PMLE Scaling Prototypes into ML Models Practice Test 7 — 15 Questions

Question 1

Your team is deploying a large model on edge devices and needs to reduce its size by 80% while maintaining reasonable accuracy. Which THREE techniques should they consider? (Choose 3.)

Accepted Answer

Quantisation to INT8. Quantisation to INT8 reduces the precision of model weights and activations from 32-bit floating point to 8-bit integers, cutting memory usage by approximately 75% (4x compression). This directly addresses the 80% size reduction target while often preserving accuracy within 1-2% through careful calibration and scaling, making it a primary technique for edge deployment.

Answer

Transfer learning from a larger model

Answer

Increasing model capacity with more layers

Question 2

A data scientist wants to train a PyTorch model on Vertex AI using a pre-built container for GPU training. She needs to use 4 NVIDIA A100 GPUs on a single machine. Which machine configuration should she select?

Accepted Answer

a2-highgpu-4g (4 A100 GPUs). The a2-highgpu-4g machine type is purpose-built for GPU workloads and provides exactly 4 NVIDIA A100 GPUs on a single machine, matching the requirement precisely. Vertex AI supports this machine type for custom training with pre-built containers, so the data scientist can select it directly in the training configuration. Choosing it avoids over-provisioning or mismatching GPU counts.

Answer

n1-highmem-16 with 4 NVIDIA V100 GPUs

Answer

n1-standard-16 with 4 NVIDIA T4 GPUs

Answer

a2-megagpu-16g (16 A100 GPUs)

Question 3

A team is building a feature pipeline for an ML model. They need to compute aggregate features over a sliding time window from streaming data. Which Google Cloud service is most appropriate for this task?

Accepted Answer

Cloud Dataflow with sliding windows. Cloud Dataflow, based on Apache Beam, natively supports sliding time windows, making it the appropriate service for computing aggregate features over overlapping time intervals from streaming data. Fixed windows alone would not satisfy the sliding-window requirement, and the other services lack native sliding window aggregation capabilities.

Answer

Cloud Pub/Sub for windowing logic

Answer

BigQuery scheduled queries

Answer

Cloud Functions with Pub/Sub triggers

Question 4

A machine learning engineer is training a TensorFlow model on Vertex AI using distributed training with the MultiWorkerMirroredStrategy. The training job uses 4 workers with 4 GPUs each. The engineer notices that the training is not scaling linearly. What is the most likely cause?

Accepted Answer

Communication overhead due to gradient synchronization. With MultiWorkerMirroredStrategy, each worker computes gradients independently on its local batch, then all-reduces gradients across workers via collective communication (e.g., NCCL or gRPC). As the number of workers increases, the communication overhead for gradient synchronization grows, often dominating the per-step time and preventing linear scaling. This is the most common bottleneck in distributed TensorFlow training, especially with many workers or small batch sizes per worker.

Answer

The model architecture is too simple to benefit from distribution

Answer

The workers are not using the same version of TensorFlow

Answer

The GPUs are not configured correctly

Question 5

A machine learning engineer is deploying a TensorFlow model on an edge device with limited memory and compute. The model needs to perform inference with low latency. The engineer has a trained float32 model. Which model compression technique should be applied first to reduce the model size and improve inference speed without significant accuracy loss?

Accepted Answer

Post-training quantization to INT8. Post-training quantization to INT8 is the correct first step because it directly reduces the model size by approximately 4x (from 32-bit floats to 8-bit integers) and speeds up inference on edge devices by leveraging integer-optimized hardware (e.g., ARM NEON or Qualcomm Hexagon). This technique requires no retraining and typically yields minimal accuracy loss for most TensorFlow models, making it the fastest path to deploy on resource-constrained devices.

Answer

Knowledge distillation

Answer

Quantization-aware training

Answer

Weight pruning

Question 6

An engineer is training a model on Vertex AI using a custom container. The training job fails with an error indicating that the container exited with a non-zero status. The engineer wants to debug the issue. What is the best way to access the logs?

Accepted Answer

Check the logs in Cloud Logging (Logs Explorer). Vertex AI automatically streams all container stdout and stderr to Cloud Logging (Logs Explorer). When a custom container exits with a non-zero status, the detailed error messages, stack traces, and application logs are captured there, making it the primary and most comprehensive debugging tool. Cloud Logging provides structured, searchable logs without requiring direct access to the container.

Answer

SSH into the training container using Vertex AI's SSH feature

Answer

View logs in Cloud Storage under the job's output directory

Answer

Use Cloud Debugger to inspect the container

Question 7

A research team is training a very large Transformer model that does not fit into the memory of a single GPU. They have access to multiple GPUs on a single machine and want to split the model layers across GPUs. Which distributed training strategy should they use?

Accepted Answer

Pipeline parallelism (model parallelism). Pipeline parallelism (model parallelism) is the correct choice because it splits the model's layers across multiple GPUs, allowing a model too large for a single GPU's memory to be trained. Each GPU holds a subset of layers and processes micro-batches in a pipelined fashion, enabling training of models that exceed single-device memory. This directly addresses the memory constraint by distributing the model parameters, not just the data.

Answer

MultiWorkerMirroredStrategy

Answer

Parameter server strategy

Answer

MirroredStrategy (data parallelism)

Question 8

An ML team wants to use Vertex AI Hyperparameter Tuning to tune a custom training job. They have a budget of 50 trials and want to use an algorithm that balances exploration and exploitation. Which algorithm should they choose?

Accepted Answer

Bayesian optimization (Vizier default). Bayesian optimization (the default algorithm in Vertex AI Vizier) is the correct choice because it explicitly balances exploration and exploitation by building a probabilistic model of the objective function and using an acquisition function to select the next hyperparameter configuration. With a budget of 50 trials, this algorithm efficiently converges to optimal regions while still exploring uncertain areas, making it ideal for tuning custom training jobs where each trial is computationally expensive.

Answer

Random search

Answer

Grid search

Answer

Manual search

Question 9

An ML engineer is using Vertex AI for distributed training of a PyTorch model across multiple nodes. The training job must use TPUs for high throughput. The engineer sets up the job configuration. Which THREE components are required for the training to work correctly? (Select 3)

Accepted Answer

A startup script to configure the TPU pod (e.g., `xla_lib.sh`). A is correct because TPU pods require a startup script (e.g., `xla_lib.sh`) to initialize the XLA runtime, configure the TPU mesh, and set environment variables like `XRT_TPU_CONFIG`. Without this script, the TPU devices will not be discoverable by the PyTorch/XLA process, causing the training to fail with device-not-found errors.

Answer

A MultiWorkerMirroredStrategy configuration

Answer

A TF_CONFIG environment variable set for each worker

Question 10

A machine learning team is building a feature engineering pipeline using Dataflow. They need to compute features from streaming data and store them in Vertex AI Feature Store for online serving. The features must be updated within 5 seconds of the event. Which TWO services should they combine? (Select 2)

Accepted Answer

Cloud Dataflow for stream processing and feature computation. Cloud Dataflow is correct because it provides unified stream and batch processing with exactly-once semantics, enabling low-latency feature computation from streaming data. It integrates natively with Vertex AI Feature Store for online serving, ensuring features are updated within the required 5-second SLA.

Answer

Cloud Storage for feature store

Answer

Cloud Functions for feature transformation

Answer

BigQuery for feature storage

Question 11

A team is scaling a prototype ML model to production on Vertex AI. The model was developed using scikit-learn and requires custom preprocessing. They want to minimize operational overhead and ensure consistency between training and serving. Which approach should they use?

Accepted Answer

Use a pre-built Vertex AI container for scikit-learn and provide a custom training Python package with preprocessing code included.. Using a pre-built Vertex AI container for scikit-learn with a custom training Python package ensures that the same preprocessing code runs during both training and serving, minimizing operational overhead. This approach leverages Vertex AI's managed infrastructure to handle scaling, monitoring, and consistency without requiring custom container maintenance.

Answer

Train on a local machine and upload the model artifacts to Cloud Storage, then create an endpoint with a pre-built container.

Answer

Deploy the model as a custom prediction routine on Vertex AI Endpoints with a custom container.

Answer

Export the model as a .pkl file and use Vertex AI's 'Import Model' with a default container for inference.

Question 12

A company wants to use Vertex AI Vizier to tune hyperparameters for a PyTorch model. They have a limited budget of 50 training jobs. The objective metric is validation accuracy, and they want to find the best configuration efficiently. Which algorithm should they choose?

Accepted Answer

Bayesian optimization using Vertex AI Vizier.. Vertex AI Vizier is a hyperparameter tuning service that uses Bayesian optimization to efficiently search the hyperparameter space. With a limited budget of 50 trials, Bayesian optimization is more sample-efficient than random or grid search, making it the best choice.

Answer

Random search with 50 random configurations.

Answer

Use a custom algorithm implemented in the training code.

Answer

Grid search with 50 evenly spaced points.

Question 13

A data engineer wants to compute feature aggregates over a large dataset stored in BigQuery and write the results to Vertex AI Feature Store. The pipeline must handle both batch and streaming data. Which Google Cloud service should they use?

Accepted Answer

Cloud Dataflow with Apache Beam. Cloud Dataflow with Apache Beam is a unified stream and batch data processing service. It can read from BigQuery, compute aggregates, and write to Vertex AI Feature Store, handling both batch and streaming data with the same pipeline code. This makes it the ideal choice for the requirement.

Answer

BigQuery scheduled queries

Answer

Cloud Functions triggered by Pub/Sub

Answer

Cloud Dataproc with Spark

Question 14

An ML team is using Vertex AI to train a deep learning model on a large dataset. To reduce costs, they want to use preemptible VMs for training jobs. However, training must complete within a bounded time. Which strategy should they use?

Accepted Answer

Use Vertex AI Training with spot VMs and ensure the training code saves checkpoints periodically to Cloud Storage.. Vertex AI Training supports spot VMs (preemptible instances) for cost savings, and periodic checkpointing to Cloud Storage ensures that training can resume from the last saved state if a VM is preempted, allowing the job to complete within a bounded time despite interruptions.

Answer

Use Cloud TPU instead of GPU; TPUs are not preemptible.

Answer

Use Vertex AI Training without spot VMs, because preemptible VMs are not supported for training.

Answer

Use a single powerful non-preemptible VM to avoid interruptions.

Question 15

A company has a TensorFlow model for image classification that must run on edge devices with limited memory. They need to reduce the model size without significant accuracy loss. Which technique should they use?

Accepted Answer

Post-training quantization using TensorFlow Lite.. Post-training quantization with TensorFlow Lite converts a trained model's weights from 32-bit floats to 8-bit integers, reducing model size by ~4x and speeding up inference on edge devices with minimal accuracy loss. TensorFlow Lite is purpose-built for edge deployment, making this the most direct and practical technique for the stated constraint.

Answer

Knowledge distillation to train a smaller student model.

Answer

Pruning the model weights to zero out unimportant connections.

Answer

Use a larger VM for training.