You used Vertex AI Workbench user-managed notebooks to develop a TensorFlow model. The model pipeline accesses data from Cloud Storage, performs feature engineering and training locally, and outputs the trained model in Vertex AI Model Registry. The end-to-end pipeline takes 10 hours on the attached optimized instance type. You want to introduce model and data lineage for automated re-training runs for this pipeline only while minimizing the cost to run the pipeline. What should you do?
#202
Your company maintains a substantial collection of audio files from phone calls to your customer call center, stored in an on-premises database. These audio files are in wav format and have an approximate duration of 5 minutes each. Your objective is to analyze these audio files for customer sentiment, and you plan to utilize the Speech-to-Text API. Your goal is to employ the most efficient approach. What steps should you take?
#203
You are developing a model to predict potential failures in a critical machine part, utilizing a dataset that includes a multivariate time series and labels indicating part failures. You have begun to experiment with various preprocessing and modeling techniques in a Vertex AI Workbench notebook. How should you manage data logging and artifact tracking for each experiment run?
#204
You are constructing a linear regression model in BigQuery ML to estimate the probability of a customer buying your company's products. The model relies on a city name variable as a significant predictive feature. To facilitate the training and deployment of the model, your data needs to be structured in columns. You aim to prepare the data with minimal coding while retaining the crucial variables. What is the recommended approach?
#205
You have developed a model to forecast daily temperatures. Initially, you randomly divided the data, followed by transforming both the training and test datasets. While the model was trained with hourly-updated temperature data and achieved 97% accuracy in testing, its accuracy plummeted to 66% post-deployment in production. What steps can you take to enhance the accuracy of your model in the production environment?
#206
You are tasked with developing an input pipeline for a machine learning training model, which needs to process images from various sources with minimal latency. Upon discovering that your input data exceeds available memory capacity, how would you construct a dataset in line with Google's recommended best practices?
#207
You work for an auto insurance company. You are preparing a proof-of-concept ML application that uses images of damaged vehicles to infer damaged parts. Your team has assembled a set of annotated images from damage claim documents in the company’s database. The annotations associated with each image consist of a bounding box for each identified damaged part and the part name. You have been given a sufficient budget to train models on Google Cloud. You need to quickly create an initial model. What should you do?
#208
You're tasked with constructing a model to predict churn probability for customers at a retail company. It's crucial for the predictions to be interpretable, enabling the development of targeted marketing campaigns for at-risk customers. What approach should you take?
#209
You developed a Vertex AI pipeline that trains a classification model on data stored in a large BigQuery table. The pipeline has four steps, where each step is created by a Python function that uses the KubeFlow v2 API. The components have the following names: Question You launch your Vertex AI pipeline as the following: Question You perform many model iterations by adjusting the code and parameters of the training step. You observe high costs associated with the development, particularly the data export and preprocessing steps. You need to reduce model development costs. What should you do?
#210
You are constructing a machine learning model for real-time anomaly detection in sensor data. To manage incoming requests, Pub/Sub will be utilized. The goal is to store the results for subsequent analytics and visualization. How should you configure the pipeline?
#211
You work for a semiconductor manufacturing company. You need to create a real-time application that automates the quality control process. High-definition images of each semiconductor are taken at the end of the assembly line in real time. The photos are uploaded to a Cloud Storage bucket along with tabular data that includes each semiconductor’s batch number, serial number, dimensions, and weight. You need to configure model training and serving while maximizing model accuracy. What should you do?
#212
As the Director of Data Science at a sizable company, your Data Science team has recently adopted the Kubeflow Pipelines SDK for managing their training pipelines. However, your team has encountered challenges when trying to seamlessly incorporate their custom Python code into the Kubeflow Pipelines SDK environment. What guidance should you provide to expedite the integration of their code with the Kubeflow Pipelines SDK?
#213
You are employed by an international manufacturing organization that ships scientific products worldwide. These products come with instruction manuals that need translation into 15 different languages. The leadership team is interested in using machine learning to reduce the costs of manual human translations and to increase translation speed. You are tasked with implementing a scalable solution that maximizes accuracy while minimizing operational overhead. Additionally, you need to incorporate a process to evaluate and correct any incorrect translations. What should you do?
#214
You work for a retail company that is using a regression model built with BigQuery ML to predict product sales. This model is being used to serve online predictions. Recently you developed a new version of the model that uses a different architecture (custom model). Initial analysis revealed that both models are performing as expected. You want to deploy the new version of the model to production and monitor the performance over the next two months. You need to minimize the impact to the existing and future model users. How should you deploy the model?
#215
You are in the process of creating an ML model to predict house prices. During the data preparation, you encounter a crucial predictor variable, which is the distance from the nearest school. However, you notice that this variable frequently has missing values and lacks significant variance. It's important to note that every instance (row) in your dataset holds significance. How should you address the issue of missing data in this context?
#216
As an ML engineer tasked with developing training pipelines for ML models, your objective is to establish a comprehensive training pipeline for a TensorFlow model. This model will undergo training using a substantial volume of structured data, amounting to several terabytes. To ensure the pipeline's effectiveness, you aim to incorporate data quality checks before training and model quality assessments after training, all while minimizing development efforts and the necessity for infrastructure management. How should you go about constructing and orchestrating this training pipeline?
#217
You have recently developed a custom model for image classification by using a neural network. You need to automatically identify the values for learning rate, number of layers, and kernel size. To do this, you plan to run multiple jobs in parallel to identify the parameters that optimize performance. You want to minimize custom code development and infrastructure management. What should you do?
#218
Your data science team is in the process of training a PyTorch model for image classification, building upon a pre-trained ResNet model. To achieve optimal performance, you now find the need to conduct hyperparameter tuning for various parameters. What steps should you take in this scenario?
#219
You are building a TensorFlow text-to-image generative model by using a dataset that contains billions of images with their respective captions. You want to create a low maintenance, automated workflow that reads the data from a Cloud Storage bucket collects statistics, splits the dataset into training/validation/test datasets performs data transformations trains the model using the training/validation datasets, and validates the model by using the test dataset. What should you do?
#220
You are tasked with the deployment of a scikit-learn classification model into a production environment. This model must be capable of continuously serving requests around the clock, and you anticipate a high volume of requests, possibly reaching millions per second, during the operational hours from 8 am to 7 pm. Your primary objective is to keep deployment costs to a minimum. How should you proceed to achieve this?
Want to track your score, take timed mock exams and get AI explanations? Create a free account
