This repo contains all the code necessary to reproduce the work published in "Predicting students' difficulties from a piece of code". It does not contain the dataset, but we can provide it upon request for research purposes, just contact us (luciana.benotti AT unc.edu.ar). Our code can also be used with a different dataset, we can provide the required details by email. Our experiments are on students learning Haskell but the features used and the deep learning architecture are independent on the programming language of the dataset.
- First Step: Install Anaconda
- Second Step: Create a conda environment. After Anaconda installation is done. run the following command to create your Anaconda virtualenv named DEDA.
>> conda create -n DEDA python=3.6 anaconda - Third Step: activate conda environment and install requirements.
>> conda activate DEDA
>> pip install -r requirements.txt - Fourth Step: Run Notebook
>> jupyter notebook To be able to run notebook project, you need to put the datasets files that you want to analyse inside the datasets directory.