The performance on the XING dataset

[hjlin0515]

I have ran the "python train_last.py --data=xing --mode=transformer --user_ --adj=adj_all --dataset=all_data --hiddenSize=100“ for many times and evaluation results like Recall@5, Recall@10 can not reach the number mentioned in the paper.
Is there anything wrong in the hyper-parameter settings or something else? Thank you!

[ZM7]

I have ran the "python train_last.py --data=xing --mode=transformer --user_ --adj=adj_all --dataset=all_data --hiddenSize=100“ for many times and evaluation results like Recall@5, Recall@10 can not reach the number mentioned in the paper.
Is there anything wrong in the hyper-parameter settings or something else? Thank you!
Thank you for your attention to our work, I set the hyper-parameter "buffer_size" as 20000 and "batch_size" as100.