You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thanks for your awesome work! Could you please release the code to evaluate minigpt4 and llava? I try to reproduce the result of minigpt4 but the results on some data type are much lower than you reported:
Accuracy for each data type:
Data type 1: 52.72%
Data type 2: 48.66%
Data type 3: 36.42%
Data type 4: 41.10%
Data type 5: 44.50%
Data type 6: 33.79%
Data type 7: 51.55%
Data type 8: 41.99%
Data type 9: 15.29%
Total accuracy: 43.31%
I want to check what's wrong with my evaluation code. Thanks in advance :)
I believe that releasing this code could significantly motivate the community to utilize your benchmark :)
The text was updated successfully, but these errors were encountered:
We greatly appreciate your interest in our work and apologize for the delayed response. We will continue to provide updates and plan to upload the evaluation code for each model in the future.
Thanks for your awesome work! Could you please release the code to evaluate minigpt4 and llava? I try to reproduce the result of minigpt4 but the results on some data type are much lower than you reported:
I want to check what's wrong with my evaluation code. Thanks in advance :)
I believe that releasing this code could significantly motivate the community to utilize your benchmark :)
The text was updated successfully, but these errors were encountered: