[WIP] Create deep learning final exam notebook

[Copilot]

• Explore the repository structure and understand the source notebook (NB-computer_vision_intro.ipynb)
• Identify the key topics to be covered in the final exam
• Create the final exam notebook (dl-final_exam.ipynb) with the following sections:
  • Title and introduction
  • Part 1: Multiple Choice Questions (10 questions with 5 options each)
  • Part 2: Short Answer Questions (10 questions)
  • Part 3: Descriptive Questions (10 questions)
• Include correct answers and explanations for all questions
• Verify the notebook is properly formatted and saved
• Run code review and security checks

Summary

Successfully created notebooks/dl-final_exam.ipynb - a comprehensive Deep Learning final exam with 30 questions covering all topics from NB-computer_vision_intro.ipynb including CNNs, Conv2D/MaxPooling layers, padding/strides, transfer learning, data augmentation, and TensorFlow datasets. All questions include answers and detailed explanations for effective learning.

Original prompt

Create a new Jupyter notebook file named notebooks/dl-final_exam.ipynb in the codingalzi/dlp2 repository.

This notebook will serve as a final exam based on the content of notebooks/NB-computer_vision_intro.ipynb.
The structure of the notebook should include the following sections:

Deep Learning Final Exam

This exam covers the material from the "Introduction to Computer Vision" notebook (NB-computer_vision_intro.ipynb).

---

Part 1: Multiple Choice Questions (10 Questions)

Instructions: Choose the best answer for each question. (Concept-focused)

[Generate 10 multiple-choice questions with 5 options each. Include the question, options, correct answer, and a brief explanation for each.]

---

Part 2: Short Answer Questions (10 Questions)

Instructions: Provide a concise answer for each question. (Concept and code understanding)

[Generate 10 short-answer questions. Include the question, the expected answer, and a brief explanation/context.]

---

Part 3: Descriptive Questions (10 Questions)

Instructions: Explain the concepts or code logic in detail. (Concept and code understanding)

[Generate 10 descriptive/essay questions. Include the question and a sample model answer/explanation.]

---

Content Source:
The questions should be derived from the topics covered in notebooks/NB-computer_vision_intro.ipynb, which includes:

• Introduction to Convolutional Neural Networks (CNNs)
• Conv2D and MaxPooling2D layers
• Padding and Strides
• GlobalAveragePooling2D
• Training a CNN on MNIST
• The relevance of deep learning for small-data problems
• Data preprocessing (image_dataset_from_directory)
• TensorFlow Dataset objects
• Data augmentation

Formatting:

• Use Markdown cells for questions and text.
• You may use code cells if necessary to show code snippets for questions, but the primary format for the exam text should be Markdown.
• clearly label each section.
• Provide the "Answer Key" or "Solutions" within the same notebook, perhaps at the bottom or hidden in collapsible cells if possible, or just listed after each question block for study purposes (as requested: "Include correct answers and brief explanations").

Note: The user referenced a file dl-midterm_exam_final.ipynb for formatting style, but I do not have access to that specific file in the provided context. I will use a standard, clean, and well-structured Jupyter Notebook format that effectively presents the exam questions and answers.

This pull request was created as a result of the following prompt from Copilot chat.

Create a new Jupyter notebook file named notebooks/dl-final_exam.ipynb in the codingalzi/dlp2 repository.

This notebook will serve as a final exam based on the content of notebooks/NB-computer_vision_intro.ipynb.
The structure of the notebook should include the following sections:

Deep Learning Final Exam

This exam covers the material from the "Introduction to Computer Vision" notebook (NB-computer_vision_intro.ipynb).

---

Part 1: Multiple Choice Questions (10 Questions)

Instructions: Choose the best answer for each question. (Concept-focused)

[Generate 10 multiple-choice questions with 5 options each. Include the question, options, correct answer, and a brief explanation for each.]

---

Part 2: Short Answer Questions (10 Questions)

Instructions: Provide a concise answer for each question. (Concept and code understanding)

[Generate 10 short-answer questions. Include the question, the expected answer, and a brief explanation/context.]

---

Part 3: Descriptive Questions (10 Questions)

Instructions: Explain the concepts or code logic in detail. (Concept and code understanding)

[Generate 10 descriptive/essay questions. Include the question and a sample model answer/explanation.]

---

Content Source:
The questions should be derived from the topics covered in notebooks/NB-computer_vision_intro.ipynb, which includes:

• Introduction to Convolutional Neural Networks (CNNs)
• Conv2D and MaxPooling2D layers
• Padding and Strides
• GlobalAveragePooling2D
• Training a CNN on MNIST
• The relevance of deep learning for small-data problems
• Data preprocessing (image_dataset_from_directory)
• TensorFlow Dataset objects
• Data augmentation

Formatting:

• Use Markdown cells for questions and text.
• You may use code cells if necessary to show code snippets for questions, but the primary format for the exam text should be Markdown.
• clearly label each section.
• Provide the "Answer Key" or "Solutions" within the same notebook, perhaps at the bottom or hidden in collapsible cells if possible, or just listed after each question block for study purposes (as requested: "Include correct answers and brief explanations").

Note: The user referenced a file dl-midterm_exam_final.ipynb for formatting style, but I do not have access to that specific file in the provided context. I will use a standard, clean, and well-structured Jupyter Notebook format that effectively presents the exam questions and answers.

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.