question about evaluation and chatgpt version?

Question

question about evaluation and chatgpt version?

Opened this issue 4 months ago · 5 comments

Thank you for the open source. May I ask when your experiment took place? Since different versions of ChatGPT can lead to different evaluation results, how did you compare it with Video-ChatGPT?

Answer 1 · 2024-07-27T06:40:05.000Z

also,The 114th line of code in save_features shows that the feature dimension after merge is 200 instead of 300. It should be replaced with the commented content.

Answer 2 · 2024-08-19T17:50:17.000Z

@vhzy Did you get the pretrained model weights that the authors provide? I can't even find the weights in the huggingface.

Answer 3 · 2024-08-26T12:30:10.000Z

@vhzy Did you get the pretrained model weights that the authors provide? I can't even find the weights in the huggingface.

Hello, have you found the weights now, or do we need to train the model ourselves?

Answer 4 · 2024-08-26T13:33:44.000Z

@vhzy Did you get the pretrained model weights that the authors provide? I can't even find the weights in the huggingface.

Hello, have you found the weights now, or do we need to train the model ourselves?

not yet

Answer 5 · 2024-11-16T10:24:19.000Z

@vhzy Did you get the pretrained model weights that the authors provide? I can't even find the weights in the huggingface.

May I ask if you have obtained the author's pre trained model and weights.