-
Hello and thank you for your work.
I would like to ask a question regarding the experimental results reported in the paper "OpenKE: An Open Toolkit for Knowledge Embedding", found here: https://ww…
-
Hello,
I am trying to organize all triples into an actual KB graph in networkx. When I tried to do so, I double checked the number of entities and it seems there is some discrepancy between what is…
-
it seems that there is no L2-norm constraints of the entities embedding in the transE code implementation, while the constraints is recommended in the transE paper, for what consideration you made thi…
-
`
def parse_line(line):
line = line.strip().split()
sub = line[0]
rel = line[1]
obj = line[2]
val = [1]
if len(line) > 3:
if line[3] == '-1':
# pri…
-
I plus my datasets in /data and then python datasets/process_datasets.py
but I still need change some code in kbc/learning/learn.py, especially the line102 dataset={'FB15K': lambda : big.FB15KDataset…
-
**SUMMARY**
a) I'm getting started, trying to test example_train_transe.py for both OpenKE TF and PyTorch
b) Tensorflow seems OK; but maybe there is a issue in the PyTorch version??
c) There may be…
drcdr updated
4 years ago
-
### Description
I am trying to replicate the performance results from the documentation but getting a runtime error when using early stopping with FB15k-237.
I have used early stopping with other…
-
**Background and Context**
The datasets will have to be hosted somewhere such that the automated downloader can access them. Google drive and Drop box were investigated but they required additional …
-
Dataset | FB15k | FB15k-237 | wn18 | wn18rr
MRR | .797 ± .001 | **.949 ± .000** | **.337 ± .001** | .477 ± .001
The MRR of datasets FB15k-237 and wn18 should be swapped with each other.
-
**Description**
We must add in docstrings:
- stats on how many tripels in each split, how many distinct entities and relations
- warning boxes for datasets with missing entities (FB15k-237, WN18…