TempKB is a comprehensive collection of knowledge graph data designed to train ML models on knowledge graph completion and reasoning tasks. This is the Event version, which contains only data where the time information is presented.
The dataset is organized into the following main components:
Train Set: 1373083 instances for training models.
Validation Set: 63421 instances for validating model performance.
Test Set: 63530… See the full description on the dataset page:
https://huggingface.co/datasets/ESITime/TempKB-Event.
