bad results

[YznMur]

Hi, thanks for your great work.

I am running SCTNet into docker:
Dockerfile :

ARG PYTORCH="1.11.0"
ARG CUDA="11.3"
ARG CUDNN="8"
FROM pytorch/pytorch:${PYTORCH}-cuda${CUDA}-cudnn${CUDNN}-devel
RUN rm /etc/apt/sources.list.d/cuda.list
RUN rm /etc/apt/sources.list.d/nvidia-ml.list

RUN pip install mmcv-full==1.6.0 -f https://download.openmmlab.com/mmcv/dist/cu113/torch1.11.0/index.html
RUN pip install timm
RUN pip install matplotlib
RUN pip install prettytable
RUN pip install einops
RUN apt update && apt install -y libgl1-mesa-glx
RUN pip install opencv-python-headless
RUN apt-get update && apt-get install ffmpeg libsm6 libxext6  -y
RUN pip install torch==1.11.0+cu113 torchvision==0.12.0+cu113 torchaudio==0.11.0 --extra-index-url https://download.pytorch.org/whl/cu113

But, I am getting such bad results:

I believe this is happening because of some function from mmseg.
Any ideas how to fix this?.

[YznMur]

@xzz777 I wish you can help with this.
thanks