This repository contains my implementation of Deep Knowledge Tracing for Udacity's Capstone Project.
Build and train a LSTM network to predict the probabilities of a student answering correctly a problem not yet seen by him using the ASSISTments Skill-builder data 2009-2010 public dataset.
This is the best results obtained by comparing the validation loss between each network configuration attempted.
| Test Data (%) | AUC |
|---|---|
| 20% | 0,85 |
The results, configuration and model's weights of each attempt can be found in the "Log" folder.
You'll need Python 3.x x64 to be able to run theses projects.
If you do not have Python installed yet, it is recommended that you install the Anaconda distribution of Python, which has almost all packages required in these projects.
You can also install Python 3.x x64 from here
- Clone the repository and navigate to the downloaded folder.
git clone https://github.com/lccasagrande/Deep-Knowledge-Tracing.git
cd Deep-Knowledge-Tracing
-
Install required packages:
- If you already has TensorFlow installed, type:
pip install -e .- If you want to install with TensorFlow-GPU, follow this guide to check the necessary NVIDIA software on your system. After that, type:
pip install -e .[tf_gpu]- If you want to install with Tensorflow-CPU, type:
pip install -e .[tf] -
Navigate to the src folder and open the notebook.
cd src
jupyter notebook DKT.ipynb
- The most important step: Have fun !!!
If you have any questions or find a bug, please contact me!