Dear students of CS-442,
We will have the second Graded Exercise on Tuesday next week (May 13th) from 10:15 to 11:45!
For those who need additional time, their time has been adjusted accordingly. The topic will be shape from X.
Room assignment:
- Until Gibbons: INM200
- Goffinet to Santelmo: INM202
- Sbicego to the end: CO 120 (this is quite far from INM, make sure to be in the right place)
Additional Information:
- You will have to perform the tasks on your computer. Please install the following libraries: numpy, matplotlib, scipy, scikit-image.
- The total grade for this exercise is worth 10% of your final grade for the course.
- It will be held in-person at the time of our usual exercise sessions. Any submissions made while not present in the lab will be discarded.
- You will be provided with a jupyter notebook with the exercises in it.
- Your goal is to fill in the cells and make algorithms work. Then answer related questions.
- When the time is up, you will have to submit your notebooks to the provided link in Moodle. The time limit is strict: make sure to upload something before time runs out. Moodle will not accept late submissions.
- Use of any internet resources, except documentation of libraries above and Moodle lectures and exercises, is strictly prohibited! This includes ChatGPT, copilot, etc.
- Please bring a power extender if you need to charge your laptop during the session.
Comments on Graded Exercise 1:
We have also released the grades for the previous graded exercise. The average is fairly high at 78/100. Some general feedback:
- Q2 and Q8: It is often a good idea to specify the axis along which to compute the mean
- Q6: Be mindful not to allow identical centroids in the initialization of K-Means
- Q3 and Q8: Make sure to return the correct shape
Course Recording:
There has been some issues with some lecture recordings in which the sound was not properly captured. You should feel encouraged to attend the lectures in person, but if it is not possible you may also watch the corresponding lectures from last year as the content is mostly identical. This issue has been fixed now.
Feel free to ask questions here and on the forum!
See you all!
TA team