return-sleep

Results 16 comments of return-sleep

> ![Screenshot from 2023-12-27 14-39-31](https://private-user-images.githubusercontent.com/31506965/292965131-87f5a3a9-bbbc-4092-83a2-00a433d8557c.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.rryvHiTmKWg3hrb5HJuMe5-80Pdd5Jf0z4sFjF8b6Nc) Apparently you have to ask the author :) This is taken from v-d4rl paper which opens their dataset. https://github.com/conglu1997/v-d4rl Thanks! V-D4RL provides a good benchmark...

same question~ how can I control the translation and rotation of a single object? use the T ?

> Thank you very much! I will have a try. Excuse me, have you solved it? I have tried the solution but haven't solved the problem.

I guess it may be caused by ConditioningMethod.grad_and_value ([here](https://github.com/DPS2022/diffusion-posterior-sampling/blob/82f313357accce87bd11c0f8d845ddb01ffee4e3/guided_diffusion/condition_methods.py#L31)), where batch size influences the gradient norm and then influences the guidance step. Another question is how can we replace the...

May I ask if you have successfully trained the dreamerv3 agent. I'm curious what the final loss of each component looks like,such as image, reward or cont. When the reconstructed...