Professional Machine Learning Engineer Exam QuestionsBrowse all questions from this exam

Professional Machine Learning Engineer Exam - Question 38


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?

Show Answer
Correct Answer: D

To minimize computational overhead for batch predictions on text data stored in BigQuery, the best option is to submit a batch prediction job on AI Platform that points to the model location in Cloud Storage. This method allows for direct integration with the trained model and efficient processing of large datasets, leveraging the capabilities of AI Platform for scalable and optimized predictions. Exporting the model to BigQuery ML, on the other hand, is not suitable for text classification models as BigQuery ML mainly supports structured data and simpler machine learning models.

Discussion

17 comments
Sign in to comment
maartenalexanderOption: A
Jun 22, 2021

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

q4exam
Sep 22, 2021

BQML doesnt support NLP model

ms_lemon
Oct 10, 2021

you can import a TF model in BQ ML

gcp2021go
Oct 29, 2021

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

harithacML
Jul 13, 2023

No need . This is a text classification problem. need to convert words to numbers and use a classifier.

chohanOption: A
Jun 18, 2021

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

Darshan12Option: D
May 17, 2023

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.

rexduoOption: A
May 21, 2023

I think D have extra compute on extrating data frm BQ

LitingOption: A
Jul 7, 2023

minimize computational overhead–>BigQuery

harithacMLOption: A
Jul 13, 2023

Model : AI Platform. pred batch data : BigQuery constraint : computational overhead Same platform as data == less computation required to load and pass it to model

Sum_SumOption: A
Nov 15, 2023

A - you can import TF models to BQ

Aastha_VashistOption: A
Mar 19, 2024

Bquery to minimize computational overhead

lucaluca1982Option: D
Apr 13, 2023

D is more straightforward

lucaluca1982Option: C
Apr 20, 2023

what about C?

tavva_prudhvi
Jul 2, 2023

This is an option that can be used to minimize computational overhead, but it is more complex to set up and requires you to have Dataflow installed.

king31
Nov 26, 2023

Although it's more complex, the question doesn't imply any restrictions on complexity, only computational overheard

lucaluca1982Option: C
Apr 25, 2023

why not C?

M25Option: D
May 9, 2023

Went with D

Voyager2
Jun 8, 2023

Not sure Text Classification Using BigQuery ML and ML.NGRAMS https://medium.com/@jeffrey.james/text-classification-using-bigquery-ml-and-ml-ngrams-6e365f0b5505

Voyager2
Jun 8, 2023

Not sure if when you have the saved model in Cloud storage that means that you don't use compute in vertex. I think that the option compute-free is bigquery

MrTracerOption: D
Dec 27, 2023

Would go with D

girguOption: D
May 26, 2024

Use the gcloud command to submit a batch prediction job, specifying the model location in Cloud Storage and the BigQuery table as the input source.

PhilipKokuOption: A
Jun 6, 2024

A) BigQuery ML