Code for Tracing Knowledge Instead of Patterns: Stable Knowledge Tracing with Diagnostic Transformer (accepted at WWW '23).
Cite this work:
@inproceedings{yin2023tracing,
author = {Yin, Yu and Dai, Le and Huang, Zhenya and Shen, Shuanghong and Wang, Fei and Liu, Qi and Chen, Enhong and Li, Xin},
title = {Tracing Knowledge Instead of Patterns: Stable Knowledge Tracing with Diagnostic Transformer},
year = {2023},
isbn = {9781450394161},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3543507.3583255},
doi = {10.1145/3543507.3583255},
booktitle = {Proceedings of the ACM Web Conference 2023},
pages = {855โ864},
numpages = {10},
keywords = {contrastive learning, knowledge tracing, DTransformer},
location = {Austin, TX, USA},
series = {WWW '23}
}
poetry install
Train DTransformer with CL loss:
python scripts/train.py -m DTransformer -d [assist09,assist17,algebra05,statics] -bs 32 -tbs 32 -p -cl --proj [-o output/DTransformer_assist09] [--device cuda]
For more options, run:
python scripts/train.py -h
Evaluate DTransformer:
python scripts/test.py -m DTransformer -d [assist09,assist17,algebra05,statics] -bs 32 -p -f [output/best_model.pt] [--device cuda]