My Copy of StyleCLIP GitHub
https://github.com/bycloudai/StyleCLI...
Colab Tutorial
• StyleCLIP Colab Tutorial
Notes:
---------------
Anaconda3 Download:
https://www.anaconda.com/products/ind...
system variable CUDA 10.2 move to the very top
CUDA 10.2 Toolkit download:
https://developer.nvidia.com/cuda-10....
conda create -n StyleCLIP python=3.7
conda activate StyleCLIP
conda install pytorch==1.7.1 torchvision==0.8.2 torchaudio==0.7.2 cudatoolkit=10.2 -c pytorch
for 30 series
conda install pytorch==1.7.1 torchvision==0.8.2 torchaudio==0.7.2 cudatoolkit=11.0 -c pytorch
pip install ftfy regex tqdm gdown
pip install opencv-python
pip install tensorflow-gpu==1.14
pip install matplotlib
pip install git+https://github.com/openai/CLIP.git
##cd to the right file directory
https://nvlabs-fi-cdn.nvidia.com/styl...
##move to model directory and rename it to ffhq.pkl
cd global
python GetCode.py --dataset_name ffhq --code_type w
builds W.npy
python GetCode.py --dataset_name ffhq --code_type s
builds S
python GetCode.py --dataset_name ffhq --code_type s_mean_std
use colab to obtain the corresponding e4e latent.pt
input image inside data/ffhq
corresponding latent space file in
python GetGUIData.py --dataset_name ffhq --real
python PlayInteractively.py --dataset_name ffhq
---------------
My main channel where I introduce the latest fascinating AI tools
/ bycloudai
Related video for info/explanation
• You Describe & AI Photoshops Faces For You...
[Discord] / discord
[Twitter] / ytcloud_
[Patreon] / bycloud