Install requirements
pip3 install -r requirements.txtGet data
git lfs pullRun
python3 manage.py --model [GCAKE] --dataset [FB15K-237, WN18RR]
./train_all_dataset.sh [GCAKE]Get raw data:
./data_downloader.shGet trainable data
./data_preprocessing.sh
- Knowledge Graph with Entity Description
- xrb92/DKRL: Representation Learning of Knowledge Graphs with Entity Descriptions (AAAI'16) - FB15k with description, FB20k-new
- villmow/datasets_knowledge_embedding: Datasets for Knowledge Graph Completion with textual information about the entities - FB15K, FB15k-237, WN18, WN18RR
- Knowledge Graph
- Relation Extraction/Classification
- davidsbatista/Annotated-Semantic-Relationships-Datasets
- SemEval 2007 Task 4
- SemEval 2010 Task 8
- ReRelEM
- BioNLP Shared Task
- DBpediaRelations-PT
- davidsbatista/Annotated-Semantic-Relationships-Datasets
head\trelation\ttail\t[cls] head_description [sep] relation_text [sep] tail_description [sep]\n
data/FB15K-237/*.tsv
- State-of-the-art table for Link Prediction on FB15k
- State-of-the-art table for Link Prediction on FB15k-237
data/WN18RR/*.tsv
- State-of-the-art table for Link Prediction on WN18
- State-of-the-art table for Link Prediction on WN18RR
all-in-one json
- Train: 38720
- Valid: 1000
- Test: 1000
each relation (total 7) a file
- Train (0-140): 140 x 7
- Test (141-?)
- relation 1: 220
- relation 2: 218
- relation 3: 233
- relation 4: 221
- relation 5: 211
- relation 6: 212
- relation 7: 214
all-in-one txt
- Train (1-8000): 8000
- Test (8001-10717): 2717
all-in-one xml
- Total
- Element 'DOC': 129
each sample three file (rel, a1, txt)
- Total: 798
all-in-one txt
grep -o 'SENTENCE' data/DBpediaRelations-PT/DBpediaRelations-PT-0.2.txt | wc -l
- Total: 98023