-
I've cloned the repository and downloaded the models and data. How can I reproduce the CQDA numbers for e.g. 2i?
I tried:
```bash
PYTHONPATH=. python cli/cq
da-cli.py --checkpoint_path "./model…
-
How does your model works on the FB15K-237 Dataset, as mentioned by your paper?
-
I ran python main.py --train --test --data fb15k-237 --max_path_len 4 --model fb15k-237 --gpu 2 --get_rule --topk 500 and I get this. I dont get such fb15k-237/fb15k-237_500_2.txt what should I do?
!…
-
Are there any other hyperparameter settings for different datasets to reproduce their performance in the original paper,such as FB15K_237? I tried to use the hyperparameters for FB15K to train FB15K_2…
-
Hi @rajarshd, @shehzaadzd,
Using the config for kinship and fb15k-237, we only get the following results, which are quite different from the ones in the paper.
```
FB15k-237
[ Hits@1: 0.1511…
-
### 🛠 Proposed Refactor
The datasets FB15k_237 and RelLinkPredDataset implement the same FB15k-237 relational link prediction dataset.
### Suggest a potential alternative/fix
FB15k-237 seems to be…
-
Could you please share the FB15k-num, FB15k-237-num used in the paper of "KBLRN: End-to-End Learning of Knowledge Base Representation with Latent, Relational, and Numerical Features"?
I tried to re…
-
I'm sorry to bother you. I want to ask some questions about Inductive Reasoning. When I was doing the experiment on Ind_FB15k-237-V2 dataset, the result of Hits@10 was always 0. Is it convenient to pr…
-
Namespace(dropout=0.2, evaluate_every=500, gpu=-1, grad_norm=1.0, graph_batch_size=30000, graph_split_size=0.5, lr=0.01, n_bases=4, n_epochs=10000, negative_sample=1, regularization=0.01)
load data f…
-
After I configured the environment, "AttributeError: 'tuple' object has no attribute 'ndim'" appeared when training on the WN18RR and FB15k-237 data sets. The error code is in line 107 of main.py (tra…
sjlmg updated
2 months ago