mega.pytorch

mega.pytorch copied to clipboard

Thanks for your sharing. I wonder how to obtain AP (fast) | AP (med) | AP (slow) after training? It would be helpful if it is convenient to tell. Thank...

ruiyuanlu

Hello, I want to know how the pretrained 'flownet.ckpt' model was trained. Was it also trained using flownet.py in mega_core/modeling/backbone with average pooling in forward? Or trained using another method...

pyCERN

error: identifier "AT_CHECK" is undefined

2

I had an error during installation:  I'm not sure where the Macro came from , how to fix it? Thanks

liuky74

some questions about the dff

1

Thanks for your sharing! And I test the dff part,, the inference speed is about 15-20fps on Titan 2080ti. But original dff(mxnet) is more than 30fps on Titan 2080ti. What...

dylan110

jhxu-org

请问，有人只使用ImageNet VID数据集训练的吗？mAP是多少。我这边是0.71.

liwenjielongren

Thanks for the code. I just run it for baseline. But I met the PIL error. UnboundLocalError: local variable 'photoshop' referenced before assignment during the training. What it is your...

lxtGH

No module named 'mega_core'

5

执行python -m torch.distributed.launch --nproc_per_node 4 tools/test_net.py --config-file configs/MEGA/vid_R_101_C4_MEGA_1x.yaml --motion-specific MODEL.WEIGHT MEGA_R_101.pth，报错：ModuleNotFoundError: No module named 'mega_core' ，请问是引入方式需要调整吗

wspspring

Segmentation Error

2

Please while trying to run demo.py I'm having a segmentation error. Anyone had the same problem?

khadijakhaldi

run the model on my dataset

1

Please I'm confused, how can I run the model on my videos? I'm using this command, but I'm having a segmentation error python demo/demo.py base configs/vid_R_101_C4_1x.yaml R_101.pth --video --visualize-path datasets/20180820_115017_260A.mkv...

khadijakhaldi