Pytorch implementation of REDet, ACCV 2022
'resnet50': 'https://download.pytorch.org/models/resnet50-19c8e357.pth',
'resnet101': 'https://download.pytorch.org/models/resnet101-5d3b4d8f.pth',
After downloading, please put it into "pretrained/"
Then download the images and annotations from the official website of LVIS.
Finally the file structure of folder lvis will be like this:
$lvis
├── annotations
│ ├── lvis_v1_val.json
│ ├── lvis_v1_train.json
├── train2017
│ ├── 000000004134.png
│ ├── 000000031817.png
│ ├── ......
├── val2017
├── test2017
# create environment
conda create --name REDet python=3.7
conda activate REDet
# install pytorch
conda install pytorch==1.8.1 torchvision==0.9.1 torchaudio==0.8.1 cudatoolkit=10.1edit easy_setup.sh：
#!/bin/bash
export PATH=/your/path/to/gcc-5.3.0/bin/:$PATH # gcc path
export LD_LIBRARY_PATH=/your/path/to/gmp-4.3.2/lib/:/your/path/to/mpfr-2.4.2/lib/:/your/path/to/mpc-0.8.1/lib/:$LD_LIBRARY_PATH # lib path
export TORCH_CUDA_ARCH_LIST='3.5;5.0+PTX;6.0;7.0' # cuda list
python setup.py build_ext -iThen:
# setup
./easy_setup.sh
# pip install requirements.txt
pip install -r requirements.txt
# install other packages
pip uninstall protobuf
pip install protobuf==3.20.1
pip install pyyaml
pip install scikit-imageWe provide training scripts experiments/test.sh as:
#!/bin/bash
ROOT=../../
T=`date +%m%d%H%M`
export ROOT=$ROOT
cfg=configs/Table1_r50_REDet.yaml
export PYTHONPATH=$ROOT:$PYTHONPATH
python -m up train \
--nm=1 \
--ng=1 \
--launch=pytorch \
--config=$cfg \
2>&1 | tee experiments/train_log/log.train.$T.$(basename $cfg) We provide testing scripts experiments/test.sh as:
#!/bin/bash
ROOT=../../
T=`date +%m%d%H%M`
export ROOT=$ROOT
cfg=configs/Table1_r50_REDet.yaml
export PYTHONPATH=$ROOT:$PYTHONPATH
python -m up train \
-e \
--nm=1 \
--ng=1 \
--launch=pytorch \
--config=$cfg \
2>&1 | tee experiments/test_log/log.test.$T.$(basename $cfg) 
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent