geobao

This is the config file that I am using: ```python _base_ = [ '../_base_/datasets/hrsc.py', '../_base_/schedules/schedule_3x.py', '../../_base_/default_runtime.py' ] # model settings model = dict( type='FCOSOBB', pretrained='open-mmlab://detectron/resnet50_caffe', backbone=dict( type='ResNet', depth=50, num_stages=4, out_indices=(0,...

我也是rtx3090 I am trying: ARG PYTORCH="1.9.0" ARG CUDA="11.1" ARG CUDNN="8" FROM pytorch/pytorch:${PYTORCH}-cuda${CUDA}-cudnn${CUDNN}-devel ENV CUDALONG=111 ENV TORCH_CUDA_ARCH_LIST="8.6" It doesn't work

Before I had: `pip install mmcv-full>=1.3 -f https://download.openmmlab.com/mmcv/dist/cu${CUDALONG}/torch${PYTORCH}/index.html` Now I fixed it to `pip install mmcv-full==1.3 -f https://download.openmmlab.com/mmcv/dist/cu${CUDALONG}/torch${PYTORCH}/index.html` This fixed it for me

I did not solve it. To get performances I executed several experiments and used the median. I cannot explain why this happens...