The official repository for the homework assignments of the lecture <Geometric Methods for High-Dimensional Data Analysis> instructed by Frank C. Park in Seoul National University.

• Syllabus

Schedule
Instructions for Settings
Homework 1
Homework 2
Homework 3

• Younghun Kim ([email protected], Head TA)
• Yonghyeon Lee ([email protected])
• Seungyeon Kim ([email protected])

For those unfamiliar with environment setup, Anaconda3 installation is recommended.

• For Windows user, download and install Anaconda3 from the following link: https://www.anaconda.com/products/distribution.
• For Linux user, download shell file from the following link: Anaconda3-2022.05-Linux-x86_64.sh. After that, execute the shell using the following command:

  chmod +x Anaconda3-2022.05-Linux-x86_64.sh      # change permission
  ./Anaconda3-2022.05-Linux-x86_64.sh             # execute shell file

After Anaconda3 installation, jupyter is automatically installed and you can run jupyter notebook with the following command:

jupyter notebook

Here's some commands for conda environment:

conda create -n {name} python=3.9   # create conda environment
conda activate {name}               # activate conda environment
conda deactivate {name}             # deactivate conda environment

If you want to use the created conda environment in jupyter notebook, you need to register the kernel. The command to register is following:

pip install ipykernel
python -m ipykernel install --user --name {name} --display-name "{name-in-jupyter}"

• {name} and {name-in-jupyter} can be replaced by your choices.

The project codes are tested in the following environment.

• python 3.9
• numpy
• matplotlib
• scikit-learn
• pytorch
• torchvision

To setup the environment except pytorch and torchvision, run the following script in the command line:

pip install -r requirements.txt

Pytorch and torchvision need to be installed separately using the method below.

Install PyTorch and torchvision from the following link: https://pytorch.org.

Follow the instructions in the HW1.ipynb file. After you complete and run the HW ipython file, send the result file to [email protected].
For problem 1~5, submit your answer sheet in the class.

Follow the instructions in the HW2.ipynb file. After you complete and run the HW ipython file, send the result file to [email protected].
For problem 1~4, submit your answer sheet to 302-413, or [email protected]

• In problem 5 of the pdf file, change $\mathcal{D} =$ { $(x_i,x_j)|x_i, x_j \in \mathcal{X}$ } to $\mathcal{D} =$ { $(x_i,x_j)$ }

Follow the instructions in the HW3.ipynb file. After you complete and run the HW ipython file, send the result file to [email protected].
For problem 1~4, submit your answer sheet to 302-413, or [email protected]