daiquocnguyen / ConvKB

A Novel Embedding Model for Knowledge Base Completion Based on Convolutional Neural Network (NAACL 2018) (Pytorch and Tensorflow)
Apache License 2.0
203 stars 39 forks source link

Issue try running the code #2

Closed yuhaozhang97 closed 6 years ago

yuhaozhang97 commented 6 years ago

Hi there,

When I tried running "python train.py --embedding_dim 100 --num_filters 50 --learning_rate 0.000005 --name FB15k-237 --useConstantInit --model_name fb15k237", It seems the program does not create a new "runs" directory. Do you guys know why?

daiquocnguyen commented 6 years ago

You should use "--run_folder" to specify the path to the training folder.

lileilai commented 5 years ago

Hi, I read your paper published in AAAI18, i notice that the performance of TransE is below: WN18 RR: 3384 0.226 50.1 FB15K-237 347 0.294 46.5 But, i can not generate this result by the existing open-source code Could you mind share your TransE model. Thank you ~~~

daiquocnguyen commented 5 years ago

As mentioned in our paper, I employed a TransE implementation available at: https://github.com/datquocnguyen/STransE .