Han Li

> Our xgboost models use the `binary:logistic'` objective function, however the m2cgen converted version of the models return raw scores instead of the transformed scores. > > This is fine...

For 4090 machine, you need to change the followings in dockfile: ARG CUDA=11.1.1------->ARG CUDA=11.8.0 FROM nvidia/cuda:${CUDA}-cudnn8-runtime-ubuntu18.04------->FROM nvidia/cuda:${CUDA}-cudnn8-devel-ubuntu20.04 Then rebuild.

For 4090 machine, you need to change the followings in dockfile: ARG CUDA=11.1.1------->ARG CUDA=11.8.0 FROM nvidia/cuda:${CUDA}-cudnn8-runtime-ubuntu18.04------->FROM nvidia/cuda:${CUDA}-cudnn8-devel-ubuntu20.04 Then, rebuild.

I mean how to edit the code if I want to use your code to test semantic segmentation on real images

I found a way to do do segmentations on real images. I can shre my codes and contribute to the code extending.

same issue

Hi @NattapolChan, My evaluation process stops at "start evaluation" when running with multi GPUs: vbench evaluate --ngpus=8 --dimension 'motion_smoothness' --videos_path data/demofusion_controlnet.mp4 --mode=custom_input WARNING:__main__: ***************************************** Setting OMP_NUM_THREADS environment variable for each...

> Can you try running these: > > 1. `vbench evaluate --ngpus=1 --dimension 'motion_smoothness' --videos_path data/demofusion_controlnet.mp4 --mode=custom_input ` > 2. `vbench evaluate --ngpus=8 --dimension 'temporal_flickering' --videos_path data/demofusion_controlnet.mp4 --mode=custom_input` > >...

I tried torchrun --nproc_per_node=2 --standalone evaluate.py --dimension 'temporal_flickering' --videos_path duck/ --mode=custom_input, it will also ayuck at start evaluation, following is the log: WARNING:torch.distributed.run: ***************************************** Setting OMP_NUM_THREADS environment variable for each...