MLS-C01 Practice Questions 2 — 50 Questions

Question 1

A data engineer runs the AWS CLI command above to inspect a file in S3. They need to determine if the file was modified after a Glue ETL job processed it. What additional information could they obtain from this command?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

The object's last modified timestamp.. The AWS CLI command `aws s3api head-object --bucket my-bucket --key my-key` returns metadata about an S3 object without downloading it. The `LastModified` field in the response provides the exact timestamp of the last modification, which can be compared to the Glue ETL job's execution time to determine if the file was modified after processing. This is the only field that directly answers the question about modification timing.

Answer

The object's content type.

Answer

The object's storage class.

Answer

The object's ETag.

Question 2

A data scientist is performing EDA on a dataset of 1 million images stored in Amazon S3. Each image is 100x100 pixels in RGB format. The data scientist wants to compute the mean pixel value per channel across the entire dataset. Which approach is most efficient?

Accepted Answer

Use Amazon SageMaker Processing with a custom Python script that iterates over S3 objects and aggregates pixel values.. (Amazon SageMaker Processing with a custom Python script) is the most efficient because it can distribute the computation across multiple instances, processing images in parallel without loading all into memory at once. This is ideal for a large dataset of 1 million images. Option B (Athena) is designed for querying structured data, not image processing. Option C (AWS Glue ETL) is for ETL on tabular data, not image processing. Option D (SageMaker notebook with large instance) would require loading all images into memory, which is not feasible for 1 million images.

Answer

Use Amazon Athena with a SQL query on the image metadata stored in a CSV file.

Answer

Use AWS Glue ETL to read images and compute the mean.

Answer

Use a SageMaker notebook instance with a large instance type to load all images into memory and compute the mean.

Question 3

Which TWO of the following are appropriate methods for handling missing data in a dataset?

Accepted Answer

Multiple imputation. Multiple imputation and using algorithms that handle missing values (e.g., XGBoost) are valid. Listwise deletion reduces sample size. Mean imputation may bias distributions. Dropping features with many missing values may lose information.

Answer

Dropping features with more than 50% missing values

Answer

Mean imputation for all features

Answer

Listwise deletion (removing rows with missing values)

Question 4

A data scientist is using SageMaker to train a linear regression model. The target variable has a long-tail distribution. Which data transformation is LEAST likely to improve model performance?

Accepted Answer

Normalize all feature values to [0,1]. Normalizing feature values to [0,1] is a scaling technique that does not address the long-tail distribution of the target variable. Long-tail distributions typically require transformations that compress the tail (e.g., log or Box-Cox) to make the relationship more linear and reduce the influence of extreme values. Feature normalization helps with gradient descent convergence but does not fix skewness in the target, so it is least likely to improve model performance for this specific issue.

Answer

Add interaction terms between features

Answer

Apply log transformation to the target variable

Answer

Remove outliers from the target variable

Question 5

A machine learning team is reviewing a dataset for a regression problem. They notice that the target variable has a right-skewed distribution. Which transformation should they consider applying to the target variable to improve model performance?

Accepted Answer

Apply log transformation to the target variable.. Log transformation is commonly applied to right-skewed data to make it more normally distributed, which can improve model performance. Option A (StandardScaler) is for scaling, not skewness. Option B (MinMaxScaler) also doesn't address skewness. Option D (One-hot encoding) is for categorical variables.

Answer

Apply StandardScaler to the target variable.

Answer

Apply MinMaxScaler to the target variable.

Answer

Apply one-hot encoding to the target variable.

Question 6

During exploratory data analysis, a data scientist notices that the Pearson correlation coefficient between two continuous variables is 0.85. What does this indicate?

Accepted Answer

A strong positive linear relationship. A correlation coefficient of 0.85 indicates a strong positive linear relationship. Option A is wrong because correlation does not imply causation. Option C is wrong because the coefficient is positive and strong, not negative and weak. Option D is wrong because a coefficient of 0.85 indicates a strong relationship, not no relationship.

Answer

A causal relationship between the two variables

Answer

A weak negative linear relationship

Answer

No relationship between the variables

Question 7

A machine learning engineer is examining a dataset containing text reviews. They want to convert the text into numerical features for a model. During EDA, they notice that the word 'the' appears in almost every review, while words like 'excellent' appear rarely. Which of the following techniques should they use to reduce the impact of very common words?

Accepted Answer

Apply TF-IDF transformation.. TF-IDF transformation downweights common words (like 'the') and emphasizes rare but informative words (like 'excellent'). Option B (removing stopwords) is insufficient because it does not adjust for frequency beyond removing a predefined list; TF-IDF handles frequency weighting. Option C (word2vec embeddings) captures semantic relationships but does not specifically reduce the impact of common words. Option D (bag-of-words) does not perform any weighting, so common words dominate.

Answer

Remove stopwords from the text.

Answer

Use word2vec embeddings.

Answer

Use a bag-of-words representation.

Question 8

A data scientist is training a binary classification model on an imbalanced dataset where the positive class represents only 5% of the data. The model currently achieves 95% accuracy but only 10% recall on the positive class. Which metric should the scientist focus on to improve the model's ability to detect the positive class?

Accepted Answer

Recall. Recall measures the proportion of actual positives the model correctly identifies (TP / (TP + FN)). With only 10% recall on the positive class, the model is missing 90% of true positives — exactly the problem the scientist needs to fix. Optimizing recall (e.g., via class weighting, resampling, or threshold tuning) directly improves the model's ability to detect the minority positive class.

Answer

Accuracy

Answer

Precision

Answer

AUC-ROC

Question 9

A machine learning team is building a fraud detection system using Amazon SageMaker. The training data is highly imbalanced (99% legitimate, 1% fraudulent). They need to maximize the recall of the fraud class while keeping precision above 90%. Which approach should they take?

Accepted Answer

Train a model using the original data, then adjust the decision threshold on the validation set to maximize recall while precision > 90%. Adjusting the decision threshold on the validation set directly optimizes the trade-off between recall and precision. By lowering the threshold, the model classifies more instances as fraud, increasing recall, while the precision constraint (≥90%) ensures the threshold is set at a point where false positives remain acceptably low. This approach works with any probabilistic classifier and does not alter the training data distribution.

Answer

Undersample the majority class to create a balanced dataset and train a Random Forest

Answer

Train an XGBoost model with scale_pos_weight parameter set to 99

Answer

Use SMOTE to oversample the fraud class and then train a logistic regression

Question 10

A data scientist is training a deep learning model for object detection using Amazon SageMaker. The training job is using a single GPU instance and is taking too long. Which THREE actions can reduce training time? (Choose THREE.)

Accepted Answer

Enable mixed precision training with FP16. Option B is correct because mixed precision training with FP16 reduces memory bandwidth and uses tensor cores on modern NVIDIA GPUs, allowing faster matrix operations and larger effective batch sizes, which shortens training time. Option C is correct because moving from a single-GPU instance to a multi-GPU instance like p3.16xlarge (8 V100 GPUs) enables data-parallel training within one instance, increasing throughput and reducing wall-clock training time. Option E is correct because distributed training across multiple instances (e.g., SageMaker distributed data parallel) scales the workload horizontally, further cutting training time for large models. Option A is incorrect because CPU instances lack the massive parallelism and tensor cores of GPUs, so training would be dramatically slower. Option D is incorrect because reducing the batch size lowers per-step compute but increases the number of steps and often reduces hardware utilization, generally increasing total training time.

Answer

Use a CPU instance instead of GPU

Answer

Reduce the batch size

Question 11

A machine learning team is deploying a time-series forecasting model using Amazon SageMaker. The model is trained on historical data and needs to be updated daily with new data. The team wants to automate the retraining pipeline and avoid manual intervention. Which approach is the most efficient?

Accepted Answer

Use Amazon SageMaker Pipelines with a scheduled Lambda function to trigger retraining daily.. Amazon SageMaker Pipelines provides a fully managed, end-to-end orchestration service for building, training, and deploying machine learning models. By combining it with a scheduled AWS Lambda function, the team can automate daily retraining without manual intervention, leveraging SageMaker's native integration for step sequencing, artifact tracking, and model registry updates.

Answer

Use AWS Step Functions to orchestrate retraining, but require a manual approval step.

Answer

Use SageMaker training jobs manually triggered by the team each day.

Answer

Use a cron job on an EC2 instance to run a training script.

Question 12

A data scientist is training a linear regression model to predict house prices. The dataset contains 10 features. After training, the data scientist notices that the model has high bias (underfitting). Which action should the data scientist take to reduce bias?

Accepted Answer

Add more features, such as polynomial features. High bias (underfitting) means the model is too simple to capture the underlying patterns in the data. Adding more features, such as polynomial features, increases model complexity, allowing the linear regression model to fit non-linear relationships and reduce bias. This directly addresses the underfitting issue by giving the model more expressive power.

Answer

Reduce the amount of training data

Answer

Increase the regularization strength

Answer

Use a simpler model, such as ridge regression

Question 13

Which THREE of the following are valid strategies to reduce overfitting in a deep neural network? (Choose 3)

Accepted Answer

Use early stopping.. Option B (Use early stopping) is correct because monitoring validation loss and halting training when it stops improving prevents the model from continuing to fit noise in the training set, which is a classic overfitting-reduction technique. Option D (Add L2 regularization to the loss function) is correct because L2 adds a penalty proportional to the sum of squared weights, shrinking weights toward zero and constraining model complexity, thereby reducing overfitting. Option E (Use dropout layers) is correct because dropout randomly deactivates a fraction of neurons during each training step, forcing the network to learn redundant, robust features and acting as an ensemble-like regularizer. Option A (Increase the number of layers) is not correct because adding layers increases model capacity, which typically worsens overfitting rather than reducing it. Option C (Increase the learning rate) is not correct because a larger learning rate affects optimization speed and stability, not regularization, and can even lead to divergence or poorer generalization rather than reducing overfitting.

Answer

Increase the number of layers.

Answer

Increase the learning rate.

Question 14

A data scientist is training a Random Forest model on Amazon SageMaker. The model performs well on the training set but poorly on the test set. Which technique should the data scientist use to address this issue?

Accepted Answer

Decrease the maximum depth of each tree. The model is overfitting, as indicated by high training performance and poor test performance. Decreasing the maximum depth of each tree limits the complexity of individual trees, reducing overfitting by preventing them from memorizing noise in the training data. This is a standard regularization technique for Random Forest models in Amazon SageMaker.

Answer

Increase the number of trees in the forest

Answer

Increase the learning rate

Answer

Increase the maximum depth of each tree

Question 15

An e-commerce company uses a linear regression model to predict customer lifetime value (LTV). The model shows high variance on the test set, with training RMSE much lower than test RMSE. Which of the following is the MOST effective approach to reduce overfitting?

Accepted Answer

Apply L2 regularization (Ridge regression). High variance (low training RMSE, high test RMSE) indicates overfitting. L2 regularization (Ridge regression) adds a penalty proportional to the square of the coefficients, shrinking them toward zero without eliminating them, which reduces model complexity and improves generalization. This directly addresses overfitting by constraining the model's sensitivity to noise in the training data.

Answer

Use a polynomial kernel in a support vector regressor

Answer

Add more features, including interaction terms

Answer

Increase training data size by duplicating existing samples

Question 16

A data scientist is working with a dataset that contains both numerical and categorical features. The target variable is continuous. Which TWO EDA techniques should the scientist use to understand relationships between features and the target?

Accepted Answer

Create scatter plots of numerical features against the target variable.. Option C is correct because with a continuous target variable, scatter plots of each numerical feature against the target directly reveal the shape, direction, and strength of the relationship (linear, nonlinear, or none), which is exactly the goal of bivariate EDA. Option D is correct because box plots of the continuous target grouped by each categorical feature's categories show how the target's distribution (median, spread, outliers) shifts across groups, exposing categorical-feature/target relationships. Option A is wrong because a confusion matrix evaluates classification predictions and requires discrete class labels, not a continuous target. Option B is wrong because the silhouette score measures clustering quality/cohesion-separation, not feature-target relationships. Option E is wrong because a histogram of the target is a univariate analysis that describes the target's own distribution and does not relate any feature to the target.

Answer

Generate a confusion matrix for the target variable.

Answer

Compute the silhouette score for each feature.

Answer

Plot a histogram of the target variable.

Question 17

A machine learning engineer is deploying a PyTorch model to SageMaker. The model requires custom inference logic. Which approach should the engineer use?

Accepted Answer

Create a custom inference script and use the SageMaker PyTorch container. SageMaker allows you to provide a custom inference script (entry point) when using the PyTorch container, enabling custom inference logic. Option A is wrong because the built-in container as-is would not incorporate custom logic. Option B is wrong because SageMaker Ground Truth is for labeling, not model deployment. Option C is wrong because SageMaker Processing is for data processing, not inference.

Answer

Use a SageMaker built-in PyTorch container as-is

Answer

Use SageMaker Ground Truth to deploy the model

Answer

Use SageMaker Processing to run inference

Question 18

A company is using Amazon SageMaker to train a model on a large dataset stored in S3. The training job is taking a long time due to slow data loading. Which action can the data scientist take to reduce data loading time?

Accepted Answer

Use Pipe mode to stream data from S3.. Pipe mode streams data directly from S3 into the training algorithm without first downloading it to the local storage, eliminating the bottleneck of disk I/O and reducing data loading time. This is especially effective for large datasets where the time to copy data to EBS (File mode) dominates the training job duration.

Answer

Use File mode and copy data to Amazon EBS.

Answer

Use a larger instance type with more memory.

Answer

Enable data augmentation during training.

Question 19

A data scientist creates the above IAM policy and attaches it to a role used by an Amazon SageMaker notebook instance. When trying to save a file to the S3 bucket, the operation fails. What is the missing permission?

Accepted Answer

kms:GenerateDataKey. (kms:GenerateDataKey) because the S3 bucket is likely encrypted with a KMS key. When SageMaker writes an object to an encrypted bucket, it needs permission to call kms:GenerateDataKey to generate a data key for encryption. Option A (kms:Decrypt) is for decryption, not encryption. Option B (s3:ListBucket) allows listing objects, not writing. Option D (s3:GetObject) allows reading objects, not writing. The error when saving indicates missing encryption permissions.

Answer

kms:Decrypt

Answer

s3:ListBucket

Answer

s3:GetObject

Question 20

A data scientist is analyzing server logs stored in Amazon CloudWatch Logs. The above snippet shows three log entries. They want to count the number of 500 errors per minute using CloudWatch Logs Insights. Which query should they use?

Accepted Answer

fields @timestamp, @message | filter @message like /500/ | stats count() by bin(1m). The correct query is D. It reads the @timestamp and @message fields, filters log entries containing '500' in the message, and then counts the number of such entries per minute using stats count() by bin(1m). This directly matches the requirement to count 500 errors per minute. Option A attempts to filter on a 'status' field, but the logs shown do not have a separate status field; the status code is embedded in the message. Option B filters for 'ERROR', but the logs use numeric status codes, not the word 'ERROR'. Option C sorts the results instead of aggregating, so it does not provide the count per minute.

Answer

fields @timestamp, status | filter status = 500 | stats count() by bin(1m)

Answer

fields @timestamp, @message | filter @message like /ERROR/ | stats count() by bin(1m)

Answer

fields @timestamp, @message | filter @message like /500/ | sort @timestamp desc