You are employed by a bank that adheres to rigorous data governance standards. Recently, you integrated a custom model designed to identify fraudulent transactions. Your intention is to configure your training code to access internal data through an API endpoint hosted within your project's network. Your primary concerns are to ensure the utmost security in accessing this data and to minimize the potential risk of data exfiltration. What steps should you take to achieve these objectives?
#242
You are in the process of training a machine learning model utilizing a dataset stored in BigQuery, and this dataset contains numerous values classified as Personally Identifiable Information (PII). Your objective is to decrease the dataset's sensitivity prior to commencing model training, and it's essential to retain all columns in the dataset as they are crucial for your model. What steps should you take in this situation?
#243
You're in charge of a data science team within a large international corporation. Your team primarily develops large-scale models using high-level TensorFlow APIs on AI Platform with GPUs. The typical iteration time for a new model version ranges from a few weeks to several months. Recently, there has been a request to assess and reduce your team's Google Cloud compute costs while ensuring that the model's performance remains unaffected. How can you achieve this cost reduction without compromising the model's quality?
#244
You have developed an AutoML tabular classification model to identify high-value customers engaging with your organization's website. The next step is deploying this model to a Vertex AI endpoint integrated with your website application. Since traffic is expected to increase during nights and weekends, you must configure the deployment settings to ensure low latency and cost efficiency. What configuration should you use?
#245
You are eager to train an AutoML model for predicting house prices using a compact public dataset stored in BigQuery. Your primary objective is to prepare the data efficiently, opting for the simplest and most straightforward approach. What step should you take to achieve this goal?
#246
You are tasked with creating a unified analytics environment that spans across various on-premises data marts. The company faces data quality and security issues during data integration across servers, stemming from the use of diverse, disconnected tools and makeshift solutions. The goal is to adopt a fully managed, cloud-native data integration service that reduces overall workload costs and minimizes repetitive tasks. Additionally, some team members favor a codeless interface for constructing Extract, Transform, Load (ETL) processes. Which service would best meet these requirements?
#247
You recently joined an enterprise-scale company that has thousands of datasets. You know that there are accurate descriptions for each table in BigQuery, and you are searching for the proper BigQuery table to use for a model you are building on AI Platform. How should you find the data that you need?
#248
You are developing a model to identify traffic signs in images extracted from videos taken from the dashboard of a vehicle. You have a dataset of 100,000 images that were cropped to show one out of ten different traffic signs. The images have been labeled accordingly for model training, and are stored in a Cloud Storage bucket. You need to be able to tune the model during each training run. How should you train the model?
#249
You've developed a custom ML model using scikit-learn, and you've encountered longer-than-expected training times. To address this issue, you've decided to migrate your model to Vertex AI Training, and you aim to enhance the model's training efficiency. What should be your initial approach to achieve this goal?
#250
You have created a Vertex AI pipeline that automates custom model training. You want to add a pipeline component that enables your team to collaborate most easily when running different executions and comparing metrics both visually and programmatically. What should you do?
#251
You work for a large technology company aiming to modernize its contact center operations. Your task is to develop a solution for classifying incoming calls by product, enabling faster routing to the appropriate support team. The calls have already been transcribed using the Speech-to-Text API. You aim to minimize data preprocessing and development time. How should you proceed to build the model?
#252
Your team is training a large number of ML models that use different algorithms, parameters, and datasets. Some models are trained in Vertex AI Pipelines, and some are trained on Vertex AI Workbench notebook instances. Your team wants to compare the performance of the models across both services. You want to minimize the effort required to store the parameters and metrics. What should you do?
#253
You are employed at a bank and tasked with developing a random forest model for fraud detection. The dataset at your disposal contains transactions, with only 1% of them being flagged as fraudulent. What data transformation strategy should you consider to enhance the classifier's performance?
#254
During the exploratory data analysis of a dataset, you've identified a crucial categorical feature with a 5% incidence of missing values. To mitigate potential bias stemming from these gaps in the data, what would be your recommended approach for handling these missing values?
#255
You are collaborating on a model prototype with your team. You need to create a Vertex AI Workbench environment for the members of your team and also limit access to other employees in your project. What should you do?
#256
You're creating machine learning models for CT scan image segmentation using AI Platform and regularly update the architectures to align with the latest research. To benchmark their performance, you need to retrain these models using the same dataset. Your goal is to reduce computational expenses and manual effort, while maintaining version control for your code. What steps should you take?
#2572টি বাছুন
You have a fully operational end-to-end ML pipeline that includes hyperparameter tuning of your ML model using AI Platform. However, the hyperparameter tuning process is taking longer than anticipated, causing delays in the downstream processes. You aim to expedite the tuning job without significantly sacrificing its effectiveness. What actions should you consider? (Choose two options)
#258
You require a rapid solution for constructing and training a model that can predict the sentiment of customer reviews, utilizing custom categories, all without the need for manual coding. However, your dataset is insufficient to train a model entirely from the ground up. The primary objective is to achieve a high level of predictive accuracy. In light of these considerations, which service should you opt for?
#259
You are employed by an online retail company developing a visual search engine. You have established an end-to-end ML pipeline on Google Cloud to determine whether an image contains your company's product. Anticipating the release of new products soon, you have incorporated a retraining functionality in the pipeline to accommodate new data for your ML models. Additionally, you aim to utilize AI Platform's continuous evaluation service to maintain high accuracy on your test dataset. What steps should you take?
#260
You have successfully trained a DNN regressor using TensorFlow to predict housing prices, utilizing a set of predictive features. The default precision for your model is tf.float64, and you've employed a standard TensorFlow estimator with the following configuration: estimator = tf.estimator.DNNRegressor( feature_columns=[YOUR_LIST_OF_FEATURES], hidden_units=[1024, 512, 256], dropout=None ) Your model's performance is satisfactory; however, as you prepare to deploy it into production, you notice that your current serving latency on CPUs is 10ms at the 90th percentile. Your production requirements dictate a model latency of 8ms at the 90th percentile, and you are open to a slight decrease in prediction performance to meet this latency requirement. To achieve this, what should be your initial approach to quickly reduce the serving latency?
