pykeen / pykeen

🤖 A Python library for learning and evaluating knowledge graph embeddings

https://pykeen.readthedocs.io/en/stable/

pykeen/pykeen Issues

Add a documentation section about constrainer vs. normalizer vs. regularizer
Closed 11 days ago2
Resolve FutureWarning with `torch>=2.5`
Updated 11 days ago
Embeddings for unseen nodes
Updated 15 days ago1
PyKEEN 1.11.1 Release
Updated 16 days ago1
Getting Embeddings of the Entity and Relations
Closed 9 months ago13
Torch 2.4 and 2.5 support
Closed 19 days ago
Evaluation loop of pipeline crashes: torch_max_mem.api: Encountered tensors on device_types={'cpu'} while only ['cuda'] are considered safe for automatic memory utilizaiton maximization.
Closed a month ago3
Change front page of docs to explain what problems pykeen can help solve
Updated 2 months ago
Cannot provide custom `entity_representations_kwargs` or `relation_representations_kwargs` via `model_kwargs` to `pipeline`
Updated 2 months ago
"Class Inheritance Diagram" of documentation about prediction in pyKeen does not show correctly
Closed 3 months ago2
Deprecated get_tail_prediction_df function in Understanding the Evaluation tutorial
Closed 3 months ago2
ImportError: cannot import name 'get_tail_prediction_df' from 'pykeen.predict'
Closed 3 months ago3
Splitting a triples factory could not find a coverage of all entities and relations
Closed 3 months ago5
Hit@10 does not match with the evaluator result
Closed 3 months ago
List of mode's best performance config for each dataset
Closed 3 months ago
`LabelBasedInitializer` sample code may be missing an argument `triples_factory`
Closed 5 months ago2
issue in "Full Inductive LP Example"
Closed 5 months ago4
pip installation broken (1.8.1)
Closed 5 months ago2
create_inverse_triples=True fails for the ILPC datasets
Closed 5 months ago
tqdm progressbar is still shown although setting `use_tqdm=False`
Closed 5 months ago
Reason for omitting validation inference triples from filtering when doing test evaluation in inductive lp example
Updated 5 months ago2
Cosine Annealing with Warm Restart LR Scheduler recieving an unexpected kwarg `T_i`
Closed 6 months ago1
OOM Crash on MPS/Apple silicon
Closed 6 months ago2
TransE - CUDA out of memory
Updated 6 months ago3
Shape Mismatch upon initializing pretrained ComplEx embeddings
Closed 7 months ago2
EarlyStopper: show progress bar
Closed 8 months ago
Unable to reproduce TransE experiment
Closed 8 months ago
Usage of a custom encoder
Closed 8 months ago6
Bug in wandb result tracker
Closed 9 months ago1
RGCN Hyper parameter optimization error
Closed 9 months ago1
Importing model_resolver
Closed 9 months ago2
MatKG
Updated 9 months ago1
HPO_Pipeline fails on AutoSF models
Updated 9 months ago1
from pykeen.pipeline import pipeline, pipeline issue
Closed 10 months ago3
Evaluating metrics on many subsets with multiple models
Closed 10 months ago2
Training loop does not update relation representations when continuing training
Updated 10 months ago2
QuatE: GPU memory is not released per epoch
Updated 10 months ago3
RGCN RuntimeError: trying to backward through graph a second time. (has parameters but no reset_parameters)
Closed a year ago2
Possible issue with model evaluation when using datasets with inverse triples
Updated a year ago1
Want to train a model without any evaluate or test dataset
Closed a year ago1
Question about the use of `create_inverse_triples`
Closed a year ago2
AttributeError: 'Module' object has no attribute 'get'
Closed a year ago2
TypeError: MarginPairwiseLoss.__init__() missing 2 required positional arguments: 'margin' and 'margin_activation'
Closed a year ago2
Protobuf requirement contriant
Closed a year ago5
TypeError: DoubleMarginLoss.__init__() got an unexpected keyword argument 'margin_positive'
Closed a year ago1
CKG(Clinical Knowledge Graph) dataset url expired
Updated a year ago5
Resuming Training gives CheckpointMismatchError
Updated a year ago10
NodePiece: A potential issue with anchors not being sorted according to distance from target node
Closed a year ago4
The GCN code will stop while running
Closed a year ago
ValueError: All arrays must be of the same length, While predicting a relation with CompGCN
Closed a year ago5