Free Practice Questions for Snowflake DSA-C03 Certification

- Machine Learning - Supervised learning - Unsupervised learning - Reinforcement learning

- Supervised Learning - Structured Data - Linear regression - Binary classification - Multi-class classification - Time-series forecasting - Unstructured Data - Image classification - Segmentation

- Unsupervised Learning - Clustering

- GenAI - Association models

- Data collection - Data visualization and exploration - Feature engineering - Training models - Model deployment - Model monitoring and evaluation (e.g., model explainability, precision, recall, accuracy, confusion matrix) - Model versioning

- Normal versus skewed distributions (e.g., mean, outliers) - Central limit theorem - Z and T tests - Bootstrapping - Confidence intervals

Use Snowpark for Python and SQL for the following tasks:

- Aggregate - Joins - Identify critical data - Remove duplicates - Remove irrelevant fields - Handle missing values - Data type casting - Sampling data

Use Snowpark and SQL to:

- Identify initial patterns (i.e., data profiling) - Connect external machine learning platforms and/or notebooks (e.g., Jupyter)

Use Snowflake native statistical functions to analyze and calculate descriptive data statistics:

- Window Functions - MIN/MAX/AVG/STDEV - VARIANCE - TOPn - Approximation/High Performing function

Perform Linear Regression to:

- Find the slope and intercept - Verify the dependencies on dependent and independent variables

Perform preprocessing including:

- Scaling data - Encoding - Normalization

Perform data transformations using:

- DataFrames (i.e., pandas, Snowpark, Snowpark pandas) - Derived features (e.g., average spend)

Perform binarizing data through:

- Binning continuous data into intervals - Label encoding - One hot encoding

Utilize the Snowpark Feature Store.

Create statistical summaries using:

- Snowsight with SQL - Interpret open-source graph libraries - Identify data outliers

Utilize Snowflake Notebooks for visualization and interpretation.

Connect Python to Snowflake using:

- Snowpark - Snowpark ML - Python connector with Pandas support

Connect from an external IDE (e.g., Visual Studio Code).

Utilize Snowpark languages.

Use Snowflake Cortex for:

- Vector embedding - Prompt engineering - Fine tuning - Task-specific models (e.g., categorization, summarization, sentiment analysis, information extraction)

Build a data science pipeline with:

- Automation of data transformation (e.g., dynamic tables) - Python User-Defined Functions (UDFs) - Python User-Defined Table Functions (UDTFs)

Perform hyperparameter tuning.

Select optimization metrics (e.g., log loss, AUC, RMSE).

Partition data using:

- Cross validation - Train validation hold-out

Perform down/up-sampling.

Train models with Python stored procedures.

Train models outside Snowflake through external functions.

Train models with Python User-Defined Table Functions (UDTFs).

Use a ROC curve/confusion matrix to calculate the expected payout of the model.

Validate regression problems.

Use a residuals plot and interpret graphics with context.

Evaluate model metrics.

Interpret a model using:

- Feature impact - Partial dependence plots - Confidence intervals - SHAP values, implemented with Python stored procedures

Use an external hosted model via:

- External functions - Pre-built models

Deploy a model in Snowflake using:

- Vectorized/Scalar Python User-Defined Functions (UDFs) - Pre-built models - Storing predictions - Stage commands - Snowflake Model Registry for model logging and retrieving, and Snowpark Container Services

Use metrics for model evaluation:

- Data drift / Model decay, including data distribution comparisons (Do the data making predictions look similar to the training data? Do the same data points give the same predictions once a model is deployed?) - Area under the curve - Accuracy, precision, recall - RMSE (regression)

- Metadata tagging - Model versioning with Snowflake Model Registry - Automation of model retraining