Exploratory analysis part of the Data Mining course. All the utility functions are used from the ICML 2020 Workshop on Graph Representation Learning and Beyond repository.

We are also comparing our COPAC implementation with the ELKI implementation which is why the *.jar is part of this repository.

Just clone this repository and install packages with pip from the requirements.txt file. I altered some of the code from the TUDataset repo so that we don't have to deal with the correct version numnber of torch since we don't need torch for our uses anyway.

Clone the repo.

git clone https://github.com/chrisonntag/imdb-analysis-copac.git

(or use some UI client)

Create a virtual environement. This creates a directory env/ where all the dependencies will be installed.

python3 -m venv env

Choose the environment.

source env/bin/activate

Install all requirements.

pip install -r requirements.txt

Open Jupyter lab.

jupyter lab

Everything should work from hereon since all needed artifacts are part of this repo as well. You can find the data analysis notebook for the exploratory part and our COPAC implementation in /analysis/eda.ipynb.

Feel free to move the COPAC implementation into its own module.

Send me your GitHub username and i'll add you to the repository as a contributor. Clone the repository and switch to the development branch

git checkout development

From there you can create a new branch with

git checkout -b YourBranchName

Do all your changes here and push it to GitHub with

git push origin YourBranchName

We can then merge it to our development/main branch. Alternatively merge it locally into the development branch.