Free A00-408 SAS Viya Natural Language Processing and Computer Vision Exam Dumps

dumps2024-05-11T09:47:34+08:00

A00-408 SAS Viya Natural Language Processing and Computer Vision real exam contains 60 multiple-choice and short-answer questions, and you have 110 minutes to complete the test. The required passing score for the A00-408 exam is 70%. SAS Certified Specialist: Natural Language Processing and Computer Vision Using SAS Viya certification is valid for 5 years. We provide the latest SAS Institute A00-408 exam dumps, which are valuable for you to pass. Free SAS Viya Natural Language Processing and Computer Vision A00-408 exam dumps are available below.

Page 1 of 3

1. Given the code specifying the two initial layers of a convolutional network:

deepLearn.addLayer / layer={type="INPUT"

nchannels=1 width=10 height=10}

modelTable={name="simple"} name="data";

deepLearn.addLayer / layer={type="CONVO"

nFilters=10 width=5 height=5 includeBias=True}

modelTable={name="simple"} name="conv1"

srcLayers={"data"};

deepLearn.addLayer / layer={type='POOL'

width=2 height=2 stride=2 pool='max' }

modelTable={name="simple"} name="pool1"

srcLayers={"conv1"}

How many parameters are in the model?

259

260

250

274

2. Which statements are true regarding the Topics node? (Choose 2)

It is possible to promote a topic to a category.

For a topic to be assigned to a document, all terms in the topic must be present in the document.

System generated topics can be split or merged.

By default, 10 system generated topics are created.

3. Enter the keyword for the blank line below that will return the string EM if both the string EM and the words Expectation Maximization are in a document. CLASSIFIER: [ _______ =EM]: Expectation Maximization

4. Which statement is true regarding a SAS Visual Text Analytics project?

There are only three roles the user can assign for variables: Text, Category, and none.

A variable with a role of Category is required to define a project.

Text and Category variables are the only variable types that can be used as Display variables.

There is only a single Visual Text Analytics pipeline available to users.

5. Which of the following is a best practice before building deep learning models after loading images into a CAS Table using the loadImages action?

Resize the images so they are all 32 x 32 pixels.

Convert the images into a flattened format.

Remove color channels from the image so they are greyscale.

Shuffle the images so that they are randomly ordered.

6. Which of the following best describes when and how to use batch normalization?

Can be applied after any layer between the input and output layers and the bias term is included when performing the normalization.

Can be applied after any layer between the input and output layers and the bias term is dropped prior to performing the normalization.

Can be applied after convolution layers and the bias term is dropped when performing the normalization.

Can be applied after convolution layers and the bias term is included when performing the normalization.

7. Which of the following is a valid summary function for the pool option when using a pooling layer in a convolutional network?

SUM

MEDIAN

PRODUCT

AVERAGE

8. Given the code specifying the two initial layers of a convolutional network:

deepLearn.addLayer / layer={type=”INPUT”

nchannels=1 width=10 height=10}

modelTable={name=”simple”}

name=”data”;

deepLearn.addLayer / layer={type=”CONVO”

nFilters=1 width=5 height=5 stride=1}

modelTable={name=”simple”}

name=”conv1”

srcLayers={“data”};

Which of the following statements describes the output tensor (image) of conv1?

The output tensor will be larger in width and height than the input data tensor.

The output tensor dimensions cannot be determined from the information above.

The output tensor will be the same width and height as the input data tensor.

The output tensor will be smaller in width and height than the input data tensor

9. Examine the term map.

What does the symbol ~ mean for the terms different and low?

The symbol and term appear in documents as shown.

Any synonyms of the terms are matched.

Additional terms appear when you click in the map.

The terms are excluded from documents.

10. What can the Sandbox be used for in Visual Text Analytics nodes?

Evaluate results of user topics before activating them in production.

Evaluate scoring code of the concepts node before activating it in production.

Evaluate rule syntax of a categories rule before activating it in a production rule.

Evaluate rule syntax of a concepts rule before activating it in a production rule.

Page 2 of 3

11. Which statement is FALSE regarding the loadImages action?

All images are represented in a single table after loading.

Images can be 3-dimensional.

Images from multiple sources can be loaded in a single call.

Encoded images must be decoded prior to loading.

12.

13. Given the exhibit below taken from the Topics node, which statements are true? (Choose 2)

All terms have a Relevancy of 1 because the selected topic is a user created topic.

The selected topic appears in 13 documents.

There are 21,918 kept terms that will be used in the analysis.

The selected topic has been promoted to a Category.

14. Which rule types are valid for the LITI language in the Concepts Node? (Choose 2)

CLASSIFIER_RULE

TERM

NO_BREAK

C_CONCEPT

15. Given the CASL code:

Which statement would assign the correct values to build a bi-directional recurrent neural network capable of sequence labeling?

r1 = True, r2= True, src1={'rnn11'}, src2={'data','rnn12'}

r1 = False, r2= True, src1={'rnn11'}, src2={'rnn11','rnn12'}

r1 = True, r2= False, src1={'data'}, src2={'rnn11','rnn12'}

r1 = False, r2= True, src1={'data'}, src2={'rnn11','data'}

16. Which of the following is a feature of the Hyperband method for tuning hyperparameters?

Computational resources are reallocated from poorly performing hyperparameter combinations during the training process, speeding up the training process.

The final model selected by the Hyperband method will have the lowest error on the training data.

The model is trained to convergence for each hyperparameter combination, and the best model is selected from all hyperparameter combinations.

A genetic algorithm is used to intelligently select hyperparameter combinations.

17. Given the following property settings for the Text Parsing node, which of the following statements is true?

There is a default start list and default stop list and both will be used in the analysis.

Misspelled words will automatically be placed in the Stop list.

Only the default synonym list will be used in the analysis.

Terms in 5 and fewer documents will be placed in the Stop list.

18. Which statements are true regarding the data source for a SAS Visual Text Analytics project? (Choose 2)

There must be a variable with the role of CATEGORY in a data source.

You can set multiple variables to the role of CATEGOR

A variable must have a role of CATEGORY before it can be set as a display variable.

A data source requires one variable with a role of TEX

19. What is a reason for using a small training dataset for an image classification task?

Small datasets do not have to be partitioned.

Classification networks work well with small datasets.

To prevent overfitting.

To ensure the network is capable of learning.

20. What are ways to modify the predefined concepts to include your own user-defined rule statements when using Model Studio to perform Text Analytics? (Choose 2)

You can create a custom concept that refers to a predefined concept and adds rules.

You can add your own LITI concept rules to any nlp rule to customize its behavior.

You can only edit predefined concepts if a Text Parsing node precedes the concepts node.

You can include or exclude creation of specific predefined system generated concepts.

Page 3 of 3

21. Which statements are TRUE regarding the creation of automatic custom concept rules from textual elements in the Concept node? (Choose two)

Text Parsing node must precede a Concept node in the pipeline when creating automatic custom concept rules from textual elements.

The automatic custom concept rules generated from textual elements in the Concept node is the PREDICATE_RULE type only.

The textual elements available for writing concept rules are comprised of all terms generated by the Text Parsing node.

The automatic custom concept rules generated from textual elements in the concept node is CONCEPT_RULE type only.

Free A00-408 SAS Viya Natural Language Processing and Computer Vision Exam Dumps

Free A00-408 SAS Viya Natural Language Processing and Computer Vision Exam Dumps

Share this post

Leave a Reply Cancel reply

Related Posts