Scribd
Upload
48 views3 pages

Google Professional ML Engineer Exam Guide

Professional Machine Learning Engineer

Uploaded by

aliasings
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
48 views3 pages

Google Professional ML Engineer Exam Guide

Professional Machine Learning Engineer

Uploaded by

aliasings
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Exam : Professional-Machine-

Learning-Engineer

Title : Google Professional Machine


Learning Engineer

Vendor : Google

Version : DEMO
NO.1 You work with a data engineering team that has developed a pipeline to clean your dataset
and save it in a Cloud Storage bucket. You have created an ML model and want to use the data to
refresh your model as soon as new data is available. As part of your CI/CD workflow, you want to
automatically run a Kubeflow Pipelines training job on Google Kubernetes Engine (GKE). How should
you architect this workflow?
A. Configure your pipeline with Dataflow, which saves the files in Cloud Storage After the file is
saved, start the training job on a GKE cluster
B. Use Cloud Scheduler to schedule jobs at a regular interval. For the first step of the job. check the
timestamp of objects in your Cloud Storage bucket If there are no new files since the last run, abort
the job.
C. Configure a Cloud Storage trigger to send a message to a Pub/Sub topic when a new file is
available in a storage bucket. Use a Pub/Sub-triggered Cloud Function to start the training job on a
GKE cluster
D. Use App Engine to create a lightweight python client that continuously polls Cloud Storage for new
files As soon as a file arrives, initiate the training job
Answer: C

NO.2 You are working on a Neural Network-based project. The dataset provided to you has columns
with different ranges. While preparing the data for model training, you discover that gradient
optimization is having difficulty moving weights to a good solution. What should you do?
A. Use the representation transformation (normalization) technique.
B. Change the partitioning step to reduce the dimension of the test set and have a larger training set.
C. Use feature construction to combine the strongest features.
D. Improve the data cleaning step by removing features with missing values.
Answer: D

NO.3 You recently designed and built a custom neural network that uses critical dependencies
specific to your organization's framework. You need to train the model using a managed training
service on Google Cloud. However, the ML framework and related dependencies are not supported
by Al Platform Training. Also, both your model and your data are too large to fit in memory on a
single machine. Your ML framework of choice uses the scheduler, workers, and servers distribution
structure. What should you do?
A. Build your custom container to run jobs on Al Platform Training
B. Use a built-in model available on Al Platform Training
C. Reconfigure your code to a ML framework with dependencies that are supported by Al Platform
Training
D. Build your custom containers to run distributed training jobs on Al Platform Training
Answer: D

NO.4 You work for an online retail company that is creating a visual search engine. You have set up
an end-to-end ML pipeline on Google Cloud to classify whether an image contains your company's
product. Expecting the release of new products in the near future, you configured a retraining
functionality in the pipeline so that new data can be fed into your ML models. You also want to use Al
Platform's continuous evaluation service to ensure that the models have high accuracy on your test
data set. What should you do?
A. Extend your test dataset with images of the newer products when they are introduced to
retraining
B. Update your test dataset with images of the newer products when your evaluation metrics drop
below a pre-decided threshold.
C. Replace your test dataset with images of the newer products when they are introduced to
retraining.
D. Keep the original test dataset unchanged even if newer products are incorporated into retraining
Answer: C

NO.5 You manage a team of data scientists who use a cloud-based backend system to submit
training jobs. This system has become very difficult to administer, and you want to use a managed
service instead. The data scientists you work with use many different frameworks, including Keras,
PyTorch, theano. Scikit-team, and custom libraries. What should you do?
A. Create a library of VM images on Compute Engine; and publish these images on a centralized
repository
B. Configure Kubeflow to run on Google Kubernetes Engine and receive training jobs through TFJob
C. Set up Slurm workload manager to receive jobs that can be scheduled to run on your cloud
infrastructure.
D. Use the Al Platform custom containers feature to receive training jobs using any framework
Answer: C

Common questions

Powered by AI

To architect a machine learning workflow that automatically starts a training job on Google Kubernetes Engine (GKE) when new data is available in Google Cloud Storage, you should configure a Cloud Storage trigger to send a message to a Pub/Sub topic. This message can then trigger a Cloud Function, which initiates the training job on a GKE cluster. This approach leverages Pub/Sub to detect new data and automatically start the downstream processes necessary for training the model .

A Slurm workload manager might be inadequate for handling a cloud-based machine learning setup with various frameworks because it is primarily designed for HPC environments and could struggle with the flexibility and scalability demands of diverse ML frameworks used in a cloud environment. Alternative solutions like custom containers on AI Platform are better suited for adapting to various frameworks and distributed training needs .

When experiencing difficulties with gradient optimization due to varying scales of features in a neural network project, the representation transformation technique, such as normalization, should be used. Normalization scales the feature ranges and helps in improving the convergence of the gradient descent algorithm .

When new products are introduced, a company should replace its test dataset with images of the newer products for retraining the ML models. This ensures that the models are trained and evaluated on the most current data, improving accuracy in identifying new products .

To train a custom neural network using Google's AI Platform when the ML framework and dependencies are unsupported, you should build and use custom containers for your training jobs. This allows you to package the specific dependencies and configurations required for your training setup, ensuring compatibility and resource efficiency on AI Platform .

To manage diverse machine learning frameworks in a cloud-based backend system, using the AI Platform custom containers feature is an effective strategy. This allows for receiving training jobs using any framework by encapsulating them in containers, facilitating ease of management and scalability .

You might also like