weizequan / dmin Goto Github PK

First of all, thank u for ur work and the valuable research you've shared with the community.

Recently, I attempted to reproduce the results presented in your article using the provided code and dataset. While I was able to successfully reproduce the experiment, I found that the accuracy I achieved (77.6%) was lower than the accuracy reported in your article (79.6%).

I have carefully verified my code and configurations to ensure that I followed the implementation details outlined in your repository and paper. However, despite my efforts, I couldn't achieve the reported accuracy.
In light of this, I am reaching out to inquire if you could kindly provide access to a pre-trained model used in the article. Having access to the pre-trained model would significantly help the community to validate and compare results more accurately.

If possible, could you please share the pre-trained model or any additional insights that might help me better reproduce the reported accuracy?

Can you please tell me when will you be releasing the code?

come on! come on !