jpy.cas
which pytorch version do you use? I downgrade tesorboardX version to 1.6, another issue occured as below: @TeeboneTing @leoxiaobin => init weights from normal distribution => loading pretrained model models/pytorch/imagenet/hrnet_w32-36af842e.pth...
I use pytorch==1.2 torchvision==0.4, Does anyone have similar experience? no problem when running "python tools/test.py", in coco/mpii
it works when torch==1.0, tensorboardX==1.6
@TeeboneTing thanks. @eng100200 pls describe your issue.
@hongsukchoi have you download AI Challenge datasets? I cann't access to it's website
@hongsukchoi one friend share me the AI challenge datasets, FYI: https://pan.baidu.com/s/1YFzoIkv51Yy_WjCvGv77qw#list/path=%2F
I also met this issue for eval.py, try to reinstall 1.12. or 1.11, no work. reinstall numpy==1.17, it works.
Q1: convert solo to onnx File "/home/jiapy/workspace/segmentation/SOLO/mmdet/core/fp16/decorators.py", line 49, in new_func return old_func(*args, **kwargs) File "/home/jiapy/workspace/segmentation/SOLO/mmdet/models/necks/fpn.py", line 115, in forward laterals[i], scale_factor=2, mode='nearest') RuntimeError: The size of tensor a (167)...
