- Timeframe: It will open Monday, 5/1, at 12:01 AM and close Wednesday, 5/3, at 11:59 PM.
- The exam will close at 11:59 pm regardless of when you started.
- It assesses the same thing as Exam 3.
- But it will be different than the original and practice exams.
- You may use things that you have learned that were not in the modules that this exam is testing, but you can answer it without knowing any modules beyond what this exam is testing.
- The data sets will be different.
- There is no regrade window due to time constraints.
- All other information is similar to Exam 3’s. Such as getting the files, Gradescope, Sakai, asking for help, grading policy, etc.
- Reminder: You do not need to do both parts. You can do only one part if you wish. You must do ALL of the questions in that part, though. We will take the max score per part.
Month: April 2023
Exam 3 Logistics
- Modules covered: 7, 8, and 9
- Practice Exam (Part1 – LINK, Part2 – LINK)
- Timeframe: It will open Thursday, 4/20, at 12:01 AM, and close Saturday, 4/22, at 11:59 PM.
- The exam will close at 11:59 pm regardless of when you started.
- The exam will be take-home. It is open book, open note, open internet, but closed to people and AI tools (such as ChatGPT).
- This means you cannot receive help on this exam from anyone, including (but not limited to) communicating with a person while taking the exam, such as asking someone through the Internet (like stackoverflow) to receive help.
- In addition, you cannot give a question on the exam to an AI tool and ask it to generate an answer.
- Your submission must represent your own work only and is your evidence that you have mastered the material.
- Like prior Exams, it consists of 2 parts. However, each part has a time limit of 2.5 hours. Both parts will have data sets, and they will be different.
- Note that each part is 30 minutes longer than the prior exam parts.
- Note about the ESNU grading:
- There are ~100 points possible and fewer than 10 questions. The number of points earned are evenly distributed across the problems based on the number of concepts they are testing. The rubric to point conversation ensures that earning an E or S on all problems means an A. While a single U means an A is very unlikely, which is reasonable since a U on a problem clearly shows a lack of mastery of at least some content for this exam.
- All other information is similar to Exam 1 Part 2. Such as getting the files, Gradescope, Sakai, asking for help, grading ESNU policy, etc.
- Grading Clarification: A simple copy+paste and find+replace replacement from the practice exam is considered a Satisfactory answer. You must use your own words or elaborate beyond the practice exam text to show an Exemplary level of content mastery.
Exam 2 Retake Logistics
- Timeframe: It will open Wednesday, 4/19, at 12:01 AM, and close Saturday, 4/22, at 11:59 PM.
- Note it is open one day earlier since Exam 3 will take longer.
- The exam will close at 11:59 pm regardless of when you started.
- It assesses the same thing as Exam 2.
- You may use things that you have learned that were not in the modules that this exam is testing, but you can answer it without knowing any modules beyond what this exam is testing.
- The data sets will be different.
- You do not need to do both parts. You can do only one part if you wish. You must do ALL of the questions in that part, though. We will take the max score per part.
- All other information is similar to Exam 1 Part 2. Such as getting the files, Gradescope, Sakai, asking for help, grading policy, etc.
- Prepare (due M 4/10)
- Content below
- Sakai quizzes
- Peer Instructions – See on the class forum
- Homework (due Sun 4/16) [Link]
- There are no worked examples
Content
10 Deep Learning
- Neural Networks and Applications (16 min.)
- Forward Propagation (10 min.)
- Gradient Descent (14 min.)
- Back Propagation (11 min.)
- Convolutional Neural Network (15 min.)
- Introducing Pytorch (23 min.)
Optional Supplements
Pytorch
Unlike most other libraries for this course, Pytorch is not included in the basic Anaconda installation. To use Pytorch, we suggest you choose one of two options.
- Install Pytorch locally (for free). You can see the directions on the website: Select the stable build, your operating system, Conda (for Anaconda), Python, and CPU to see install directions for your particular setup. (CUDA is used to support hardware acceleration with NVIDIA graphics cards and is not necessary for this course).
- Use Pytorch in a Jupyter notebook in the cloud (also for free). The easiest way to do this if you have a Google account is with a Google colab notebook; Pytorch will already be available to you in this cloud environment.
You can find the official Pytorch documentation here. Of particular note are the Pytorch tutorials, including Pytorch recipes which serve as small examples of common tasks.
Book
The deep learning book is available free online and is authored by some of the leading experts in machine learning with deep artificial neural networks. It is very detailed and in-depth and is purely for those who are interested in learning more about deep learning theory now or in the future; you do not need to read the book for this course.