Google Professional Machine Learning Engineer practice questions

1. Question 1 of 25

   You are building an ML model to detect anomalies in real-time sensor data. You will use Pub/Sub to handle incoming requests. You want to store the results for analytics and visualization. How should you configure the pipeline?

   • A1 = Dataflow, 2 = AI Platform, 3 = BigQuery
   • B1 = DataProc, 2 = AutoML, 3 = Cloud Bigtable
   • C1 = BigQuery, 2 = AutoML, 3 = Cloud Functions
   • D1 = BigQuery, 2 = AI Platform, 3 = Cloud Storage

   Show answer

   Correct answer: A

   How others answered

   • A100%

   Explanation

   B. 1 = DataProc, 2 = AutoML, 3 = Cloud Bigtable - wrong because bigtable is not for analytics and dataflow is preferred over dataproc. C. 1 = BigQuery, 2 = AutoML, 3 = Cloud Functions - wrong because cloud function is not a data store product D. 1 = BigQuery, 2 = AI Platform, 3 = Cloud Storage - wrong because cloud storage is not best way to store data for analytics

   Adapted from community discussion by ashii007 · 2021-12-12 · 6 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

   Supporting references

   • Google Cloud documentation
   • Google Cloud documentation
   • Google developer documentation
   • Google Cloud documentation
   • Google Cloud documentation

   View community discussion4

   Candidate context

   Question 1 discussion

   These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

   1. I

      inder0007

      2021-06-02

      12 upvotes

      Even if I follow the link, it should be dataflow, AI-Platform and Bigquery. Real answer should be A

   2. A

      ashii007

      2021-12-12

      6 upvotes

      B. 1 = DataProc, 2 = AutoML, 3 = Cloud Bigtable - wrong because bigtable is not for analytics and dataflow is preferred over dataproc. C. 1 = BigQuery, 2 = AutoML, 3 = Cloud Functions - wrong because cloud function is not a data store product D. 1 = BigQuery, 2 = AI Platform, 3 = Cloud Storage - wrong because cloud storage is not best way to store data for analytics

   3. G

      giaZ

      2022-03-15

      3 upvotes

      I think it's B, by elimination. When they tell you "minimize effort" you immediately think AutoML, but here they make you rule that possibility out by specifying on purpose "without early stopping" which goes against "minimize training time" ) and "with RMSLE..." (which goes against "can include negative values"). So you're are left with B as the best option, even though with less than 100,000 records you should usually go for pre-built models.

   4. N

      nktyagi

      2024-09-20

      Selected A2 upvotes

      To preprocess data you will use Dataflow, and then you can use the Vertex AI platform for training and serving. Since it's a recommendation use case, Cloud BigQuery is the recommended NoSQL store to manage this use case storage at scale and reduce latency.

2. Question 2 of 25

   You are designing an ML recommendation model for shoppers on your company's ecommerce website. You will use Recommendations AI to build, test, and deploy your system. How should you develop recommendations that increase revenue while following best practices?

   • AUse the ג€Other Products You May Likeג€ recommendation type to increase the click-through rate.
   • BUse the ג€Frequently Bought Togetherג€ recommendation type to increase the shopping cart size for each order.
   • CImport your user events and then your product catalog to make sure you have the highest quality event stream.
   • DBecause it will take time to collect and record product data, use placeholder values for the product catalog to test the viability of the model.

   Show answer

   Correct answer: B

   How others answered

   • B91%

   Explanation

   ANS:B https://cloud.google.com/recommendations-ai/docs/placements#fbt Frequently bought together (shopping cart expansion) The "Frequently bought together" recommendation predicts items frequently bought together for a specific product within the same shopping session. If a list of products is being viewed, then it predicts items frequently bought with that product list. This recommendation is useful when the user has indicated an intent to purchase a particular product (or list of products) already, and you are looking to recommend complements (as opposed to substitutes). This recommendation is commonly displayed on the "add to cart" page, or on the "shopping cart" or "registry" pages (for shopping cart expansion).

   Adapted from community discussion by Celia20210714 · 2021-07-18 · 7 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

   Supporting references

   • Google Cloud documentation
   • Google Cloud documentation
   • Google Cloud documentation
   • Google Cloud documentation
   • Google Cloud documentation

   View community discussion4

   Candidate context

   Question 2 discussion

   These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

   1. C

      chohan

      2021-06-15

      19 upvotes

      Answer should be B https://cloud.google.com/recommendations-ai/docs/placements#rps

   2. C

      Celia20210714

      2021-07-18

      7 upvotes

      ANS:B https://cloud.google.com/recommendations-ai/docs/placements#fbt Frequently bought together (shopping cart expansion) The "Frequently bought together" recommendation predicts items frequently bought together for a specific product within the same shopping session. If a list of products is being viewed, then it predicts items frequently bought with that product list. This recommendation is useful when the user has indicated an intent to purchase a particular product (or list of products) already, and you are looking to recommend complements (as opposed to substitutes). This recommendation is commonly displayed on the "add to cart" page, or on the "shopping cart" or "registry" pages (for shopping cart expansion).

   3. R

      ramen_lover

      2021-11-05

      4 upvotes

      I don't know the correct answer, but it seems C and D are not correct: - "Do not record user events for product items that have not been imported yet."; i.e., import your product catalog first and then your user events. - "Make sure that all required catalog information is included and correct. Do not use dummy or placeholder values." https://cloud.google.com/retail/recommendations-ai/docs/upload-catalog#catalog_import_best_practices I think the correct answer is B, because the "default optimization objective" for FBT is "revenue per order", whereas the "default optimization objective" for OYML is "click-through rate". https://cloud.google.com/retail/recommendations-ai/docs/placements#fbt

   4. G

      gcp2021go

      2021-06-07

      5 upvotes

      the correct answer should be C there is a diagram on the webpage, discuss how it works https://cloud.google.com/recommendations

3. Question 3 of 25

   You work for a biotech startup that is experimenting with deep learning ML models based on properties of biological organisms. Your team frequently works on early-stage experiments with new architectures of ML models, and writes custom TensorFlow ops in C++. You train your models on large datasets and large batch sizes. Your typical batch size has 1024 examples, and each example is about 1 MB in size. The average size of a network with all weights and embeddings is 20 GB. What hardware should you choose for your models?

   • AA cluster with 2 n1-highcpu-64 machines, each with 8 NVIDIA Tesla V100 GPUs (128 GB GPU memory in total), and a n1-highcpu-64 machine with 64 vCPUs and 58 GB RAM
   • BA cluster with 2 a2-megagpu-16g machines, each with 16 NVIDIA Tesla A100 GPUs (640 GB GPU memory in total), 96 vCPUs, and 1.4 TB RAM
   • CA cluster with an n1-highcpu-64 machine with a v2-8 TPU and 64 GB RAM
   • DA cluster with 4 n1-highcpu-96 machines, each with 96 vCPUs and 86 GB RAM

   Show answer

   Correct answer: D

   How others answered

   • D67%
   • B33%

   Only 67% agreed on this answer — treat it as contested rather than settled.

   Explanation

   “writes custom TensorFlow ops in C++” -> use CPUs when “Models that contain many custom TensorFlow operations written in C++”: https://cloud.google.com/tpu/docs/intro-to-tpu#when_to_use_tpus

   Adapted from community discussion by M25 · 2023-05-09 · 2 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

   Supporting references

   • Google Cloud documentation
   • Google Cloud documentation
   • Google Cloud documentation

   View community discussion4

   Candidate context

   Question 3 discussion

   These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

   1. A

      aw_49

      2023-05-21

      Selected D7 upvotes

      D: use CPU when models that contain many custom TensorFlow operations written in C++ https://cloud.google.com/tpu/docs/intro-to-tpu#cpus

   2. H

      hiromi

      2022-12-21

      Selected D3 upvotes

      D CPUs are recommended for TensorFlow ops written in C++ - https://cloud.google.com/tpu/docs/tensorflow-ops (Cloud TPU only supports Python)

   3. M

      M25

      2023-05-09

      Selected D2 upvotes

      “writes custom TensorFlow ops in C++” -> use CPUs when “Models that contain many custom TensorFlow operations written in C++”: https://cloud.google.com/tpu/docs/intro-to-tpu#when_to_use_tpus

   4. T

      tavva_prudhvi

      2023-07-30

      1 upvote

      While it is true that using CPUs can be more efficient when dealing with custom TensorFlow operations written in C++, it is important to consider the specific requirements of your models. In his case, we mentioned large batch sizes (1024 examples), large example sizes (1 MB each), and large network sizes (20 GB). 4 n1-highcpu-96 machines, each with 96 vCPUs and 86 GB RAM. While this configuration would provide a high number of vCPUs for custom TensorFlow operations, it lacks the GPU memory and overall RAM necessary to handle the large batch sizes and network sizes of your models.

4. Question 4 of 25

   You are working on a system log anomaly detection model for a cybersecurity organization. You have developed the model using TensorFlow, and you plan to use it for real-time prediction. You need to create a Dataflow pipeline to ingest data via Pub/Sub and write the results to BigQuery. You want to minimize the serving latency as much as possible. What should you do?

   • AContainerize the model prediction logic in Cloud Run, which is invoked by Dataflow.
   • BLoad the model directly into the Dataflow job as a dependency, and use it for prediction.
   • CDeploy the model to a Vertex AI endpoint, and invoke this endpoint in the Dataflow job.
   • DDeploy the model in a TFServing container on Google Kubernetes Engine, and invoke it in the Dataflow job.

   Show answer

   Correct answer: C

   How others answered

   • C54%
   • B46%

   Only 54% agreed on this answer — treat it as contested rather than settled.

   Explanation

   C. According Google: "Instead of deploying the model to an endpoint, you can use the RunInference API to serve machine learning models in your Apache Beam pipeline. This approach has several advantages, including flexibility and portability. However, deploying the model in Vertex AI offers many additional benefits, such as the platform's built-in tools for model monitoring, TensorBoard, and model registry governance. Vertex AI also provides the ability to use Optimized TensorFlow runtime in your endpoints. To do this, simply specify the TensorFlow runtime container when you deploy your model." https://cloud.google.com/blog/products/ai-machine-learning/streaming-prediction-with-dataflow-and-vertex

   Adapted from community discussion by guilhermebutzke · 2024-02-01 · 7 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

   Supporting references

   • Google Cloud documentation
   • Google Cloud documentation
   • Google Cloud documentation
   • Google Cloud documentation
   • Google Cloud documentation

   View community discussion4

   Candidate context

   Question 4 discussion

   These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

   1. G

      guilhermebutzke

      2024-02-01

      Selected C7 upvotes

      C. According Google: "Instead of deploying the model to an endpoint, you can use the RunInference API to serve machine learning models in your Apache Beam pipeline. This approach has several advantages, including flexibility and portability. However, deploying the model in Vertex AI offers many additional benefits, such as the platform's built-in tools for model monitoring, TensorBoard, and model registry governance. Vertex AI also provides the ability to use Optimized TensorFlow runtime in your endpoints. To do this, simply specify the TensorFlow runtime container when you deploy your model." https://cloud.google.com/blog/products/ai-machine-learning/streaming-prediction-with-dataflow-and-vertex

   2. T

      tavva_prudhvi

      2023-07-08

      Selected C3 upvotes

      In this case, the best way to minimize the serving latency of the system log anomaly detection model is to deploy it to a Vertex AI endpoint. This will allow Dataflow to invoke the model directly, without having to load it into the job as a dependency. This will significantly reduce the serving latency, as Dataflow will not have to wait for the model to load before it can make a prediction. Option B would involve loading the model directly into the Dataflow job as a dependency. This would also add an additional layer of latency, as Dataflow would have to load the model into memory before it could make a prediction.

   3. H

      hiromi

      2022-12-22

      Selected C4 upvotes

      C - https://cloud.google.com/architecture/detecting-anomalies-in-financial-transactions - https://cloud.google.com/blog/topics/financial-services/detect-anomalies-in-real-time-forex-data-with-ml

   4. T

      TNT87

      2023-03-07

      Selected C2 upvotes

      C. Deploy the model to a Vertex AI endpoint, and invoke this endpoint in the Dataflow job. The reason for this choice is that deploying the model to a Vertex AI endpoint and invoking it in the Dataflow job is the most efficient and scalable option for real-time prediction. Vertex AI provides a fully managed, serverless platform for deploying and serving machine learning models. It allows for high availability and low-latency serving of models, and can handle a large volume of requests in parallel. Invoking the model via an endpoint in the Dataflow job minimizes the latency for model prediction, as it avoids any unnecessary data transfers or containerization

5. Question 5 of 25

   You are developing an image recognition model using PyTorch based on ResNet50 architecture. Your code is working fine on your local laptop on a small subsample. Your full dataset has 200k labeled images. You want to quickly scale your training workload while minimizing cost. You plan to use 4 V100 GPUs. What should you do?

   • ACreate a Google Kubernetes Engine cluster with a node pool that has 4 V100 GPUs. Prepare and submit a TFJob operator to this node pool.
   • BCreate a Vertex AI Workbench user-managed notebooks instance with 4 V100 GPUs, and use it to train your model.
   • CPackage your code with Setuptools, and use a pre-built container. Train your model with Vertex AI using a custom tier that contains the required GPUs.
   • DConfigure a Compute Engine VM with all the dependencies that launches the training. Train your model with Vertex AI using a custom tier that contains the required GPUs.

   Show answer

   Correct answer: C

   How others answered

   • C74%

   Only 74% agreed on this answer — treat it as contested rather than settled.

   Explanation

   Custom trainer , don't overthink 1000%, this is google recommendation. you don't need Vertex AI Workbench user-managed notebooks,Google Kubernetes Engine, Compute Engine at at all , it is a waste of your effort https://cloud.google.com/vertex-ai/docs/training/configure-compute#specifying_gpus You can choose as your want

   Adapted from community discussion by John_Pongthorn · 2023-09-07 · 11 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

   Supporting references

   • Google Cloud documentation
   • Google Cloud documentation
   • Google Cloud documentation

   View community discussion4

   Candidate context

   Question 5 discussion

   These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

   1. J

      John_Pongthorn

      2023-09-07

      Selected C11 upvotes

      Custom trainer , don't overthink 1000%, this is google recommendation. you don't need Vertex AI Workbench user-managed notebooks,Google Kubernetes Engine, Compute Engine at at all , it is a waste of your effort https://cloud.google.com/vertex-ai/docs/training/configure-compute#specifying_gpus You can choose as your want

   2. F

      fitri001

      2024-10-22

      Selected C3 upvotes

      Pre-built container: Utilizing a pre-built PyTorch container image eliminates the need to manage dependencies within your container, saving time and simplifying the process. Vertex AI custom tier: Vertex AI custom tiers allow you to configure a machine type with the desired GPUs (4 V100 in this case) and pay only for the resources you use. This is more cost-effective than managing a dedicated VM instance. Setuptools packaging: Packaging your code with tools like Setuptools ensures all necessary libraries and scripts are included within the container, creating a self-contained training environment.

   3. M

      Mickey321

      2024-05-14

      Selected C1 upvote

      Using Vertex AI allows you to easily leverage multiple GPUs without managing infrastructure yourself. The custom tier gives you control to specify 4 V100 GPUs. Packaging with Setuptools and using a pre-built container ensures a consistent and portable environment with all dependencies readily available. This approach minimizes overhead and cost by relying on Vertex AI's managed service instead of setting up your own Kubernetes cluster or VMs.

   4. M

      M25

      2023-11-10

      Selected C1 upvote

      “Vertex AI provides flexible and scalable hardware and secured infrastructure to train PyTorch based deep learning models with pre-built containers and custom containers. (…) use PyTorch ResNet-50 as the example model and train it on ImageNet validation data (50K images) to measure the training performance for different training strategies”: https://cloud.google.com/blog/products/ai-machine-learning/efficient-pytorch-training-with-vertex-ai

6. Question 6 of 25

   You developed a Vertex AI ML pipeline that consists of preprocessing and training steps and each set of steps runs on a separate custom Docker image. Your organization uses GitHub and GitHub Actions as CI/CD to run unit and integration tests. You need to automate the model retraining workflow so that it can be initiated both manually and when a new version of the code is merged in the main branch. You want to minimize the steps required to build the workflow while also allowing for maximum flexibility. How should you configure the CI/CD workflow?

   • ATrigger a Cloud Build workflow to run tests, build custom Docker images, push the images to Artifact Registry, and launch the pipeline in Vertex AI Pipelines.
   • BTrigger GitHub Actions to run the tests, launch a job on Cloud Run to build custom Docker images, push the images to Artifact Registry, and launch the pipeline in Vertex AI Pipelines.
   • CTrigger GitHub Actions to run the tests, build custom Docker images, push the images to Artifact Registry, and launch the pipeline in Vertex AI Pipelines.
   • DTrigger GitHub Actions to run the tests, launch a Cloud Build workflow to build custom Docker images, push the images to Artifact Registry, and launch the pipeline in Vertex AI Pipelines.

   Show answer

   Correct answer: D

   How others answered

   • D55%
   • C39%

   Only 55% agreed on this answer — treat it as contested rather than settled.

   Explanation

   Choose D: GitHub Actions should be used to run tests and initiate the workflow upon code merges. Then, Cloud Build is a suitable service for building Docker images and handling the subsequent steps of pushing the images to Artifact Registry. So, Vertex AI Pipelines can be launched as part of the Cloud Build workflow for model retraining. In A Using Cloud Build directly from GitHub Actions would bypass GitHub Actions' capabilities for triggering and testing. In B, Cloud Run for building Docker images can introduce potential compatibility issues with Vertex AI Pipelines. In C, Skipping Cloud Build for image building limits the workflow's portability and integration with Vertex AI. https://cloud.google.com/vertex-ai/docs/pipelines/introduction https://medium.com/@cait.ray13/serving-ml-model-using-google-pub-sub-python-f569c46e7eb0

   Adapted from community discussion by guilhermebutzke · 2024-02-06 · 2 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

   Supporting references

   • Google Cloud documentation
   • Google Cloud documentation
   • Google Cloud documentation

   View community discussion4

   Candidate context

   Question 6 discussion

   These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

   1. A

      AnnaR

      2024-04-27

      Selected D4 upvotes

      Not A: does not leverage the integration capabilities of GitHub Actions with GitHub for initial testing, which is more efficient when managing repo triggers and workflows directly from Github. Not B: Cloud Run for running stateless containers, not for CI/CD tasks like building and pushing images Not C: building docker images directly in github Actions can encounter limits in terms of build performance and resource availability, esp. for complex images

   2. G

      guilhermebutzke

      2024-02-06

      Selected D2 upvotes

      Choose D: GitHub Actions should be used to run tests and initiate the workflow upon code merges. Then, Cloud Build is a suitable service for building Docker images and handling the subsequent steps of pushing the images to Artifact Registry. So, Vertex AI Pipelines can be launched as part of the Cloud Build workflow for model retraining. In A Using Cloud Build directly from GitHub Actions would bypass GitHub Actions' capabilities for triggering and testing. In B, Cloud Run for building Docker images can introduce potential compatibility issues with Vertex AI Pipelines. In C, Skipping Cloud Build for image building limits the workflow's portability and integration with Vertex AI. https://cloud.google.com/vertex-ai/docs/pipelines/introduction https://medium.com/@cait.ray13/serving-ml-model-using-google-pub-sub-python-f569c46e7eb0

   3. F

      fitri001

      2024-04-21

      Selected D1 upvote

      Security: GitHub Actions are ideal for running unit and integration tests within the controlled environment of your GitHub repository. This keeps your test code separate from the production pipeline code running in Cloud Build. Scalability and Resource Management: Cloud Build is a managed service specifically designed for building container images in Google Cloud. It offers better resource management and scalability for building Docker images compared to Cloud Run, which is primarily designed for running stateless containers. Flexibility: This configuration allows for independent scaling of test execution (in GitHub Actions) and image building (in Cloud Build). You can modify the workflow files in each platform independently without affecting the other.

   4. S

      shadz10

      2024-01-18

      Selected D3 upvotes

      D https://cloud.google.com/build/docs/building/build-containers https://cloud.google.com/build/docs/build-push-docker-image

7. Question 7 of 25

   You have developed a BigQuery ML model that predicts customer chum, and deployed the model to Vertex AI Endpoints. You want to automate the retraining of your model by using minimal additional code when model feature values change. You also want to minimize the number of times that your model is retrained to reduce training costs. What should you do?

   • A1 Enable request-response logging on Vertex AI Endpoints 2. Schedule a TensorFlow Data Validation job to monitor prediction drift 3. Execute model retraining if there is significant distance between the distributions
   • B1. Enable request-response logging on Vertex AI Endpoints 2. Schedule a TensorFlow Data Validation job to monitor training/serving skew 3. Execute model retraining if there is significant distance between the distributions
   • C1. Create a Vertex AI Model Monitoring job configured to monitor prediction drift 2. Configure alert monitoring to publish a message to a Pub/Sub queue when a monitoring alert is detected 3. Use a Cloud Function to monitor the Pub/Sub queue, and trigger retraining in BigQuery
   • D1. Create a Vertex AI Model Monitoring job configured to monitor training/serving skew 2. Configure alert monitoring to publish a message to a Pub/Sub queue when a monitoring alert is detected 3. Use a Cloud Function to monitor the Pub/Sub queue, and trigger retraining in BigQuery

   Show answer

   Correct answer: C

   How others answered

   • C50%
   • D50%

   Only 50% agreed on this answer — treat it as contested rather than settled.

   Explanation

   After reconsidering, I think it is C: - No need to use TF to enable model monitoring as stated here: https://cloud.google.com/vertex-ai/docs/model-monitoring/using-model-monitoring (even if it uses it under the hood: https://cloud.google.com/vertex-ai/docs/model-monitoring/overview#calculating-skew-and-drift) - The problem speaks about alerting of model feature changes, which happens over time, and uses a baseline of the historical production data -> prediction skew. (if the problem specified that it changes compared to training data, then it would be training-skew) (https://cloud.google.com/vertex-ai/docs/model-monitoring/monitor-explainable-ai#feature_attribution_training-serving_skew_and_prediction_drift)

   Adapted from community discussion by b1a8fae · 2024-01-15 · 3 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

   Supporting references

   • Google Cloud documentation
   • Google Cloud documentation
   • Google Cloud documentation

   View community discussion4

   Candidate context

   Question 7 discussion

   These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

   1. B

      b1a8fae

      2024-01-15

      Selected C3 upvotes

      After reconsidering, I think it is C: - No need to use TF to enable model monitoring as stated here: https://cloud.google.com/vertex-ai/docs/model-monitoring/using-model-monitoring (even if it uses it under the hood: https://cloud.google.com/vertex-ai/docs/model-monitoring/overview#calculating-skew-and-drift) - The problem speaks about alerting of model feature changes, which happens over time, and uses a baseline of the historical production data -> prediction skew. (if the problem specified that it changes compared to training data, then it would be training-skew) (https://cloud.google.com/vertex-ai/docs/model-monitoring/monitor-explainable-ai#feature_attribution_training-serving_skew_and_prediction_drift)

   2. B

      bobjr

      2024-06-06

      Selected C4 upvotes

      Skew should be detected at the beginning of the productionalisation of the model -> skew test the training data Vs the real data -> a skew indicates you trained in a dataset that is not alined with your data that you have in input Drift is used when the model works well at the beginning, but the world change and the data input changes -> drift is more long term here it is a drift issue

   3. D

      ddogg

      2024-01-31

      Selected C3 upvotes

      Option C: This option directly addresses your requirements: Vertex AI Model Monitoring: It allows efficient monitoring of prediction drift through metrics like Mean Squared Error or AUC-ROC. Pub/Sub alerts: Alert triggers notification upon significant drift, minimizing unnecessary retraining. Cloud Function: It reacts to Pub/Sub messages and triggers retraining in BigQuery using minimal additional code.

   4. 3

      36bdc1e

      2024-01-13

      2 upvotes

      C The best option for automating the retraining of your model by using minimal additional code when model feature values change, and minimizing the number of times that your model is retrained to reduce training costs, is to create a Vertex AI Model Monitoring job configured to monitor prediction drift, configure alert monitoring to publish a message to a Pub/Sub queue when a monitoring alert is detected, and use a Cloud Function to monitor the Pub/Sub queue, and trigger retraining in BigQuery. This option allows you to leverage the power and simplicity of Vertex AI, Pub/Sub, and Cloud Functions to monitor your model performance and retrain your model when needed. Vertex AI is a unified platform for building and deploying machine learning solutions on Google Cloud.

8. Question 8 of 25

   You are implementing a batch inference ML pipeline in Google Cloud. The model was developed using TensorFlow and is stored in SavedModel format in Cloud Storage. You need to apply the model to a historical dataset containing 10 TB of data that is stored in a BigQuery table. How should you perform the inference?

   • AExport the historical data to Cloud Storage in Avro format. Configure a Vertex AI batch prediction job to generate predictions for the exported data
   • BImport the TensorFlow model by using the CREATE MODEL statement in BigQuery ML. Apply the historical data to the TensorFlow model
   • CExport the historical data to Cloud Storage in CSV format. Configure a Vertex AI batch prediction job to generate predictions for the exported data
   • DConfigure a Vertex AI batch prediction job to apply the model to the historical data in BigQuery

   Show answer

   Correct answer: B

   How others answered

   • B62%
   • D24%

   Only 62% agreed on this answer — treat it as contested rather than settled.

   Explanation

   The choice is between B and D, both good BUT: Importing and making batch predictions is quite straightforward in BQ ML https://cloud.google.com/bigquery/docs/making-predictions-with-imported-tensorflow-models if not pre-processing needed on the data. If we need a more complete pipeline I'd chose D, but the tables need partitioning (100GB is the limit in Vertex AI): https://cloud.google.com/vertex-ai/docs/tabular-data/classification-regression/get-batch-predictions#input_data_requirements

   Adapted from community discussion by edoo · 2024-03-08 · 4 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

   Supporting references

   • Google Cloud documentation
   • Google Cloud documentation
   • Google Cloud documentation

   View community discussion4

   Candidate context

   Question 8 discussion

   These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

   1. E

      edoo

      2024-03-08

      Selected B4 upvotes

      The choice is between B and D, both good BUT: Importing and making batch predictions is quite straightforward in BQ ML https://cloud.google.com/bigquery/docs/making-predictions-with-imported-tensorflow-models if not pre-processing needed on the data. If we need a more complete pipeline I'd chose D, but the tables need partitioning (100GB is the limit in Vertex AI): https://cloud.google.com/vertex-ai/docs/tabular-data/classification-regression/get-batch-predictions#input_data_requirements

   2. Z

      Zwi3b3l

      2024-01-28

      Selected B4 upvotes

      Has to be B, because D has limitations: BigQuery data source tables must be no larger than 100 GB. https://cloud.google.com/vertex-ai/docs/tabular-data/classification-regression/get-batch-predictions#input_data_requirements

   3. D

      ddogg

      2024-02-01

      Selected B2 upvotes

      https://cloud.google.com/bigquery/docs/making-predictions-with-imported-tensorflow-models#:~:text=Import%20TensorFlow%20models,-To%20import%20TensorFlow&text=In%20the%20Google%20Cloud%20console%2C%20go%20to%20the%20BigQuery%20page.&text=In%20the%20query%20editor%2C%20enter,MODEL%20statement%20like%20the%20following.&text=The%20preceding%20query%20imports%20a,BigQuery%20ML%20model%20named%20imported_tf_model%20.

   4. S

      sonicclasps

      2024-01-30

      Selected B2 upvotes

      https://cloud.google.com/bigquery/docs/reference/standard-sql/bigqueryml-syntax-create-tensorflow#limitations

9. Question 9 of 25

   Your company stores a large number of audio files of phone calls made to your customer call center in an on-premises database. Each audio file is in wav format and is approximately 5 minutes long. You need to analyze these audio files for customer sentiment. You plan to use the Speech-to-Text API You want to use the most efficient approach. What should you do?

   • A1. Upload the audio files to Cloud Storage 2. Call the speech:longrunningrecognize API endpoint to generate transcriptions 3. Call the predict method of an AutoML sentiment analysis model to analyze the transcriptions.
   • B1. Upload the audio files to Cloud Storage. 2. Call the speech:longrunningrecognize API endpoint to generate transcriptions 3. Create a Cloud Function that calls the Natural Language API by using the analyzeSentiment method
   • C1. Iterate over your local files in Python 2. Use the Speech-to-Text Python library to create a speech.RecognitionAudio object, and set the content to the audio file data 3. Call the speech:recognize API endpoint to generate transcriptions 4. Call the predict method of an AutoML sentiment analysis model to analyze the transcriptions.
   • D1. Iterate over your local files in Python 2. Use the Speech-to-Text Python Library to create a speech.RecognitionAudio object and set the content to the audio file data 3. Call the speech:longrunningrecognize API endpoint to generate transcriptions. 4. Call the Natural Language API by using the analyzeSentiment method

   Show answer

   Correct answer: B

   How others answered

   • B68%
   • A32%

   Only 68% agreed on this answer — treat it as contested rather than settled.

   Explanation

   My answer: B According to https://cloud.google.com/vertex-ai/docs/text-data/sentiment-analysis/prepare-data, AutoML sentiment analysis requires a minimum of 10 labeled training documents per sentiment category, with a maximum of 100,000 total training documents. This means you need to ensure you have an adequate amount of labeled data to train a reliable model. Therefore, option B is more suitable since the API will return the sentiment and there is no mention of a customized problem that justifies the use of AutoML.

   Adapted from community discussion by guilhermebutzke · 2024-02-11 · 8 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

   Supporting references

   • Google Cloud documentation
   • Google Cloud documentation
   • Google Cloud documentation

   View community discussion4

   Candidate context

   Question 9 discussion

   These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

   1. G

      guilhermebutzke

      2024-02-11

      Selected B8 upvotes

      My answer: B According to https://cloud.google.com/vertex-ai/docs/text-data/sentiment-analysis/prepare-data, AutoML sentiment analysis requires a minimum of 10 labeled training documents per sentiment category, with a maximum of 100,000 total training documents. This means you need to ensure you have an adequate amount of labeled data to train a reliable model. Therefore, option B is more suitable since the API will return the sentiment and there is no mention of a customized problem that justifies the use of AutoML.

   2. F

      fitri001

      2024-04-21

      Selected B2 upvotes

      Scalability: Uploading audio files to Cloud Storage provides a scalable and reliable storage solution for your large dataset. Asynchronous Processing: The speech:longrunningrecognize API enables asynchronous transcription, allowing your code to proceed without waiting for each file to finish processing. This improves overall throughput. Managed Service: Cloud Functions are serverless functions that automatically scale to handle the workload. You don't need to manage servers or infrastructure. Natural Language API Integration: The Cloud Function can directly call the Natural Language API's analyzeSentiment method for sentiment analysis, streamlining the workflow.

   3. B

      BlehMaks

      2024-01-14

      Selected B4 upvotes

      Vertex AI AutoML is overkill as the build-in NL API provides sentiment analysis.

   4. S

      shadz10

      2024-01-16

      Selected B1 upvote

      I’m going with b and agree with BlehMaks - For your convenience, the Natural Language API can perform sentiment analysis directly on a file located in Cloud Storage, without the need to send the contents of the file in the body of your request. Googles best practices try api first then auto ml then custom training. https://cloud.google.com/natural-language/docs/analyzing-sentiment

10. Question 10 of 25

    You have deployed multiple versions of an image classification model on AI Platform. You want to monitor the performance of the model versions over time. How should you perform this comparison?

    • ACompare the loss performance for each model on a held-out dataset.
    • BCompare the loss performance for each model on the validation data.
    • CCompare the receiver operating characteristic (ROC) curve for each model using the What-If Tool.
    • DCompare the mean average precision across the models using the Continuous Evaluation feature.

    Show answer

    Correct answer: D

    How others answered

    • D53%
    • B27%

    Only 53% agreed on this answer — treat it as contested rather than settled.

    Explanation

    Guys, I not sure about the answer D ... And maybe you could help me in my arguments. I think choose loss to compare the model performance is better than see for metrics. For example, when can build an image model classification that has good precision metrics, because the class in unbalanced, but the loss could be terrible because of kind of loss choose that penalizes classes. so, losses are better than metrics to available models, and the answer is in A or B. I thought that the A could be the answer because I see validation as a part of the training process. So, If we want to test the model performance over time, we have to use new data, which I suppose to be the held-out data.

    Adapted from community discussion by guilhermebutzke · 2023-01-31 · 3 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Google Cloud documentation
    • Google Cloud documentation
    • Google Cloud documentation

    View community discussion4

    Candidate context

    Question 10 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

    1. D

       Danny2021

       2021-09-08

       6 upvotes

       D is correct. Choose the feature / capability GCP provides is always a good bet. :)

    2. G

       guilhermebutzke

       2023-01-31

       3 upvotes

       Guys, I not sure about the answer D ... And maybe you could help me in my arguments. I think choose loss to compare the model performance is better than see for metrics. For example, when can build an image model classification that has good precision metrics, because the class in unbalanced, but the loss could be terrible because of kind of loss choose that penalizes classes. so, losses are better than metrics to available models, and the answer is in A or B. I thought that the A could be the answer because I see validation as a part of the training process. So, If we want to test the model performance over time, we have to use new data, which I suppose to be the held-out data.

    3. S

       Sum_Sum

       2023-11-15

       Selected D4 upvotes

       D - because you are using a Google provided feature. remember in this exam its important to always choose the google services over anything else

    4. J

       Jarek7

       2023-05-09

       3 upvotes

       Please, How? B is not monitoring. It is a validation. The definition of monitoring states: "observe and check the progress or quality of (something) over a period of time" So it is a continuous process. Each option A,B,C are just one time check, not monitoring.

11. Question 11 of 25

    You work for a manufacturing company. You need to train a custom image classification model to detect product defects at the end of an assembly line. Although your model is performing well, some images in your holdout set are consistently mislabeled with high confidence. You want to use Vertex AI to understand your model’s results. What should you do?

    • AConfigure feature-based explanations by using Integrated Gradients. Set visualization type to PIXELS, and set clip_percent_upperbound to 95.
    • BCreate an index by using Vertex AI Matching Engine. Query the index with your mislabeled images.
    • CConfigure feature-based explanations by using XRAI. Set visualization type to OUTLINES, and set polarity to positive.
    • DConfigure example-based explanations. Specify the embedding output layer to be used for the latent space representation.

    Show answer

    Correct answer: D

    How others answered

    • D60%
    • A40%

    Only 60% agreed on this answer — treat it as contested rather than settled.

    Explanation

    https://cloud.google.com/vertex-ai/docs/explainable-ai/overview#example-based "Improve your data or model: One of the core use cases for example-based explanations is helping you understand why your model made certain mistakes in its predictions, and using those insights to improve your data or model. [...] For example, suppose we have a model that classifies images as either a bird or a plane, and that it is misclassifying the following bird as a plane with high confidence. You can use Example-based explanations to retrieve similar images from the training set to figure out what is happening." Not A: Integrated Gradients is recommended for low-contrast images, such as X-rays https://cloud.google.com/vertex-ai/docs/explainable-ai/overview#compare-methods Not C: Cannot set Outlines for XRAI https://cloud.google.com/ai-platform/prediction/docs/ai-explanations/visualizing-explanations

    Adapted from community discussion by eico · 2024-09-01 · 3 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Google Cloud documentation
    • Google Cloud documentation
    • Google Cloud documentation

    View community discussion4

    Candidate context

    Question 11 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

    1. G

       guilhermebutzke

       2024-02-15

       4 upvotes

       My Answer: A According to this documentation: https://cloud.google.com/vertex-ai/docs/explainable-ai/visualization-settings This option A aligns with using Integrated Gradients, which is suitable for feature-based explanations. Setting the visualization type to PIXELS allows for per-pixel attribution, which can help in understanding the specific regions of the image influencing the model's decision. Additionally, setting the clip_percent_upperbound parameter to 95 helps in filtering out noise and focusing on areas of strong attribution, which is crucial for understanding mislabeled images with high confidence. Option C suggests using XRAI for feature-based explanations and setting the visualization type to OUTLINES, along with setting the polarity to positive. However, based on the provided documentation, XRAI is recommended to have its visualization type set to PIXELS, not OUTLINES.

    2. E

       eico

       2024-09-01

       Selected D3 upvotes

       https://cloud.google.com/vertex-ai/docs/explainable-ai/overview#example-based "Improve your data or model: One of the core use cases for example-based explanations is helping you understand why your model made certain mistakes in its predictions, and using those insights to improve your data or model. [...] For example, suppose we have a model that classifies images as either a bird or a plane, and that it is misclassifying the following bird as a plane with high confidence. You can use Example-based explanations to retrieve similar images from the training set to figure out what is happening." Not A: Integrated Gradients is recommended for low-contrast images, such as X-rays https://cloud.google.com/vertex-ai/docs/explainable-ai/overview#compare-methods Not C: Cannot set Outlines for XRAI https://cloud.google.com/ai-platform/prediction/docs/ai-explanations/visualizing-explanations

    3. V

       VipinSingla

       2024-03-15

       Selected D4 upvotes

       Improve your data or model: One of the core use cases for example-based explanations is helping you understand why your model made certain mistakes in its predictions, and using those insights to improve your data or model. https://cloud.google.com/vertex-ai/docs/explainable-ai/overview

    4. Y

       YangG

       2024-10-13

       Selected D2 upvotes

       It is to understand why model is making specific mistakes, so example-based explanation makes sense to me.

12. Question 12 of 25

    You are an ML engineer at a global shoe store. You manage the ML models for the company's website. You are asked to build a model that will recommend new products to the user based on their purchase behavior and similarity with other users. What should you do?

    • ABuild a classification model
    • BBuild a knowledge-based filtering model
    • CBuild a collaborative-based filtering model
    • DBuild a regression model using the features as predictors

    Show answer

    Correct answer: C

    How others answered

    • C100%

    Explanation

    Chat gPT: Collaborative filtering models are specifically designed for recommendation systems. They work by analyzing the interactions and behaviors of users and items, then making predictions about what users will like based on similarities with other users. In this case, since you're looking at purchase behavior and user similarities, a collaborative filtering approach is well-suited to identify and recommend products that users with similar behaviors have liked or purchased. Classification models (Option A) and regression models (Option D) are generally used for different types of predictive modeling tasks, not specifically for recommendations. A knowledge-based filtering model (Option B), while useful in recommendation systems, relies more on explicit knowledge about users and items, rather than on user interaction patterns and similarities, which seems to be the focus in this scenario.

    Adapted from community discussion by Sum_Sum · 2023-11-15 · 2 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Google Cloud documentation
    • Google developer documentation
    • Google Cloud documentation

    View community discussion4

    Candidate context

    Question 12 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

    1. M

       maartenalexander

       2021-06-22

       22 upvotes

       C. Collaborative filtering is about user similarity and product recommendations. Other models won't work

    2. S

       Sum_Sum

       2023-11-15

       Selected C2 upvotes

       Chat gPT: Collaborative filtering models are specifically designed for recommendation systems. They work by analyzing the interactions and behaviors of users and items, then making predictions about what users will like based on similarities with other users. In this case, since you're looking at purchase behavior and user similarities, a collaborative filtering approach is well-suited to identify and recommend products that users with similar behaviors have liked or purchased. Classification models (Option A) and regression models (Option D) are generally used for different types of predictive modeling tasks, not specifically for recommendations. A knowledge-based filtering model (Option B), while useful in recommendation systems, relies more on explicit knowledge about users and items, rather than on user interaction patterns and similarities, which seems to be the focus in this scenario.

    3. M

       mousseUwU

       2021-10-19

       4 upvotes

       C - https://cloud.google.com/architecture/recommendations-using-machine-learning-on-compute-engine#filtering_the_data

    4. D

       DaleR

       2024-10-22

       2 upvotes

       C. Collaborative filtering is a foundational model for building a recommendation system as the input dataset is simple and the embeddings are learned for you. Matrix factorization is simply the model that applies to the collaborative filtering.

13. Question 13 of 25

    You recently used BigQuery ML to train an AutoML regression model. You shared results with your team and received positive feedback. You need to deploy your model for online prediction as quickly as possible. What should you do?

    • ARetrain the model by using BigQuery ML, and specify Vertex AI as the model registry. Deploy the model from Vertex AI Model Registry to a Vertex AI endpoint,
    • BRetrain the model by using Vertex Al Deploy the model from Vertex AI Model. Registry to a Vertex AI endpoint.
    • CAlter the model by using BigQuery ML, and specify Vertex AI as the model registry. Deploy the model from Vertex AI Model Registry to a Vertex AI endpoint.
    • DExport the model from BigQuery ML to Cloud Storage. Import the model into Vertex AI Model Registry. Deploy the model to a Vertex AI endpoint.

    Show answer

    Correct answer: C

    How others answered

    • C65%
    • D32%

    Only 65% agreed on this answer — treat it as contested rather than settled.

    Explanation

    the answer is C, no need to export the model : "You can register BigQuery ML models with the Model Registry, in order to manage them alongside your other ML models without needing to export them" https://cloud.google.com/bigquery/docs/managing-models-vertex a simple update is sufficient : https://cloud.google.com/bigquery/docs/update_vertex

    Adapted from community discussion by sonicclasps · 2024-01-31 · 4 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Google Cloud documentation
    • Google Cloud documentation
    • Google Cloud documentation
    • Google Cloud documentation
    • Google Cloud documentation

    View community discussion4

    Candidate context

    Question 13 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

    1. D

       Dagogi96

       2024-01-21

       7 upvotes

       Friend is the C, and with Alter MODEL you can register the model in Vertex AI, I work in a company and I myself have registered models like this.

    2. C

       cruise93

       2024-04-24

       Selected C6 upvotes

       https://cloud.google.com/bigquery/docs/reference/standard-sql/bigqueryml-syntax-alter-model#alter_model_statement

    3. S

       sonicclasps

       2024-01-31

       Selected C4 upvotes

       the answer is C, no need to export the model : "You can register BigQuery ML models with the Model Registry, in order to manage them alongside your other ML models without needing to export them" https://cloud.google.com/bigquery/docs/managing-models-vertex a simple update is sufficient : https://cloud.google.com/bigquery/docs/update_vertex

    4. V

       vaibavi

       2024-02-04

       5 upvotes

       I think it's C Exported models for model types AUTOML_REGRESSOR and AUTOML_CLASSIFIER do not support AI Platform deployment for online prediction.

14. Question 14 of 25

    You work at an organization that maintains a cloud-based communication platform that integrates conventional chat, voice, and video conferencing into one platform. The audio recordings are stored in Cloud Storage. All recordings have an 8 kHz sample rate and are more than one minute long. You need to implement a new feature in the platform that will automatically transcribe voice call recordings into a text for future applications, such as call summarization and sentiment analysis. How should you implement the voice call transcription feature following Google-recommended best practices?

    • AUse the original audio sampling rate, and transcribe the audio by using the Speech-to-Text API with synchronous recognition.
    • BUse the original audio sampling rate, and transcribe the audio by using the Speech-to-Text API with asynchronous recognition.
    • CUpsample the audio recordings to 16 kHz, and transcribe the audio by using the Speech-to-Text API with synchronous recognition.
    • DUpsample the audio recordings to 16 kHz, and transcribe the audio by using the Speech-to-Text API with asynchronous recognition.

    Show answer

    Correct answer: B

    How others answered

    • B64%
    • D36%

    Only 64% agreed on this answer — treat it as contested rather than settled.

    Explanation

    We have longer than minute, 8KHz recordings. https://cloud.google.com/speech-to-text/docs/best-practices-provide-speech-data "avoid re-sampling. For example, in telephony the native rate is commonly 8000 Hz, which is the rate that should be sent to the service." -> 8KHz https://cloud.google.com/speech-to-text/docs/sync-recognize "Synchronous speech recognition returns the recognized text for short audio (less than 60 seconds). To process a speech recognition request for audio longer than 60 seconds, use Asynchronous Speech Recognition." -> asynchronous So, the correct answer is B.

    Adapted from community discussion by asmgi · 2024-07-17 · 6 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Google Cloud documentation
    • Google Cloud documentation
    • Google Cloud documentation
    • Google Cloud documentation

    View community discussion4

    Candidate context

    Question 14 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

    1. A

       asmgi

       2024-07-17

       Selected B6 upvotes

       We have longer than minute, 8KHz recordings. https://cloud.google.com/speech-to-text/docs/best-practices-provide-speech-data "avoid re-sampling. For example, in telephony the native rate is commonly 8000 Hz, which is the rate that should be sent to the service." -> 8KHz https://cloud.google.com/speech-to-text/docs/sync-recognize "Synchronous speech recognition returns the recognized text for short audio (less than 60 seconds). To process a speech recognition request for audio longer than 60 seconds, use Asynchronous Speech Recognition." -> asynchronous So, the correct answer is B.

    2. P

       PhilipKoku

       2024-06-12

       Selected B4 upvotes

       B) Use original sampling rate and use asynchronous recognition... "If possible, set the sampling rate of the audio source to 16000 Hz. Otherwise, set the sample_rate_hertz to match the native sample rate of the audio source (instead of re-sampling)." https://cloud.google.com/speech-to-text/docs/best-practices-provide-speech-data#sampling_rate

    3. P

       pinimichele01

       2024-04-21

       Selected B2 upvotes

       https://cloud.google.com/speech-to-text/docs/best-practices-provide-speech-data: Capture audio with a sampling rate of 16,000 Hz or higher. Lower sampling rates may reduce accuracy. However, avoid re-sampling. For example, in telephony the native rate is commonly 8000 Hz, which is the rate that should be sent to the service. https://cloud.google.com/speech-to-text/docs/optimizing-audio-files-for-speech-to-text#sample_rate_frequency_range: It's possible to convert from one sample rate to another. However, there's no benefit to up-sampling the audio, because the frequency range information is limited by the lower sample rate and can't be recovered by converting to a higher sample rate. -----> B, not D

    4. L

       livewalk

       2024-05-22

       Selected B3 upvotes

       According to google recommandation on Sampling rate: "If possible, set the sampling rate of the audio source to 16000 Hz. Otherwise, set the sample_rate_hertz to match the native sample rate of the audio source (instead of re-sampling)." So we should match the native sample (8kHz) in the question.

15. Question 15 of 25

    Your team trained and tested a DNN regression model with good results. Six months after deployment, the model is performing poorly due to a change in the distribution of the input data. How should you address the input differences in production?

    • ACreate alerts to monitor for skew, and retrain the model.
    • BPerform feature selection on the model, and retrain the model with fewer features.
    • CRetrain the model, and select an L2 regularization parameter with a hyperparameter tuning service.
    • DPerform feature selection on the model, and retrain the model on a monthly basis with fewer features.

    Show answer

    Correct answer: A

    How others answered

    • A100%

    Explanation

    A Data values skews: These skews are significant changes in the statistical properties of data, which means that data patterns are changing, and you need to trigger a retraining of the model to capture these changes. https://developers.google.com/machine-learning/guides/rules-of-ml/#rule_37_measure_trainingserving_skew

    Adapted from community discussion by celia20200410 · 2022-01-20 · 34 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Google developer documentation
    • Google Cloud documentation
    • Google Cloud documentation

    View community discussion4

    Candidate context

    Question 15 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

    1. C

       celia20200410

       2022-01-20

       34 upvotes

       A Data values skews: These skews are significant changes in the statistical properties of data, which means that data patterns are changing, and you need to trigger a retraining of the model to capture these changes. https://developers.google.com/machine-learning/guides/rules-of-ml/#rule_37_measure_trainingserving_skew

    2. [

       [Removed]

       2021-12-02

       4 upvotes

       C. "A problem is said to be ill-posed if small changes in the given information cause large changes in the solution. This instability with respect to the data makes solutions unreliable because small measurement errors or uncertainties in parameters may be greatly magnified and lead to wildly different responses. […] The idea behind regularization is to use supplementary information to restate an ill-posed problem in a stable form." Reference: https://machinelearningmastery.com/introduction-to-regularization-to-reduce-overfitting-and-improve-generalization-error/

    3. P

       Paul_Dirac

       2021-12-25

       5 upvotes

       A Data drift doesn't necessarily require feature reselection (e.g. by L2 regularization). https://cloud.google.com/architecture/mlops-continuous-delivery-and-automation-pipelines-in-machine-learning#challenges

    4. T

       tavva_prudhvi

       2024-01-02

       Selected A2 upvotes

       When the distribution of input data changes, the model may not perform as well as it did during training. It is important to monitor the performance of the model in production and identify any changes in the distribution of input data. By creating alerts to monitor for skew, you can detect when the input data distribution has changed and take action to retrain the model using more recent data that reflects the new distribution. This will help ensure that the model continues to perform well in production.

16. Question 16 of 25

    You have trained a text classification model in TensorFlow using AI Platform. You want to use the trained model for batch predictions on text data stored in BigQuery while minimizing computational overhead. What should you do?

    • AExport the model to BigQuery ML.
    • BDeploy and version the model on AI Platform.
    • CUse Dataflow with the SavedModel to read the data from BigQuery.
    • DSubmit a batch prediction job on AI Platform that points to the model location in Cloud Storage.

    Show answer

    Correct answer: A

    How others answered

    • A55%
    • D38%

    Only 55% agreed on this answer — treat it as contested rather than settled.

    Explanation

    There are some drawbacks to option D. Cost: Submitting a batch prediction job on AI Platform is a paid service. The cost will depend on the size of the model and the amount of data that you are predicting. Complexity: Submitting a batch prediction job on AI Platform requires you to write some code. This can be a challenge if you are not familiar with AI Platform. Performance: Submitting a batch prediction job on AI Platform may not be as efficient as using BigQuery ML. This is because AI Platform needs to load the model into memory before it can run the predictions. Overall, option D is a viable option, but it may not be the best option for all situations.

    Adapted from community discussion by Darshan12 · 2023-05-17 · 2 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Google Cloud documentation
    • Google Cloud documentation
    • Google Cloud documentation
    • Google Cloud documentation

    View community discussion4

    Candidate context

    Question 16 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

    1. M

       maartenalexander

       2021-06-22

       21 upvotes

       A. You would want to minimize computational overhead–BigQuery minimizes such overhead

    2. C

       chohan

       2021-06-18

       11 upvotes

       I think it's A https://cloud.google.com/bigquery-ml/docs/making-predictions-with-imported-tensorflow-models#importing_models

    3. G

       gcp2021go

       2021-10-29

       5 upvotes

       agree. https://cloud.google.com/bigquery-ml/docs/making-predictions-with-imported-tensorflow-models

    4. D

       Darshan12

       2023-05-17

       2 upvotes

       There are some drawbacks to option D. Cost: Submitting a batch prediction job on AI Platform is a paid service. The cost will depend on the size of the model and the amount of data that you are predicting. Complexity: Submitting a batch prediction job on AI Platform requires you to write some code. This can be a challenge if you are not familiar with AI Platform. Performance: Submitting a batch prediction job on AI Platform may not be as efficient as using BigQuery ML. This is because AI Platform needs to load the model into memory before it can run the predictions. Overall, option D is a viable option, but it may not be the best option for all situations.

17. Question 17 of 25

    Your team is building a convolutional neural network (CNN)-based architecture from scratch. The preliminary experiments running on your on-premises CPU-only infrastructure were encouraging, but have slow convergence. You have been asked to speed up model training to reduce time-to-market. You want to experiment with virtual machines (VMs) on Google Cloud to leverage more powerful hardware. Your code does not include any manual device placement and has not been wrapped in Estimator model-level abstraction. Which environment should you train your model on?

    • AAVM on Compute Engine and 1 TPU with all dependencies installed manually.
    • BAVM on Compute Engine and 8 GPUs with all dependencies installed manually.
    • CA Deep Learning VM with an n1-standard-2 machine and 1 GPU with all libraries pre-installed.
    • DA Deep Learning VM with more powerful CPU e2-highcpu-16 machines with all libraries pre-installed.

    Show answer

    Correct answer: C

    How others answered

    • C57%
    • A24%
    • D19%

    Only 57% agreed on this answer — treat it as contested rather than settled.

    Explanation

    Code without manual device placement => default to CPU if TPU is present or to the lowest order GPU if multiple GPUs are present. => Not A, B. D: already using CPU and needing GPU for CNN. Ans: C

    Adapted from community discussion by Paul_Dirac · 2021-08-01 · 13 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Google Cloud documentation
    • Google Cloud documentation
    • developers.google.com documentation

    View community discussion4

    Candidate context

    Question 17 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

    1. C

       celia20200410

       2021-07-20

       16 upvotes

       ANS: C to support CNN, you should use GPU. for preliminary experiment, pre-installed pkgs/libs are good choice. https://cloud.google.com/deep-learning-vm/docs/cli#creating_an_instance_with_one_or_more_gpus https://cloud.google.com/deep-learning-vm/docs/introduction#pre-installed_packages

    2. P

       Paul_Dirac

       2021-08-01

       13 upvotes

       Code without manual device placement => default to CPU if TPU is present or to the lowest order GPU if multiple GPUs are present. => Not A, B. D: already using CPU and needing GPU for CNN. Ans: C

    3. S

       shankalman717

       2023-02-22

       3 upvotes

       Critical sentece: Your code does not include any manual device placement and has not been wrapped in Estimator model-level abstraction. So only answer we have. it's D.

    4. S

       suresh_vn

       2022-08-24

       3 upvotes

       "has not been wrapped in Estimator model-level abstraction" How you can use GPU? D in my opinion, E-family using for high CPU tasks

18. Question 18 of 25

    You recently joined a machine learning team that will soon release a new project. As a lead on the project, you are asked to determine the production readiness of the ML components. The team has already tested features and data, model development, and infrastructure. Which additional readiness check should you recommend to the team?

    • AEnsure that training is reproducible.
    • BEnsure that all hyperparameters are tuned.
    • CEnsure that model performance is monitored.
    • DEnsure that feature expectations are captured in the schema.

    Show answer

    Correct answer: C

    How others answered

    • C94%

    Explanation

    Hey! all guys A+B+D=The team has already tested features and data, model development, and infrastructure. we are about to go live with production. Monitoring readiness is the last thing to account for. It will be very rediculous if you launch model as production regardless of how we will have about monitoring. you will lauch model as production for while and will make plan to model performance monitoring later ??? you are too reckless. Pls . Read it carefully https://developers.google.com/machine-learning/testing-debugging/pipeline/production https://developers.google.com/machine-learning/testing-debugging/pipeline/overview#what-is-an-ml-pipeline. You Most guys prefer A : https://developers.google.com/machine-learning/testing-debugging/pipeline/deploying I think that it is all about model development prior to deploying .

    Adapted from community discussion by John_Pongthorn · 2023-08-16 · 4 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Google developer documentation
    • Google developer documentation
    • Google developer documentation
    • Google Cloud documentation
    • Google developer documentation

    View community discussion4

    Candidate context

    Question 18 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

    1. V

       vivid_cucumber

       2022-05-13

       7 upvotes

       To my understanding, I think A might be correct since model performance monitoring is happens "in production". but the question said the project "will soon release" which means right now is before launching, so to me testing the reproducible would make more sense. (I was confused about A and C for a long time) reference: - Testing reproducibility: https://developers.google.com/machine-learning/testing-debugging/pipeline/deploying - Testing in Production: https://developers.google.com/machine-learning/testing-debugging/pipeline/production

    2. J

       John_Pongthorn

       2023-08-16

       Selected C4 upvotes

       Hey! all guys A+B+D=The team has already tested features and data, model development, and infrastructure. we are about to go live with production. Monitoring readiness is the last thing to account for. It will be very rediculous if you launch model as production regardless of how we will have about monitoring. you will lauch model as production for while and will make plan to model performance monitoring later ??? you are too reckless. Pls . Read it carefully https://developers.google.com/machine-learning/testing-debugging/pipeline/production https://developers.google.com/machine-learning/testing-debugging/pipeline/overview#what-is-an-ml-pipeline. You Most guys prefer A : https://developers.google.com/machine-learning/testing-debugging/pipeline/deploying I think that it is all about model development prior to deploying .

    3. S

       salsabilsf

       2022-01-27

       5 upvotes

       Testing for Deploying Machine Learning Models: - Test Model Updates with Reproducible Training https://developers.google.com/machine-learning/testing-debugging/pipeline/deploying

    4. S

       sensev

       2022-01-29

       4 upvotes

       The question was not about model QA but production readiness, thus I think the answer is C because monitor model performance in production is important. As regard to A, I would I argue it could fall under "model development", since reproducible training is already important during model development.

19. Question 19 of 25

    You work for a large retailer and have been asked to segment your customers by their purchasing habits. The purchase history of all customers has been uploaded to BigQuery. You suspect that there may be several distinct customer segments, however you are unsure of how many, and you don’t yet understand the commonalities in their behavior. You want to find the most efficient solution. What should you do?

    • ACreate a k-means clustering model using BigQuery ML. Allow BigQuery to automatically optimize the number of clusters.
    • BCreate a new dataset in Dataprep that references your BigQuery table. Use Dataprep to identify similarities within each column.
    • CUse the Data Labeling Service to label each customer record in BigQuery. Train a model on your labeled data using AutoML Tables. Review the evaluation metrics to understand whether there is an underlying pattern in the data.
    • DGet a list of the customer segments from your company’s Marketing team. Use the Data Labeling Service to label each customer record in BigQuery according to the list. Analyze the distribution of labels in your dataset using Data Studio.

    Show answer

    Correct answer: A

    How others answered

    • A96%

    Explanation

    ans: A, pretty sure. C, D => discarded, very time consuming. B => yes, you can identify similarities within each column, but when i read "you don’t yet understand the commonalities in their behavior" i understand that this job would be difficult, because there could be many columns to analyze, and i don't think that this would be efficient. A => BigQuery ML is compatible with kmeans clustering, it's easy and efficient to create, and i would automatically detect the number of clusters. Also from the BigQuery ML docs: "K-means clustering for data segmentation; for example, identifying customer segments." (Source: https://cloud.google.com/bigquery-ml/docs/introduction#supported_models_in)

    Adapted from community discussion by wish0035 · 2023-06-15 · 4 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Google Cloud documentation
    • Google Cloud documentation
    • Google Cloud documentation

    View community discussion4

    Candidate context

    Question 19 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

    1. W

       wish0035

       2023-06-15

       Selected A4 upvotes

       ans: A, pretty sure. C, D => discarded, very time consuming. B => yes, you can identify similarities within each column, but when i read "you don’t yet understand the commonalities in their behavior" i understand that this job would be difficult, because there could be many columns to analyze, and i don't think that this would be efficient. A => BigQuery ML is compatible with kmeans clustering, it's easy and efficient to create, and i would automatically detect the number of clusters. Also from the BigQuery ML docs: "K-means clustering for data segmentation; for example, identifying customer segments." (Source: https://cloud.google.com/bigquery-ml/docs/introduction#supported_models_in)

    2. C

       CloudKida

       2023-11-09

       Selected A4 upvotes

       when to use k-means : Your data may contain natural groupings or clusters of data. You may want to identify these groupings descriptively in order to make data-driven decisions. For example, a retailer may want to identify natural groupings of customers who have similar purchasing habits or locations. This process is known as customer segmentation. https://cloud.google.com/bigquery/docs/kmeans-tutorial

    3. L

       LearnSodas

       2023-06-15

       Selected A3 upvotes

       K-means is a good unsupervised learning algorithm to segment a population based on similarity We can usa K-means directly in BQ, so I think it's "the most efficient way" Labeling is not a good option since we don't really know what make a customer similar to another, and why dataprep if we can use directly BQ?

    4. H

       hiromi

       2023-06-16

       Selected A3 upvotes

       A https://cloud.google.com/bigquery-ml/docs/kmeans-tutorial https://towardsdatascience.com/how-to-use-k-means-clustering-in-bigquery-ml-to-understand-and-describe-your-data-better-c972c6f5733b

20. Question 20 of 25

    You lead a data science team at a large international corporation. Most of the models your team trains are large-scale models using high-level TensorFlow APIs on AI Platform with GPUs. Your team usually takes a few weeks or months to iterate on a new version of a model. You were recently asked to review your team’s spending. How should you reduce your Google Cloud compute costs without impacting the model’s performance?

    • AUse AI Platform to run distributed training jobs with checkpoints.
    • BUse AI Platform to run distributed training jobs without checkpoints.
    • CMigrate to training with Kuberflow on Google Kubernetes Engine, and use preemptible VMs with checkpoints.
    • DMigrate to training with Kuberflow on Google Kubernetes Engine, and use preemptible VMs without checkpoints.

    Show answer

    Correct answer: C

    How others answered

    • C82%
    • A18%

    Explanation

    C is the best approach because it allows you to reduce your compute costs without impacting the model's performance. Preemptible VMs are much cheaper than standard VMs, but they can be terminated at any time. By using checkpoints, you can ensure that your training job can be resumed if a preemptible VM is terminated. Also, even if training takes days, the checkpoints will prevent lossing the progress if preemtible VM are down.

    Adapted from community discussion by joaquinmenendez · 2023-09-19 · 4 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Google Cloud documentation
    • Google Cloud documentation
    • Google Cloud documentation
    • Google Cloud documentation

    View community discussion4

    Candidate context

    Question 20 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

    1. S

       seifou

       2022-12-16

       Selected C11 upvotes

       https://cloud.google.com/blog/products/ai-machine-learning/reduce-the-costs-of-ml-workflows-with-preemptible-vms-and-gpus?hl=en

    2. J

       joaquinmenendez

       2023-09-19

       4 upvotes

       C is the best approach because it allows you to reduce your compute costs without impacting the model's performance. Preemptible VMs are much cheaper than standard VMs, but they can be terminated at any time. By using checkpoints, you can ensure that your training job can be resumed if a preemptible VM is terminated. Also, even if training takes days, the checkpoints will prevent lossing the progress if preemtible VM are down.

    3. H

       hiromi

       2022-12-16

       Selected C4 upvotes

       It's seem C - https://www.kubeflow.org/docs/distributions/gke/pipelines/preemptible/ - https://cloud.google.com/optimization/docs/guide/checkpointing

    4. M

       MultiCloudIronMan

       2024-04-01

       Selected C3 upvotes

       Pre-emptive VMs are cheaper and checkpoints will enable termination if the result is acceptable

21. Question 21 of 25

    You need to execute a batch prediction on 100 million records in a BigQuery table with a custom TensorFlow DNN regressor model, and then store the predicted results in a BigQuery table. You want to minimize the effort required to build this inference pipeline. What should you do?

    • AImport the TensorFlow model with BigQuery ML, and run the ml.predict function.
    • BUse the TensorFlow BigQuery reader to load the data, and use the BigQuery API to write the results to BigQuery.
    • CCreate a Dataflow pipeline to convert the data in BigQuery to TFRecords. Run a batch inference on Vertex AI Prediction, and write the results to BigQuery.
    • DLoad the TensorFlow SavedModel in a Dataflow pipeline. Use the BigQuery I/O connector with a custom function to perform the inference within the pipeline, and write the results to BigQuery.

    Show answer

    Correct answer: A

    How others answered

    • A78%
    • D17%

    Explanation

    for this: https://cloud.google.com/bigquery-ml/docs/reference/standard-sql/bigqueryml-syntax-inference-overview Predict the label, either a numerical value for regression tasks or a categorical value for classification tasks on DNN regresion

    Adapted from community discussion by enghabeth · 2023-08-09 · 2 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Google Cloud documentation
    • Google Cloud documentation
    • Google Cloud documentation

    View community discussion4

    Candidate context

    Question 21 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

    1. H

       hiromi

       2023-06-18

       Selected A11 upvotes

       A should work with less effort - https://cloud.google.com/bigquery-ml/docs/making-predictions-with-imported-tensorflow-models#api - https://towardsdatascience.com/how-to-do-batch-predictions-of-tensorflow-models-directly-in-bigquery-ffa843ebdba6

    2. E

       enghabeth

       2023-08-09

       Selected A2 upvotes

       for this: https://cloud.google.com/bigquery-ml/docs/reference/standard-sql/bigqueryml-syntax-inference-overview Predict the label, either a numerical value for regression tasks or a categorical value for classification tasks on DNN regresion

    3. P

       pawan94

       2024-07-09

       2 upvotes

       Simplest doesn't mean it is the most effecient/optimal. If I follow the Best practices offered by Google for Serving / Inference Pipeline I would go with Vertex AI predictions. Read More for correct details : https://cloud.google.com/architecture/ml-on-gcp-best-practices#machine-learning-development

    4. J

       JamesDoe

       2023-09-28

       Selected A2 upvotes

       https://cloud.google.com/bigquery-ml/docs/making-predictions-with-imported-tensorflow-models

22. Question 22 of 25

    Your data science team has requested a system that supports scheduled model retraining, Docker containers, and a service that supports autoscaling and monitoring for online prediction requests. Which platform components should you choose for this system?

    • AVertex AI Pipelines and App Engine
    • BVertex AI Pipelines, Vertex AI Prediction, and Vertex AI Model Monitoring
    • CCloud Composer, BigQuery ML, and Vertex AI Prediction
    • DCloud Composer, Vertex AI Training with custom containers, and App Engine

    Show answer

    Correct answer: B

    How others answered

    • B78%
    • D17%

    Explanation

    The Cloud Compose may be good consideration if you are involved in getting Google Data Engineer Cert App enging is relevant to Dev-Op Cert Pls. if you know a bit about ML Google Cloud, we are preparing to take Google ML Cert, if there is no specifically particular requirement in the question. We must emphasize on use of Vertext AI as much as possible.

    Adapted from community discussion by John_Pongthorn · 2023-07-27 · 8 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Google Cloud documentation
    • Google Cloud documentation
    • Google Cloud documentation

    View community discussion4

    Candidate context

    Question 22 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

    1. J

       John_Pongthorn

       2023-07-27

       Selected B8 upvotes

       The Cloud Compose may be good consideration if you are involved in getting Google Data Engineer Cert App enging is relevant to Dev-Op Cert Pls. if you know a bit about ML Google Cloud, we are preparing to take Google ML Cert, if there is no specifically particular requirement in the question. We must emphasize on use of Vertext AI as much as possible.

    2. R

       rosenr0

       2023-11-28

       2 upvotes

       B. Vertext AI also supports Docker container https://cloud.google.com/vertex-ai/docs/training/containers-overview

    3. H

       hiromi

       2023-06-23

       2 upvotes

       I think Vertex AI Pipeline includes schedule/trigger runs, so my vote is B

    4. M

       mil_spyro

       2023-06-20

       1 upvote

       Scheduling is not possible without the Cloud Scheduler https://cloud.google.com/vertex-ai/docs/pipelines/schedule-cloud-scheduler

23. Question 23 of 25

    You work on the data science team for a multinational beverage company. You need to develop an ML model to predict the company’s profitability for a new line of naturally flavored bottled waters in different locations. You are provided with historical data that includes product types, product sales volumes, expenses, and profits for all regions. What should you use as the input and output for your model?

    • AUse latitude, longitude, and product type as features. Use profit as model output.
    • BUse latitude, longitude, and product type as features. Use revenue and expenses as model outputs.
    • CUse product type and the feature cross of latitude with longitude, followed by binning, as features. Use profit as model output.
    • DUse product type and the feature cross of latitude with longitude, followed by binning, as features. Use revenue and expenses as model outputs.

    Show answer

    Correct answer: C

    How others answered

    • C81%

    Explanation

    Option C is the best option because it takes into account both the product type and location, which can affect profitability. Binning the feature cross of latitude and longitude can help capture the nonlinear relationship between location and profitability, and using profit as the model output is appropriate because it's the target variable we want to predict.

    Adapted from community discussion by tavva_prudhvi · 2023-09-23 · 3 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Google developer documentation
    • Google developer documentation
    • Google developer documentation

    View community discussion4

    Candidate context

    Question 23 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

    1. H

       hiromi

       2023-06-19

       Selected C7 upvotes

       C (not sure) - https://developers.google.com/machine-learning/crash-course/feature-crosses/video-lecture - https://developers.google.com/machine-learning/crash-course/regularization-for-sparsity/l1-regularization

    2. T

       tavva_prudhvi

       2023-09-23

       Selected C3 upvotes

       Option C is the best option because it takes into account both the product type and location, which can affect profitability. Binning the feature cross of latitude and longitude can help capture the nonlinear relationship between location and profitability, and using profit as the model output is appropriate because it's the target variable we want to predict.

    3. T

       taxberg

       2023-08-03

       Selected C2 upvotes

       Must be C. Always feature cross lat and lon on geographical problems. Also, D can not be right as we do not have revenue in the dataset.

    4. A

       andresvelasco

       2024-03-21

       2 upvotes

       Most people have chosen C but: Does it make sense to do binning after feature cross? Isnt it the other way around?

24. Question 24 of 25

    You work as an ML engineer at a social media company, and you are developing a visual filter for users’ profile photos. This requires you to train an ML model to detect bounding boxes around human faces. You want to use this filter in your company’s iOS-based mobile phone application. You want to minimize code development and want the model to be optimized for inference on mobile phones. What should you do?

    • ATrain a model using AutoML Vision and use the “export for Core ML” option.
    • BTrain a model using AutoML Vision and use the “export for Coral” option.
    • CTrain a model using AutoML Vision and use the “export for TensorFlow.js” option.
    • DTrain a custom TensorFlow model and convert it to TensorFlow Lite (TFLite).

    Show answer

    Correct answer: A

    How others answered

    • A77%
    • D15%

    Explanation

    Updated Vertex AI link:https://cloud.google.com/vertex-ai/docs/export/export-edge-model Trained AutoML Edge image classification models can be exported in the following formats: TF Lite - to run your model on edge or mobile devices. Edge TPU TF Lite - to run your model on Edge TPU devices. Container - to run on a Docker container. Core ML - to run your model on iOS and macOS devices. Tensorflow.js - to run your model in the browser and in Node.js.

    Adapted from community discussion by maukaba · 2024-10-20 · 5 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Google Cloud documentation
    • Google Cloud documentation
    • Google Cloud documentation
    • Google Cloud documentation

    View community discussion4

    Candidate context

    Question 24 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

    1. P

       pshemol

       2023-12-20

       Selected A16 upvotes

       https://cloud.google.com/vision/automl/docs/export-edge Core ML -> iOS and macOS Coral -> Edge TPU-based device TensorFlow.js -> web

    2. M

       maukaba

       2024-10-20

       5 upvotes

       Updated Vertex AI link:https://cloud.google.com/vertex-ai/docs/export/export-edge-model Trained AutoML Edge image classification models can be exported in the following formats: TF Lite - to run your model on edge or mobile devices. Edge TPU TF Lite - to run your model on Edge TPU devices. Container - to run on a Docker container. Core ML - to run your model on iOS and macOS devices. Tensorflow.js - to run your model in the browser and in Node.js.

    3. H

       hiromi

       2023-12-19

       Selected A2 upvotes

       A "You want to minimize code development" -> AutoML - https://cloud.google.com/vision/automl/docs/tflite-coreml-ios-tutorial - https://cloud.google.com/vertex-ai/docs/training-overview#image

    4. T

       tavva_prudhvi

       2024-03-23

       1 upvote

       Its wrong, While TFLite is a mobile-optimized version of TensorFlow, it requires more code development than using AutoML Vision and exporting for Core ML. Therefore, it's not the best option for minimizing code development time.

25. Question 25 of 25

    You are developing an ML model that uses sliced frames from video feed and creates bounding boxes around specific objects. You want to automate the following steps in your training pipeline: ingestion and preprocessing of data in Cloud Storage, followed by training and hyperparameter tuning of the object model using Vertex AI jobs, and finally deploying the model to an endpoint. You want to orchestrate the entire pipeline with minimal cluster management. What approach should you use?

    • AUse Kubeflow Pipelines on Google Kubernetes Engine.
    • BUse Vertex AI Pipelines with TensorFlow Extended (TFX) SDK.
    • CUse Vertex AI Pipelines with Kubeflow Pipelines SDK.
    • DUse Cloud Composer for the orchestration.

    Show answer

    Correct answer: C

    How others answered

    • C59%
    • B41%

    Only 59% agreed on this answer — treat it as contested rather than settled.

    Explanation

    From: https://cloud.google.com/vertex-ai/docs/pipelines/build-pipeline#sdk "1. If you use TensorFlow in an ML workflow that processes terabytes of structured data or text data, we recommend that you build your pipeline using TFX. To learn more about building a TFX pipeline, follow the TFX getting started tutorials. To learn more about using Vertex AI Pipelines to run a TFX pipeline, follow the TFX on Google Cloud tutorials. 2. For other use cases, we recommend that you build your pipeline using the Kubeflow Pipelines SDK. By building a pipeline with the Kubeflow Pipelines SDK, you can implement your workflow by building custom components or reusing prebuilt components, such as the Google Cloud Pipeline Components. Google Cloud Pipeline Components make it easier to use Vertex AI services like AutoML in your pipeline." So I guess since it is image processing, it should be Kubeflow - answer C (TFX is for structured or text data).

    Adapted from community discussion by qaz09 · 2023-02-06 · 13 community upvotes. Third-party contribution; reviewed by our quality filter, not independently verified.

    Supporting references

    • Google Cloud documentation
    • Google Cloud documentation
    • Google Cloud documentation
    • Google Cloud documentation
    • Google Cloud documentation

    View community discussion4

    Candidate context

    Question 25 discussion

    These are quality-filtered excerpts from third-party community discussion. They can be useful reasoning signals, but they are not official answers and have not been independently verified by PMLE or CertExamHQ.

    1. Q

       qaz09

       2023-02-06

       13 upvotes

       From: https://cloud.google.com/vertex-ai/docs/pipelines/build-pipeline#sdk "1. If you use TensorFlow in an ML workflow that processes terabytes of structured data or text data, we recommend that you build your pipeline using TFX. To learn more about building a TFX pipeline, follow the TFX getting started tutorials. To learn more about using Vertex AI Pipelines to run a TFX pipeline, follow the TFX on Google Cloud tutorials. 2. For other use cases, we recommend that you build your pipeline using the Kubeflow Pipelines SDK. By building a pipeline with the Kubeflow Pipelines SDK, you can implement your workflow by building custom components or reusing prebuilt components, such as the Google Cloud Pipeline Components. Google Cloud Pipeline Components make it easier to use Vertex AI services like AutoML in your pipeline." So I guess since it is image processing, it should be Kubeflow - answer C (TFX is for structured or text data).

    2. H

       hiromi

       2022-12-19

       Selected C4 upvotes

       C "If you are using other frameworks, we recommend using Kubeflow Pipeline, which is very flexible and allows you to use simple code to construct pipelines. Kubeflow Pipeline also provides Google Cloud pipeline components such as Vertex AI AutoML." (Journey to Become a Google Cloud Machine Learning Engineer: Build the mind and hand of a Google Certified ML professional)

    3. P

       pawan94

       2024-05-11

       3 upvotes

       You have to understand the ML lifecycle and difference between TFX and KFP better here. For ML end to end life cycle TFX is a better option, you can ensure Drift Detection/ Train Serve SKew with TFDV, you can easily perform serving with Tf.Serve and easily integrate TFX with Vertex AI pipeline which runs serverless. All these features are not directly available/managed in KFP ( as its user centric and user managed library). So I would go here with B.

    4. T

       tavva_prudhvi

       2023-03-25

       Selected C2 upvotes

       Vertex AI Pipelines with Kubeflow Pipelines SDK provides a high-level interface for building end-to-end machine learning pipelines. This approach allows for easy integration with Google Cloud services, including Cloud Storage for data ingestion and preprocessing, Vertex AI for training and hyperparameter tuning, and deployment to an endpoint. The Kubeflow Pipelines SDK also allows for easy orchestration of the entire pipeline, minimizing cluster management.

Keyboard:J Kmove between questionsA–Dselect an answerRshow or hide the answer← →change page

PMLE practice, question 1 of 25