EricGuo5513 / HumanML3D

HumanML3D: A large and diverse 3d human motion-language dataset.
MIT License
767 stars 76 forks source link

motion_representation.ipynb to check if it's alight with the given reference #138

Open Daye-Lee18 opened 2 months ago

Daye-Lee18 commented 2 months ago

Hello,

I was running the motion_representation.ipynb notebook, but I encountered an issue where the difference between an existing reference and a newly made reference is not zero.

Specifically, they are 0.00322 and 0.0185, respectively, as shown in the attached photo.

Is it okay to proceed to the next step for preprocessing the data, or should I restart the download process?

If I shouldn't proceed, what could be the reason for this issue?

Thank you.

Screenshot 2024-07-11 at 5 09 20 PM
EricGuo5513 commented 2 months ago

It’s okay. It’s quite small.

On Thu, Jul 11, 2024 at 4:13 AM Daye-Lee18 @.***> wrote:

Hello,

I was running the motion_representation.ipynb notebook, but I encountered an issue where the difference between an existing reference and a newly made reference is not zero.

Specifically, they are 0.00322 and 0.0185, respectively, as shown in the attached photo.

Is it okay to proceed to the next step for preprocessing the data, or should I restart the download process?

If I shouldn't proceed, what could be the reason for this issue?

Thank you. Screenshot.2024-07-11.at.5.09.20.PM.png (view on web) https://github.com/EricGuo5513/HumanML3D/assets/102942350/7f302303-8b39-4717-8ab0-4321f51ff004

— Reply to this email directly, view it on GitHub https://github.com/EricGuo5513/HumanML3D/issues/138, or unsubscribe https://github.com/notifications/unsubscribe-auth/AKRYNB5LHKCEITC5LJNWZVLZLY5DPAVCNFSM6AAAAABKWOGM3WVHI2DSMVQWIX3LMV43ASLTON2WKOZSGQYDENJVGMZTGMI . You are receiving this because you are subscribed to this thread.Message ID: @.***>