-latency-1 model
  • CC. Vertex AI AutoML Vision model
  • DD. Vertex AI AutoML Vision Edge mobile-versatile-1 model
  Show answer
• #4

  You are employed by a gaming company specializing in massively multiplayer online (MMO) games. You have constructed a TensorFlow model designed to forecast whether players will engage in in-app purchases exceeding $10 within the next two weeks. These predictions are intended to tailor each user's game experience. All user data is stored in BigQuery. What is the most effective approach for deploying your model to strike a balance between cost optimization, user experience enhancement, and ease of management?

  • AA. Import the model into BigQuery ML. Make predictions using batch reading data from BigQuery, and push the data to Cloud SQL
  • BB. Deploy the model to Vertex AI Prediction. Make predictions using batch reading data from Cloud Bigtable, and push the data to Cloud SQL.
  • CC. Embed the model in the mobile application. Make predictions after every in-app purchase event is published in Pub/Sub, and push the data to Cloud SQL.
  • DD. Embed the model in the streaming Dataflow pipeline. Make predictions after every in-app purchase event is published in Pub/Sub, and push the data to Cloud SQL.
  Show answer
• #5

  You are investigating the root cause of a misclassification error made by one of your models. You used Vertex AI Pipelines to train and deploy the model. The pipeline reads data from BigQuery. creates a copy of the data in Cloud Storage in TFRecord format, trains the model in Vertex AI Training on that copy, and deploys the model to a Vertex AI endpoint. You have identified the specific version of that model that misclassified, and you need to recover the data this model was trained on. How should you find that copy of the data?

  • AA. Use Vertex AI Feature Store. Modify the pipeline to use the feature store, and ensure that all training data is stored in it. Search the feature store for the data used for the training.
  • BB. Use the lineage feature of Vertex AI Metadata to find the model artifact. Determine the version of the model and identify the step that creates the data copy and search in the metadata for its location.
  • CC. Use the logging features in the Vertex AI endpoint to determine the timestamp of the model’s deployment. Find the pipeline run at that timestamp. Identify the step that creates the data copy, and search in the logs for its location.
  • DD. Find the job ID in Vertex AI Training corresponding to the training for the model. Search in the logs of that job for the data used for the training.
  Show answer
• #6

  You are training a TensorFlow model on a structured dataset with 100 billion records stored in several CSV files. You need to improve the input/output execution performance. What should you do?

  • AA. Load the data into BigQuery, and read the data from BigQuery.
  • BB. Load the data into Cloud Bigtable, and read the data from Bigtable.
  • CC. Convert the CSV files into shards of TFRecords, and store the data in Cloud Storage.
  • DD. Convert the CSV files into shards of TFRecords, and store the data in the Hadoop Distributed File System.
  Show answer
• #7

  You have a task to train a regression model using a dataset stored in BigQuery, consisting of 50,000 records. The dataset contains 20 features, a mix of categorical and numerical, and the target variable can have negative values. Your goal is to achieve high model performance while minimizing both effort and training time. What is the most suitable approach to train this regression model efficiently?

  • AA. Create a custom TensorFlow DNN model
  • BB. Use BQML XGBoost regression to train the model.
  • CC. Use Vertex AI AutoML Tables to train the model without early stopping.
  • DD. Use Vertex AI AutoML Tables to train the model with RMSLE as the optimization objective.
  Show answer
• #8

  You've recently created a custom neural network that relies on essential dependencies unique to your organization's framework. Now, you want to train this model using a managed training service in Google Cloud. However, there's a challenge: the ML framework and its related dependencies aren't compatible with AI Platform Training. Additionally, both your model and data exceed the capacity of a single machine's memory. Your preferred ML framework is designed around a distribution structure involving schedulers, workers, and servers. What steps should you take in this situation?

  • AA. Use a built-in model available on AI Platform Training.
  • BB. Build your custom container to run jobs on AI Platform Training.
  • CC. Build your custom containers to run distributed training jobs on AI Platform Training.
  • DD. Reconfigure your code to a ML framework with dependencies that are supported by AI Platform Training.
  Show answer
• #9

  You've trained a deep neural network model on Google Cloud that shows low loss on training data but underperforms on validation data, indicating overfitting. What strategy should be adopted to enhance the model's resilience against overfitting during retraining?

  • AA. Apply a dropout parameter of 0.2, and decrease the learning rate by a factor of 10.
  • BB. Apply a L2 regularization parameter of 0.4, and decrease the learning rate by a factor of 10.
  • CC. Run a hyperparameter tuning job on AI Platform to optimize for the L2 regularization and dropout parameters.
  • DD. Run a hyperparameter tuning job on AI Platform to optimize for the learning rate, and increase the number of neurons by a factor of 2.
  Show answer
• #10

  You have trained a model using data that was preprocessed in a batch Dataflow pipeline, and now you need real-time inference while ensuring consistent data preprocessing between training and serving. What should you do?

  • AA. Perform data validation to ensure that the input data to the pipeline matches the input data format for the endpoint.
  • BB. Refactor the transformation code in the batch data pipeline so that it can be used outside of the pipeline and employ the same code in the endpoint.
  • CC. Refactor the transformation code in the batch data pipeline so that it can be used outside of the pipeline, and share this code with the endpoint's end users.
  • DD. Batch the real-time requests using a time window, preprocess the batched requests using the Dataflow pipeline, and then send the preprocessed requests to the endpoint.
  Show answer
• #11

  You are currently in the process of training an object detection model utilizing a Cloud TPU v2, and you've noticed that the training duration is exceeding your initial expectations. To address this issue in a manner that is both cost-effective and efficient, what course of action should you pursue, as indicated by this simplified Cloud TPU profile trace?

  • AA. Move from Cloud TPU v2 to Cloud TPU v3 and increase batch size.
  • BB. Move from Cloud TPU v2 to 8 NVIDIA V100 GPUs and increase batch size.
  • CC. Rewrite your input function to resize and reshape the input images.
  • DD. Rewrite your input function using parallel reads, parallel processing, and prefetch.
  Show answer
• #12

  You have deployed a model on Vertex AI for real-time inference. While processing an online prediction request, you encounter an 'Out of Memory' error. What should be your course of action?

  • AA. Use batch prediction mode instead of online mode.
  • BB. Send the request again with a smaller batch of instances.
  • CC. Use base64 to encode your data before using it for prediction.
  • DD. Apply for a quota increase for the number of prediction requests.
  Show answer
• #13

  You work for a delivery company. You need to design a system that stores and manages features such as parcels delivered and truck locations over time. The system must retrieve the features with low latency and feed those features into a model for online prediction. The data science team will retrieve historical data at a specific point in time for model training. You want to store the features with minimal effort. What should you do?

  • AA. Store features in Bigtable as key/value data.
  • BB. Store features in Vertex AI Feature Store.
  • CC. Store features as a Vertex AI dataset, and use those features to train the models hosted in Vertex AI endpoints.
  • DD. Store features in BigQuery timestamp-partitioned tables, and use the BigQuery Storage Read API to serve the features.
  Show answer
• #14

  You work for a company that sells corporate electronic products to thousands of businesses worldwide. Your company stores historical customer data in BigQuery. You need to build a model that predicts customer lifetime value over the next three years. You want to use the simplest approach to build the model and you want to have access to visualization tools. What should you do?

  • AA. Create a Vertex AI Workbench notebook to perform exploratory data analysis. Use IPython magics to create a new BigQuery table with input features. Use the BigQuery console to run the CREATE MODEL statement. Validate the results by using the ML.EVALUATE and ML.PREDICT statements.
  • BB. Run the CREATE MODEL statement from the BigQuery console to create an Vertex AI AutoML model. Validate the results by using the ML.EVALUATE and ML.PREDICT statements.
  • CC. Create a Vertex AI Workbench notebook to perform exploratory data analysis and create input features. Save the features as a CSV file in Cloud Storage. Import the CSV file as a new BigQuery table. Use the BigQuery console to run the CREATE MODEL statement. Validate the results by using the ML.EVALUATE and ML.PREDICT statements.
  • DD. Create a Vertex AI Workbench notebook to perform exploratory data analysis. Use IPython magics to create a new BigQuery table with input features, create the model, and validate the results by using the CREATE MODEL, ML.EVALUATE, and ML.PREDICT statements.
  Show answer
• #15

  You have constructed a Vertex AI pipeline consisting of two key stages. The initial step involves the preprocessing of a substantial 10 TB dataset, completing this task within approximately 1 hour, and then saving the resulting data in a Cloud Storage bucket. The subsequent step utilizes this preprocessed data to train a model. Your current objective is to make adjustments to the model's code, facilitating the testing of different algorithms. Throughout this process, you aim to reduce both the pipeline's execution time and cost while keeping any alterations to the pipeline itself to a minimum. What actions should you take to meet these goals?

  • AA. Add a pipeline parameter and an additional pipeline step. Depending on the parameter value, the pipeline step conducts or skips data preprocessing, and starts model training.
  • BB. Create another pipeline without the preprocessing step, and hardcode the preprocessed Cloud Storage file location for model training.
  • CC. Configure a machine with more CPU and RAM from the compute-optimized machine family for the data preprocessing step.
  • DD. Enable caching for the pipeline job, and disable caching for the model training step.
  Show answer
• #16

  As an employee at a social media company, your task is to identify whether uploaded images feature cars. Each training sample belongs to precisely one category. Having trained an object detection neural network, you've deployed this model version to AI Platform Prediction for evaluation, also setting up an evaluation job linked to this model version. You observe that the model's precision falls short of the required business standards. What adjustments should you make to the softmax threshold in the model's final layer to improve precision?

  • AA. Increase the recall.
  • BB. Decrease the recall.
  • CC. Increase the number of false positives.
  • DD. Decrease the number of false negatives.
  Show answer
• #17

  You have received a request to construct a model using a dataset residing in a medium-sized BigQuery table, approximately 10 GB in size. Your objective is to swiftly assess the suitability of this data for model development. You intend to generate a one-time report that encompasses informative data distribution visualizations as well as advanced statistical analyses, which you will share with fellow ML engineers on your team. To achieve maximum flexibility in creating your report, what steps should you take?

  • AA. Use Vertex AI Workbench user-managed notebooks to generate the report.
  • BB. Use the Google Data Studio to create the report.
  • CC. Use the output from TensorFlow Data Validation on Dataflow to generate the report.
  • DD. Use Dataprep to create the report.
  Show answer
• #18

  You collaborate with a data engineering team that has developed a pipeline to clean the dataset and store it in a Cloud Storage bucket. You've created an ML model and aim to refresh it as soon as new data becomes available. As part of your CI/CD workflow, you intend to automate the execution of a Kubeflow Pipelines training job on a Google Kubernetes Engine (GKE) cluster. How should you design this workflow?

  • AA. Configure your pipeline with Dataflow, which saves the files in Cloud Storage. After the file is saved, start the training job on a GKE cluster.
  • BB. Use App Engine to create a lightweight python client that continuously polls Cloud Storage for new files. As soon as a file arrives, initiate the training job.
  • CC. Configure a Cloud Storage trigger to send a message to a Pub/Sub topic when a new file is available in a storage bucket. Use a Pub/Sub-triggered Cloud Function to start the training job on a GKE cluster.
  • DD. Use Cloud Scheduler to schedule jobs at a regular interval. For the first step of the job, check the timestamp of objects in your Cloud Storage bucket. If there are no new files since the last run, abort the job.
  Show answer
• #19

  You have recently used TensorFlow to train a classification model on tabular data. You have created a Dataflow pipeline that can transform several terabytes of data into training or prediction datasets consisting of TFRecords. You now need to productionize the model, and you want the predictions to be automatically uploaded to a BigQuery table on a weekly schedule. What should you do?

  • AA. Import the model into Vertex AI and deploy it to a Vertex AI endpoint. On Vertex AI Pipelines, create a pipeline that uses the DataflowPythonJobOp and the ModelBatchPredictOp components.
  • BB. Import the model into Vertex AI and deploy it to a Vertex AI endpoint. Create a Dataflow pipeline that reuses the data processing logic, sends requests to the endpoint, and then uploads predictions to a BigQuery table.
  • CC. Import the model into Vertex AI. On Vertex AI Pipelines, create a pipeline that uses the DataflowPythonJobOp and the ModelBatchPredictOp components.
  • DD. Import the model into BigQuery. Implement the data processing logic in a SQL query. On Vertex AI Pipelines, create a pipeline that uses the BigqueryQueryJobOp and the BigqueryPredictModelJobOp components.
  Show answer
• #20

  You have a dataset that is split into training, validation, and test sets. All the sets have similar distributions. You have sub-selected the most relevant features and trained a neural network in TensorFlow. TensorBoard plots show the training loss oscillating around 0.9, with the validation loss higher than the training loss by 0.3. You want to update the training regime to maximize the convergence of both losses and reduce overfitting. What should you do?

  • AA. Decrease the learning rate to fix the validation loss, and increase the number of training epochs to improve the convergence of both losses.
  • BB. Decrease the learning rate to fix the validation loss, and increase the number and dimension of the layers in the network to improve the convergence of both losses.
  • CC. Introduce L1 regularization to fix the validation loss, and increase the learning rate and the number of training epochs to improve the convergence of both losses.
  • DD. Introduce L2 regularization to fix the validation loss.
  Show answer

Want to track your score, take timed mock exams and get AI explanations? Create a free account