EduStudio v1.1.3

⚡️ New

• add IdentifiablityEvalTPL including Identifiability Score (IDS) metric that is proposed in [WWW 2024] IC-IDM
• add Matrix Factorization (MF) model in the CD model section

⬆️ Improvements

• [optimize] get_stu_status API is iterative in TrainTPL

EduStudio v1.1.1

⬆️ Improvements

• change the dataset download source from GitLab to HuggingFace
• add external datasets.yaml support for downloading datasets automatically

🔧 Bug Fixes

• fix DT_INFO_DICT option in frame_cfg (specified in settings.py) not working bug

EduStudio v1.1.0

⚡️New Features

• Add DCD model: [NeurIPS 2023] Disentangling Cognitive Diagnosis with Limited Exercise Labels
• Add FairCD model: [SCIS 2023] Understanding and Improving Fairness in Cognitive Diagnosis
• Add AdversarialTrainTPL for adverarial training
• Add FairnessEvalTPL for fairness evaluation, including DP, EO and FCD metrics.
• Add R2M operations of SLP_Math and SLP_English datasets, which are from SLP dataset.
• Add EDUSTUDIO_VERSION option in settings.py (i.e., frame_cfg)

⬆️ Improvements

• Divide functionality of M2C_BuildSeqInterFeats ➡️ M2C_RandomDataSplit4KT & M2C_BuildSeqInterFeats
• Rename EduTrainTPL ➡️ GeneralTrainTPL
• Rename BinaryClassificationEvalTPL ➡️ PredictionEvalTPL
• Rename CognitiveDiagnosisEvalTPL ➡️InterpretabilityEvalTPL
• Rename some atomic operations: Cpt ➡️ KC
• Enrich Docs

🔧 Bug Fixes

• K-fold split for KT

EduStudio v1.0.0

This release is meant to fix bugs:

• Fix bugs in IRT、MIRT
• Fix bugs in DeepIRT、DKVMN、DKT+、QIKT、QDKT

EduStudio v1.0.0-beta4

• Fixed Data Cache bug
• Fixed UnifyConfig Bug
• Add fixed random seed for each fold training

EduStudio v1.0.0-beta3

• Fixed EduTrainTPL bugs

EduStudio v1.0.0-beta2.1

• Fixed EduTrainTPL bugs

EduStudio v1.0.0-beta2

• Fixed Bugs: init_all, parse_cfg, ckt
• Add Option: DT_INFO_DICT

EduStudio v1.0.0-beta1

• Fix many bugs
• Update docs
• Optimize some models

EduStudio v1.0.0-alpha5

Fixed Bugs