Skip to content

Latest commit

 

History

History
63 lines (52 loc) · 2.37 KB

install.md

File metadata and controls

63 lines (52 loc) · 2.37 KB
Raw

INSTALLATION

Installation is tested and working on the following platforms:

  • Ubuntu 20.04.1
    • GPUs: RTX-3090 (Driver Version: 520.56.06)
    • RAM: 64GB

Prerequisites

The detailed package requirement can be found in requirements.txt.
To install the packages required by our LED, you could :

  1. Create a virtual enviorment and activate it: 
    conda create -y -n LED-ICCV23 python=3.8
conda activate LED-ICCV23
  2. Install the prerequisite packages: 
    pip install -r requirements.txt
  3. Install our LED for develop: 
    python setup.py develop

Attention! ONE package (RawPy) is not included in requirements.txt!
To read the cam2rgb matrix from RAW format data, we use the customized rawpy by Vandermode (the author of ELD) during data preparation.

By the way, if you don't need to reproduce the metrics in our paper, you can simply install rawpy through pip install rawpy and skip the following part. But in this way, you won't be able to use the dataloader we have already prepared.

We heavily recommend you follow the instructions from ELD to install the custumized rawpy, or you can just follow the next steps.

  1. Download custumized rawpy and LibRaw, then unzip them: 
    # download in downloads/
mkdir -p downloads/
# use our script for downloading from google drive
wget https://www.libraw.org/data/LibRaw-0.21.1.zip -O downloads/LibRaw-0.21.1.zip
python scripts/download_gdrive.py --id 1EuJsbZ_a_YJHHcGAVA9TXXPnGU90QoP4 --save-path downloads/rawpy.zip
# unzip the rawpy and LibRaw
unzip downloads/LibRaw-0.21.1.zip -d downloads/
unzip downloads/rawpy.zip -d downloads/
  2. Compile and install LibRaw: 
    cd downloads/LibRaw-0.21.1
./configure
make
sudo make install
  3. Install RawPy! (Please pay attention to whether you are in a virtual environment): 
    cd ../rawpy
RAWPY_USE_SYSTEM_LIBRAW=1 pip install -e .

All the above instructions are integrated into install.sh.
So you can simply install by bash install.sh.

Pretrained Models

If you would like to use our pretrained network, please refer to pretrained-models.md.