-
**Is your feature request related to a problem? Please describe.**
Find a cloud storage service that 1) allows anyone with a link to upload files (something like the Dropbox file request); 2) automat…
-
Suddenly, everything stopped working 😱 presumably because of the changes to torch.sparse.
Particularly, I am on PyTorch 1.10, master branch of PyKEEN and torch-ppr 0.0.5.
Problem 1: the `allclose…
-
### Describe the bug
I get a bug: Base representations comprise only 475 representations, but at least 948 are required.
### How to reproduce
```python
import torch.nn
from pykeen.model…
-
During running time, it appears an error, do you know how to solve it?
```
python train_transr_FB15K237.py
Input Files Path : ./benchmarks/FB15K237/
The toolkit is importing datasets.
The total o…
-
### Describe the bug
It seems like Node Piece isn't work with PairRE and TripleRE interaction
```python-traceback
[/usr/local/lib/python3.7/dist-packages/pykeen/models/nbase.py] in _prepare_repr…
-
### Describe the bug
The current implementation of the `AdversarialBCEWithLogits` degrades the performance with the growing number of negative samples (see the illustration) - presumably because of…
-
### Describe the bug
Transe experiment files not in the correct folder, they are placed with convkb and therefore cannot be executed
### How to reproduce
```
pykeen experiments reproduce tr…
-
为了观察、调试模型，我修改了src/neuralkg/lit_model/KGELitModel.py和src/neuralkg/eval_task/link_prediction.py中的代码。但是在使用main.py训练、测试的过程中，这些修改的代码都好像没有执行，而是保持修改之前的状态。请问这种现象出现的原因是什么，以及我如果想把这些修改的代码应用于训练，应该如何去做？
-
### Describe the bug
After #888 changed the `init_phases` function, I get the error while creating a standard documented NodePiece model
```
phases = 2 * np.pi * torch.rand_like(torch.view_as_c…
-
**Describe the bug**
As titled, only around 5 datasets passed loading.
Possible because of https://github.com/Graph-Learning-Benchmarks/gli/pull/325
```
=========================================…