awslabs/dgl-ke

IndexError: list index out of range when training on raw user defined knowledge graph

Maester-Khris opened this issue · 4 comments

Hello i try to train with my own dataset, but i have one error after the start of triple reading

! sudo pip3 install dgl==0.4.3
#sudo pip3 install dglke
! git clone https://github.com/awslabs/dgl-ke.git
%cd dgl-ke/python
! sudo python3 setup.py install
! pip install ogb

%%time
!DGLBACKEND=pytorch dglke_train
--dataset mydatas
--model_name TransE_l2
--data_path /content/mydataset/
--format raw_udd_hrt
--data_files train.tsv valid.tsv test.tsv
--batch_size 1000
--neg_sample_size 200
--hidden_dim 400
--gamma 19.9
--lr 0.25
--max_step 24000
--log_interval 100
--batch_size_eval 16 -adv
--regularization_coef 1.00E-09
--test
--save_path /content/mydataset/
--neg_sample_size_eval 10000

Traceback (most recent call last):
File "/usr/local/bin/dglke_train", line 33, in
sys.exit(load_entry_point('dglke==0.1.0.dev0', 'console_scripts', 'dglke_train')())
File "/usr/local/lib/python3.7/dist-packages/dglke-0.1.0.dev0-py3.7.egg/dglke/train.py", line 85, in main
File "/usr/local/lib/python3.7/dist-packages/dglke-0.1.0.dev0-py3.7.egg/dglke/dataloader/KGDataset.py", line 762, in get_dataset
File "/usr/local/lib/python3.7/dist-packages/dglke-0.1.0.dev0-py3.7.egg/dglke/dataloader/KGDataset.py", line 527, in init
File "/usr/local/lib/python3.7/dist-packages/dglke-0.1.0.dev0-py3.7.egg/dglke/dataloader/KGDataset.py", line 595, in load_entity_relation
IndexError: list index out of range

I need some help to solve this

It seems something wrong here: https://github.com/awslabs/dgl-ke/blob/master/python/dglke/dataloader/KGDataset.py#L594

Can you check if the input file (edges) is correct?

Hello,
Here are the files i used
valid
test
train

Hello i changed the content of files and precise the delimiter to , in the command traing