Snowflake Snowflake Certifications DSA-C02 Questions & Answers

• Question 31:

  Which are the following additional Metadata columns Stream contains that could be used for creating Efficient Data science Pipelines and helps in transforming only the New/Modified data only? Choose 3.

  A. METADATA$ACTION

  B. METADATA$FILE_ID

  C. METADATA$ISUPDATE

  D. METADATA$DELETE

  E. METADATA$ROW_ID

  Correct Answer: ACE

  Explanation:

  A stream stores an offset for the source object and not any actual table columns or data. When que-ried, a stream accesses and returns the historic data in the same shape as the source object (i.e. the same column names and ordering) with

  the following additional columns:

  METADATA$ACTION

  Indicates the DML operation (INSERT, DELETE) recorded.

  METADATA$ISUPDATE

  Indicates whether the operation was part of an UPDATE statement. Updates to rows in the source object are represented as a pair of DELETE and INSERT records inthe stream with a metadata column METADATA$ISUPDATE values set to

  TRUE. Note that streams record the differences between two offsets. If a row is added and then updated in the current offset, the delta change is a new row. The METADATA$ISUPDATE row records a FALSE value.

  METADATA$ROW_ID

  Specifies the unique and immutable ID for the row, which can be used to track changes to specific rows over time.

• Question 32:

  Which ones are the known limitations of using External function? Choose all apply.

  A. Currently, external functions cannot be shared with data consumers via Secure Data Sharing.

  B. Currently, external functions must be scalar functions. A scalar external function re-turns a single value for each input row.

  C. External functions have more overhead than internal functions (both built-in functions and internal UDFs) and usually execute more slowly

  D. An external function accessed through an AWS API Gateway private endpoint can be accessed only from a Snowflake VPC (Virtual Private Cloud) on AWS and in the same AWS region.

  Correct Answer: ABCD

• Question 33:

  Which of the following Snowflake parameter can be used to Automatically Suspend Tasks which are running Data science pipelines after specified Failed Runs?

  A. SUSPEND_TASK

  B. SUSPEND_TASK_AUTO_NUM_FAILURES

  C. SUSPEND_TASK_AFTER_NUM_FAILURES

  D. There is none as such available.

  Correct Answer: C

  Explanation:

  Automatically Suspend Tasks After Failed Runs

  Optionally suspend tasks automatically after a specified number of consecutive runs that either fail or time out. This feature can reduce costs by suspending tasks that consume Snowflake credits but fail to run to completion. Failed task runs

  include runs in which the SQL code in the task body either produces a user error or times out. Task runs that are skipped, canceled, or that fail due to a sys-tem error are considered indeterminate and are not included in the count of failed

  task runs.

  Set the SUSPEND_TASK_AFTER_NUM_FAILURES = num parameter on a standalone task or the root task in a DAG. When the parameter is set to a value greater than 0, the following behavior applies to runs of the standalone task or DAG:

  Standalone tasks are automatically suspended after the specified number of consecutive task runs either fail or time out.

  The root task is automatically suspended after the run of any single task in a DAG fails or times out the specified number of times in consecutive runs. The parameter can be set when creating a task (using CREATE TASK) or later (using

  ALTER TASK). The setting applies to tasks that rely on either Snowflake-managed compute resources (i.e. serverless compute model) or user-managed compute resources (i.e. a virtual warehouse).

  The SUSPEND_TASK_AFTER_NUM_FAILURES parameter can also be set at the account, database, or schema level. The setting applies to all standalone or root tasks contained in the modified object. Note that explicitly setting the

  parameter at a lower (i.e. more granular) level overrides the parameter value set at a higher level.

• Question 34:

  Mark the incorrect statement regarding Python UDF?

  A. Python UDFs can contain both new code and calls to existing packages

  B. For each row passed to a UDF, the UDF returns either a scalar (i.e. single) value or, if defined as a table function, a set of rows.

  C. A UDF also gives you a way to encapsulate functionality so that you can call it repeatedly from multiple places in code

  D. A scalar function (UDF) returns a tabular value for each input row

  Correct Answer: D

  Explanation:

  A scalar function (UDF) returns one output row for each input row. The returned row consists of a single column/value

• Question 35:

  Which of the following metrics are used to evaluate classification models?

  A. Area under the ROC curve

  B. F1 score

  C. Confusion matrix

  D. All of the above

  Correct Answer: D

  Explanation:

  Evaluation metrics are tied to machine learning tasks. There are different metrics for the tasks of classification and regression. Some metrics, like precision-recall, are useful for multiple tasks. Classification and regression are examples of

  supervised learning, which constitutes a majority of machine learning applications. Using different metrics for performance evaluation, we should be able to im-prove our model's overall predictive power before we roll it out for production on

  unseen data. Without doing a proper evaluation of the Machine Learning model by using different evaluation metrics, and only depending on accuracy, can lead to a problemwhen the respective model is deployed on unseen data and may

  end in poor predictions.

  Classification metrics are evaluation measures used to assess the performance of a classification model. Common metrics include accuracy (proportion of correct predictions), precision (true positives over total predicted positives), recall (true

  positives over total actual positives), F1 score (har-monic mean of precision and recall), and area under the receiver operating characteristic curve (AUC-ROC).

  Confusion Matrix

  Confusion Matrix is a performance measurement for the machine learning classification problems where the output can be two or more classes. It is a table with combinations of predicted and actual values.

  It is extremely useful for measuring the Recall, Precision, Accuracy, and AUC-ROC curves.

  The four commonly used metrics for evaluating classifier performance are:

  1.

  Accuracy: The proportion of correct predictions out of the total predictions.

  2.

  Precision: The proportion of true positive predictions out of the total positive predictions (precision = true positives / (true positives + false positives)).

  3.

  Recall (Sensitivity or True Positive Rate): The proportion of true positive predictions out of the total actual positive instances (recall = true positives / (true positives + false negatives)).

  4.

  F1 Score: The harmonic mean of precision and recall, providing a balance between the two metrics (F1 score = 2 * ((precision * recall) / (precision + recall))). These metrics help assess the classifier's effectiveness in correctly classifying

  instances of different classes.

  Understanding how well a machine learning model will perform on unseen data is the main purpose behind working with these evaluation metrics. Metrics like accuracy, precision, recall are good ways to evaluate classification models for

  balanced datasets, but if the data is imbalanced then other methods like ROC/AUC perform better in evaluating the model performance.

  ROC curve isn't just a single number but it's a whole curve that provides nuanced details about the behavior of the classifier. It is also hard to quickly compare many ROC curves to each other.

• Question 36:

  A Data Scientist as data providers require to allow consumers to access all databases and database objects in a share by granting a single privilege on shared databases. Which one is incorrect SnowSQL command used by her while doing this task?

  Assuming:

  A database named product_db exists with a schema named product_agg and a table named Item_agg.

  The database, schema, and table will be shared with two accounts named xy12345 and yz23456.

  1.USE ROLE accountadmin;

  2.CREATE DIRECT SHARE product_s;

  3.GRANT USAGE ON DATABASE product_db TO SHARE product_s;

  4.GRANT USAGE ON SCHEMA product_db. product_agg TO SHARE product_s;

  5.GRANT SELECT ON TABLE sales_db. product_agg.Item_agg TO SHARE product_s; 6.SHOW GRANTS TO SHARE product_s;

  7.ALTER SHARE product_s ADD ACCOUNTS=xy12345, yz23456;

  8.SHOW GRANTS OF SHARE product_s;

  A. GRANT USAGE ON DATABASE product_db TO SHARE product_s;

  B. CREATE DIRECT SHARE product_s;

  C. GRANT SELECT ON TABLE sales_db. product_agg.Item_agg TO SHARE product_s;

  D. ALTER SHARE product_s ADD ACCOUNTS=xy12345, yz23456;

  Correct Answer: C

  Explanation:

  CREATE SHARE product_s is the correct Snowsql command to create Share object.

  Rest are correct ones.

  https://docs.snowflake.com/en/user-guide/data-sharing-provider#creating-a-share-using-sql

• Question 37:

  All Snowpark ML modeling and preprocessing classes are in the ________ namespace?

  A. snowpark.ml.modeling

  B. snowflake.sklearn.modeling

  C. snowflake.scikit.modeling

  D. snowflake.ml.modeling

  Correct Answer: D

  Explanation:

  All Snowpark ML modeling and preprocessing classes are in the snowflake.ml.modeling namespace. The Snowpark ML modules have the same name as the corresponding module from the sklearn namespace. For example, the Snowpark

  ML module corresponding to sklearn.calibration is snow-flake.ml.modeling.calibration. The xgboost and lightgbm modules correspond to snowflake.ml.modeling.xgboost and snow-flake.ml.modeling.lightgbm, respectively.

  Not all of the classes from scikit-learn are supported in Snowpark ML.

• Question 38:

  Which type of Python UDFs let you define Python functions that receive batches of input rows as Pandas DataFrames and return batches of results as Pandas arrays or Series?

  A. MPP Python UDFs

  B. Scaler Python UDFs

  C. Vectorized Python UDFs

  D. Hybrid Python UDFs

  Correct Answer: C

  Explanation:

  Vectorized Python UDFs let you define Python functions that receive batches of input rows as Pandas DataFrames and return batches of results as Pandas arrays or Series. You call vectorized Py-thon UDFs the same way you call other

  Python UDFs. Advantages of using vectorized Python UDFs compared to the default row-by-row processing pat-tern include:

  The potential for better performance if your Python code operates efficiently on batches of rows.

  Less transformation logic required if you are calling into libraries that operate on Pandas Data-Frames or Pandas arrays.

  When you use vectorized Python UDFs:

  You do not need to change how you write queries using Python UDFs. All batching is handled by the UDF framework rather than your own code. As with non-vectorized UDFs, there is no guarantee of which instances of your handler code will

  see which batches of input.

• Question 39:

  There are a couple of different types of classification tasks in machine learning, Choose the Correct Classification which best categorized the below Application Tasks in Machine learning?

  To detect whether email is spam or not

  To determine whether or not a patient has a certain disease in medicine.

  To determine whether or not quality specifications were met when it comes to QA (Quality Assurance).

  A. Multi-Label Classification

  B. Multi-Class Classification

  C. Binary Classification

  D. Logistic Regression

  Correct Answer: C

  Explanation: The Supervised Machine Learning algorithm can be broadly classified into Regression and Classification Algorithms. In Regression algorithms, we have predicted the output for continuous values, but to predict the categorical

  values, we need Classification algorithms.

  What is the Classification Algorithm?

  The Classification algorithm is a Supervised Learning technique that is used to identify the category of new observations on the basis of training data. In Classification, a program learns from the given dataset or observations and then

  classifies new observation into a number of classes or groups. Such as, Yes or No, 0 or 1, Spam or Not Spam, cat or dog, etc. Classes can be called as targets/labels or categories. Unlike regression, the output variable of Classification is a

  category, not a value, such as "Green or Blue", "fruit or animal", etc. Since the Classification algorithm is a Supervised learning technique, hence it takes labeled input data, which means it contains input with the corresponding output.

  In classification algorithm, a discrete output function(y) is mapped to input variable(x).

  y=f(x), where y = categorical output

  The best example of an ML classification algorithm is Email Spam Detector. The main goal of the Classification algorithm is to identify the category of a given dataset, and these algorithms are mainly used to predict the output for the

  categorical data. The algorithm which implements the classification on a dataset is known as a classifier.

  There are two types of Classifications:

  Binary Classifier: If the classification problem has only two possible outcomes, then it is called as Binary Classifier.

  Examples: YES or NO, MALE or FEMALE, SPAM or NOT SPAM, CAT or DOG, etc. Multi-class Classifier: If a classification problem has more than two outcomes, then it is called as Multi-class Classifier.

  Example: Classifications of types of crops, Classification of types of music. Binary classification in deep learning refers to the type of classification where we have two class labels - one normal and one abnormal. Some examples of binary

  classification use:

  To detect whether email is spam or not

  To determine whether or not a patient has a certain disease in medicine.

  To determine whether or not quality specifications were met when it comes to QA (Quality Assurance).

  For example, the normal class label would be that a patient has the disease, and the abnormal class label would be that they do not, or vice-versa. As is with every other type of classification, it is only as good as the binary classification

  dataset that it has ?or, in other words, the more training and data it has, the better it is.

• Question 40:

  Consider a data frame df with 10 rows and index [ 'r1', 'r2', 'r3', 'row4', 'row5', 'row6', 'r7', 'r8', 'r9', 'row10']. What does the aggregate method shown in below code do?

  g = df.groupby(df.index.str.len())

  A. aggregate({'A':len, 'B':np.sum})

  B. Computes Sum of column A values

  C. Computes length of column A

  D. Computes length of column A and Sum of Column B values of each group

  E. Computes length of column A and Sum of Column B values

  Correct Answer: C

  Explanation:

  Computes length of column A and Sum of Column B values of each group