-
## 🐛 Bug
running link_predict example as:
python3 link_predict.py -d wn18 --gpu 0
produces an error:
ValueError: Feature size must be a multiplier of num_bases (36)
with full stack t…
-
In line 1315 of sampler.cc, there's a check as follows:
CHECK_EQ(num_chunks * neg_sample_size, global_neg_vids.size());
I'm wondering why you'd want num_chunks*neg_sample_size == global_neg_vids…
-
please help me. I want to use command that dglke_predict --data_path data/wn18/ --model_path ckpts/TransE_l2_wn18_0/ --format 'h_r_t' --data_files head.list rel.list tail.list --score_func logsigmoid…
-
In knowledge_graph.py,
`np.random.RandomState, self._global_rs.randint(2 ** 32) `
but the maximum of positive int32 integer is 2**-1
-
Hi! Sorry if this is a noob question.
In the paper, the authors mention 'at each main iteration of the algorithm, the embedding vectors of the entities are first normalized'. Is this missing from `Tr…
-
please I would like to know how to interpret this result
-
Right now, the `margin` and `adversarial_temperature` parameters do not have a default for NSSA loss. We need to pick reasonable ones and set it.
-
Providing a basic template of building a Dataset object for DGL. The recommend logic of processing the Dataset is as the following:
1) Check whether there is a cached Dataset (already processed and …
-
Hi!
I have a question about datasets.
After calling `python kbc/process_datasets.py` and downloading the datasets, I realized that triples are in the "index" format as such:
a triple --> `2431 89…
-
Hi,
Thanks for this wonderful framework!
I was wondering whether you have reported your official results (with hyper-parameters) on the WN18 or FB15K using this framework. I am trying to repro…