Code generation in Prompt Engineering / GenAI - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

Output of a simple code generation model snippet

What is the output of this Python code simulating a simple code generation model's token prediction?

Prompt Engineering / GenAI

tokens = ['def', ' ', 'add', '(', 'a', ',', ' ', 'b', ')', ':', '\n', '    ', 'return', ' ', 'a', ' ', '+', ' ', 'b']
predicted = ''.join(tokens)
print(predicted)

A

defadd(a,b):
returna+b

B def add(a, b): return a + b

C

def add(a, b):
    return a + b

D

def add(a, b):
return a + b

💡 Hint

Look carefully at the tokens and how they join with spaces and newlines.

The tokens include spaces and newline characters that format the function definition properly with indentation and line breaks.

❓ Model Choice

intermediate

Choosing the right model for code generation

Which model architecture is best suited for generating code snippets given a prompt?

A Transformer-based model

B Convolutional Neural Network (CNN)

C Recurrent Neural Network (RNN) with LSTM

D K-Nearest Neighbors (KNN)

💡 Hint

Consider models that handle long-range dependencies and context well.

Transformer models excel at capturing long-range dependencies and context, making them ideal for code generation tasks.

❓ Hyperparameter

advanced

Effect of temperature on code generation output

In a code generation model, what is the effect of increasing the temperature parameter during sampling?

A Improves the syntax correctness of generated code

B Generates more deterministic and repetitive code outputs

C Decreases the length of generated code

D Generates more random and diverse code outputs

💡 Hint

Think about how temperature affects randomness in sampling.

Higher temperature increases randomness, leading to more diverse but potentially less coherent outputs.

❓ Metrics

advanced

Evaluating code generation quality

Which metric is most appropriate to evaluate the functional correctness of generated code snippets?

A Number of tokens generated per second

B Code execution accuracy on test inputs

C Perplexity of the language model on generated code

D BLEU score comparing generated code to reference code

💡 Hint

Think about measuring if the code actually works as intended.

Execution accuracy tests if generated code produces correct outputs on inputs, directly measuring functional correctness.

🔧 Debug

expert

Debugging a code generation output error

A code generation model outputs this Python code snippet: "def multiply(a, b):\nreturn a * b" When running, it raises an IndentationError. What is the cause?

Prompt Engineering / GenAI

def multiply(a, b):
return a * b

A The return statement is not indented inside the function

B The function name is invalid

C The multiplication operator is incorrect

D The function is missing a colon at the end

💡 Hint

Python requires code blocks to be indented properly.

Python functions require the body to be indented; missing indentation causes IndentationError.