/SoftwareX

CoCoStyle code

Primary LanguagePythonMIT LicenseMIT

SoftwareX

CoCoStyle

CocoStyle is an AI based mixed initiative co-creative system that supports three creative fashion design process which is research, design, and improvement.

Local server setup

System requirments:

  • We tested in NodeJS 20.10.0 and npm 10.2.3
  • We tested in Python 3.8.5 and PyTorch 1.13.1 with CUDA 11.8 and RTX 4070ti.

Frontend:

Environment installation

  1. Install (NodeJS)

  2. Install (npm)

  3. Install a package through the -g flag will install the package in the system folder.

     npm install -g n

  4. Install the stable version through a plug-in that manages the Node.js version

      n stable

  5. Inside the frontend directory, install dependencies by running the following command in the terminal:

      n stable

  6. Go to the frontend directory /src/QuQu/QuQu/, build the application by running the following command in the terminal

      npm run build

  7. Launch the frontend application by running the following command in the terminal within the /src/QuQu/QuQu/ directory:

       npm start

  8. Access the frontend of the platform from React by visiting:

     127.0.0.1:3000

Enabling S3

To enable S3 for images, go to src/QuQu/QuQu/src/ququ.config.js and enter your AWS.

149. ####INPUT S3 INFO
149. ACCESS_KEY =  ''
150. SECRET_ACCESS_KEY =  ""
151. REGION =  ""
152. S3_BUCKET =  ''

Backend:

Environment installation

To work with this project on your own machine, you need to install the environment as follows:

conda env create -f environment.yml
conda activate CoCoStyle
pip install torch==1.13.1+cu117 torchvision==0.14.1+cu117 torchaudio==0.13.1 --extra-index-url https://download.pytorch.org/whl/cu117
pip install nvidia-pyindex
pip install nvidia-tensorflow[horovod]
pip install nvidia-tensorboard==1.15
pip install paddleseg==2.5.0
pip install paddle
pip install paddlepaddle
pip install -U openmim
mim install mmcv-full==1.7.1
pip install mmdet==2.27.0
pip install cvlib=0.2.7
pip install einops
pip install pytorch_transformers==1.2.0
pip install extcolors
pip install wandb

Enabling S3

To enable S3 for images, go to QuQu-backend/ququ_backend/settings.py and enter your AWS.

149. ####INPUT S3 INFO
149. ACCESS_KEY_ID =  ''
150. ACCESS_SECRET_KEY =  ""
151. AWS_DEFAULT_REGION =  ""
152. BUCKET_NAME =  ''

Input user name

Enter the user name in QuQu-backend/ququ/views.py to distinguish between users when saving the log.

38. ###INPUT USER INFO
39. global_user_name =  ""

Download the file

Go to https://drive.google.com/file/d/12vOtPhA6OcndJR5G0o-GUI30ob9SAhM3/view?usp=drive_link and move the downloaded database to src/QuQu-backend/

Go to https://drive.google.com/file/d/1lMDldYc12J1zw6AnXefpwNBUVHwpdR0W/view?usp=drive_link and move the downloaded HumanGAN file to src/QuQu-backend/ququ/apicode/

Go to https://drive.google.com/file/d/1nKUg2nqSgv4AvNlXKi-wMP5NJzUI9I0c/view?usp=drive_link and move the downloaded pth file to src/QuQu-backend/ququ/

Django

Construct Django environment.

pip3 install Django
pip install django-cors-headers
pip install djangorestframework
cd src/
python3 manage.py migrate

Run Django server.

python manage.py runserver

Acknowlegement

Part of the code is borrowed from StyleGAN-Human, FashionFormer, Vision_Transformer.