Module 10 Test: ML and DL Projects Pipeline
Choose one answer for each question. Questions cover the topics summarized in Module 10.
Questions
-
What is the first step in the ML overview?
- A. Problem Definition
- B. Deployment
- C. Monitoring
- D. Feature Engineering
-
Which sources are listed for ML data loading?
- A. Only an app
- B. A CSV, database or API
- C. Only a neural network
- D. A chart and a model
-
What does the ML preprocessing stage include?
- A. Saving and deploying the model
- B. Defining the optimizer
- C. Data cleaning and visualization (EDA)
- D. Building a DataLoader
-
What does the ML Feature Engineering stage include?
- A. Model deployment
- B. Cross-validation
- C. Data loading from an API
- D. A train/test split
-
Which activities are listed in the ML evaluation stage?
- A. Prediction, evaluation/tuning, cross-validation and pipeline building
- B. Data loading and cleaning only
- C. Defining a neural-network loss function
- D. Saving and deploying a model only
-
What is created from data before the DL Dataset and DataLoader steps?
- A. CSV files
- B. PyTorch tensors
- C. An API
- D. A monitoring chart
-
In TensorFlow/Keras, where are DL steps 10–13 usually handled?
- A. In the Dataset and DataLoader
- B. In
model.compile()andmodel.fit() - C. In an API key
- D. In the train/test split
-
What happens when new data arrives after deployment and monitoring?
- A. The pipeline ends permanently
- B. Only the data-loading step is repeated
- C. The project pipeline cycle runs again
- D. Monitoring is removed
Answer key
- A · 2. B · 3. C · 4. D · 5. A · 6. B · 7. B · 8. C