search

Pointcept / GPT4Point

[CVPR'24 Highlight] GPT4Point: A Unified Framework for Point-Language Understanding and Generation.
https://gpt4point.github.io/
MIT License
304 stars 19 forks source link

Some issues about Evaluation #18

Closed Fugtemypt123 closed 4 months ago

Fugtemypt123 commented 4 months ago

Thanks for your interesting work! I have met some problems when I tried to evaluate your model. I have downloaded the Data, Annotations, and Checkpoints following your guidance in README.md. However, when I run the evaluation code, I got all 0.000~0.001 in Bleu, METEOR, Rouge, CIDEr score. Maybe it is not correct? I even tried training the second stage on my computer, but it also didn't work. Since I followed your guidance and your code step by step, I didn't know where I got wrong. What do you think about that? I am looking forward to your reply.

Qi-Zhangyang commented 4 months ago

I will fix it as soon as possible.

On Tue, Apr 23, 2024 at 1:22 PM Fugtemypt123 @.***> wrote:

Thanks for your interesting work! I have met some problems when I tried to evaluate your model. I have downloaded the Data, Annotations, and Checkpoints following your guidance in README.md. However, when I run the evaluation code, I got all 0.000~0.001 in Bleu, METEOR, Rouge, CIDEr score. Maybe it is not correct? I even tried training the second stage on my computer, but it also didn't work. Since I followed your guidance and your code step by step, I didn't know where I got wrong. What do you think about that? I am looking forward to your reply.

— Reply to this email directly, view it on GitHub https://github.com/Pointcept/GPT4Point/issues/18, or unsubscribe https://github.com/notifications/unsubscribe-auth/AWVKZMPU6LK2VJUB5R6VE6LY6XVYFAVCNFSM6AAAAABGUCSM7WVHI2DSMVQWIX3LMV43ASLTON2WKOZSGI2TOOJXGEYDSOI . You are receiving this because you are subscribed to this thread.Message ID: @.***>

Fugtemypt123 commented 4 months ago

I will fix it as soon as possible. … On Tue, Apr 23, 2024 at 1:22 PM Fugtemypt123 @.> wrote: Thanks for your interesting work! I have met some problems when I tried to evaluate your model. I have downloaded the Data, Annotations, and Checkpoints following your guidance in README.md. However, when I run the evaluation code, I got all 0.000~0.001 in Bleu, METEOR, Rouge, CIDEr score. Maybe it is not correct? I even tried training the second stage on my computer, but it also didn't work. Since I followed your guidance and your code step by step, I didn't know where I got wrong. What do you think about that? I am looking forward to your reply. — Reply to this email directly, view it on GitHub <#18>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AWVKZMPU6LK2VJUB5R6VE6LY6XVYFAVCNFSM6AAAAABGUCSM7WVHI2DSMVQWIX3LMV43ASLTON2WKOZSGI2TOOJXGEYDSOI . You are receiving this because you are subscribed to this thread.Message ID: @.>

Thanks!

Qi-Zhangyang commented 4 months ago

I have corrected the error and now you can run the eval code directly to get the result.