You have recently created a proof-of-concept (POC) deep learning model. You are satisfied with the overall architecture, but you need to determine the value for a couple of hyperparameters. You want to perform hyperparameter tuning on Vertex AI to determine both the appropriate embedding dimension for a categorical feature used by your model and the optimal learning rate. You configure the following settings: For the embedding dimension, you set the type to INTEGER with a minValue of 16 and maxValue of 64. For the learning rate, you set the type to DOUBLE with a minValue of 10e-05 and maxValue of 10e-02. You are using the default Bayesian optimization tuning algorithm, and you want to maximize model accuracy. Training time is not a concern. How should you set the hyperparameter scaling for each hyperparameter and the maxParallelTrials?
Correct Answer: A
The best option for performing hyperparameter tuning on Vertex AI to determine the appropriate embedding dimension and the optimal learning rate is to use UNIT_LINEAR_SCALE for the embedding dimension, UNIT_LOG_SCALE for the learning rate, and a large number of parallel trials. This option has the following advantages: * It matches the appropriate scaling type for each hyperparameter, based on their range and distribution. The embedding dimension is an integer hyperparameter that varies linearly between 16 and 64, so using UNIT_LINEAR_SCALE makes sense. The learning rate is a double hyperparameter that varies exponentially between 10e-05 and 10e-02, so using UNIT_LOG_SCALE is more suitable. * It maximizes the exploration of the hyperparameter space, by using a large number of parallel trials. Since training time is not a concern, using more trials can help find the best combination of hyperparameters that maximizes model accuracy. The default Bayesian optimization tuning algorithm can efficiently sample the hyperparameter space and converge to the optimal values. The other options are less optimal for the following reasons: * Option B: Using UNIT_LINEAR_SCALE for the embedding dimension, UNIT_LOG_SCALE for the learning rate, and a small number of parallel trials, reduces the exploration of the hyperparameter space, by using a small number of parallel trials. Since training time is not a concern, using fewer trials can miss some potentially good combinations of hyperparameters that maximize model accuracy. The default Bayesian optimization tuning algorithm can benefit from more trials to sample the hyperparameter space and converge to the optimal values. * Option C: Using UNIT_LOG_SCALE for the embedding dimension, UNIT_LINEAR_SCALE for the learning rate, and a large number of parallel trials, mismatches the appropriate scaling type for each hyperparameter, based on their range and distribution. The embedding dimension is an integer hyperparameter that varies linearly between 16 and 64, so using UNIT_LOG_SCALE is not suitable. The learning rate is a double hyperparameter that varies exponentially between 10e-05 and 10e-02, so using UNIT_LINEAR_SCALE makes less sense. * Option D: Using UNIT_LOG_SCALE for the embedding dimension, UNIT_LINEAR_SCALE for the learning rate, and a small number of parallel trials, combines the drawbacks of option B and option C. It * mismatches the appropriate scaling type for each hyperparameter, based on their range and distribution, and reduces the exploration of the hyperparameter space, by using a small number of parallel trials. References: * [Vertex AI: Hyperparameter tuning overview] * [Vertex AI: Configuring the hyperparameter tuning job]
Question 137
While running a model training pipeline on Vertex Al, you discover that the evaluation step is failing because of an out-of-memory error. You are currently using TensorFlow Model Analysis (TFMA) with a standard Evaluator TensorFlow Extended (TFX) pipeline component for the evaluation step. You want to stabilize the pipeline without downgrading the evaluation quality while minimizing infrastructure overhead. What should you do?
Correct Answer: C
The best option to stabilize the pipeline without downgrading the evaluation quality while minimizing infrastructure overhead is to use Dataflow as the runner for the evaluation step. Dataflow is a fully managed service for executing Apache Beam pipelines that can scale up and down according to the workload. Dataflow can handle large-scale, distributed data processing tasks such as model evaluation, and it can also integrate with Vertex AI Pipelines and TensorFlow Extended (TFX). By using the flag -runnerDataflowRunner in beam_pipeline_args, you can instruct the Evaluator component to run the evaluation step on Dataflow, instead of using the default DirectRunner, which runs locally and may cause out-of-memory errors. Option A is incorrect because adding tfma.MetricsSpec() to limit the number of metrics in the evaluation step may downgrade the evaluation quality, as some important metrics may be omitted. Moreover, reducing the number of metrics may not solve the out-of-memory error, as the evaluation step may still consume a lot of memory depending on the size and complexity of the data and the model. Option B is incorrect because migrating the pipeline to Kubeflow hosted on Google Kubernetes Engine (GKE) may increase the infrastructure overhead, as you need to provision, manage, and monitor the GKE cluster yourself. Moreover, you need to specify the appropriate node parameters for the evaluation step, which may require trial and error to find the optimal configuration. Option D is incorrect because moving the evaluation step out of the pipeline and running it on custom Compute Engine VMs may also increase the infrastructure overhead, as you need to create, configure, and delete the VMs yourself. Moreover, you need to ensure that the VMs have sufficient memory for the evaluation step, which may require trial and error to find the optimal machine type. Reference: Dataflow documentation Using DataflowRunner Evaluator component documentation Configuring the Evaluator component
Question 138
You have created a Vertex Al pipeline that includes two steps. The first step preprocesses 10 TB data completes in about 1 hour, and saves the result in a Cloud Storage bucket The second step uses the processed data to train a model You need to update the model's code to allow you to test different algorithms You want to reduce pipeline execution time and cost, while also minimizing pipeline changes What should you do?
Correct Answer: B
Question 139
A company uses a long short-term memory (LSTM) model to evaluate the risk factors of a particular energy sector. The model reviews multi-page text documents to analyze each sentence of the text and categorize it as either a potential risk or no risk. The model is not performing well, even though the Data Scientist has experimented with many different network structures and tuned the corresponding hyperparameters. Which approach will provide the MAXIMUM performance boost?
Correct Answer: D
Question 140
You need to train a ControlNet model with Stable Diffusion XL for an image editing use case. You want to train this model as quickly as possible. Which hardware configuration should you choose to train your model?
Correct Answer: A
NVIDIA A100 GPUs are optimized for training complex models like Stable Diffusion XL. Using float32 precision ensures high model accuracy during training, whereas float16 or bfloat16 may cause lower precision in gradients, especially important for image editing. Distributing across multiple instances with T4 GPUs (Options C and D) would not speed up the process effectively due to lower power and more complex setup requirements.
