search

hongsukchoi / TCMR_RELEASE

Official Pytorch implementation of "Beyond Static Features for Temporally Consistent 3D Human Pose and Shape from a Video", CVPR 2021
MIT License
274 stars 39 forks source link

Tranning Needed memory size #29

Closed amituofo1996 closed 1 year ago

amituofo1996 commented 1 year ago

Hi, My computer have 23GB memory, when I start trainning repr_table6_h36m_model.yaml, I can only load needed 2d data and 3d data. How can I do? Can I separate 3D data and train stage by stage? are you still remeber your experiment memory?

amituofo1996 commented 1 year ago

Can you tell me about your trainning time?

hongsukchoi commented 1 year ago

Hi, it took approx 55GB.

My recommendation is to discard 1/3 - 1/2 of the whole dataset in a random manner and train with the remaining data. It would give similar accuracy

amituofo1996 commented 1 year ago

Thanks for your quick reply!