encoder4editing

encoder4editing copied to clipboard

Suggestions on training an encoder on FFHQ-256

1

Hi! Thanks for your excellent work I am trying to train an encoder on FFHQ-256(simply downsample by 4, no other difference). I followed your instructions, using pretrained model from [rosinality's...

FishWoWater

First thank you for your great work. I'm wondering if it is possible to train the encoder which is segmap to face? Just like the pSp style mixing method?

Augest8

yunus-temurlenk

Model parameters at other resolutions

2

Can you provide the model parameters on the FFHQ dataset with a resolution of 256?

li-car-fei

Hello, I have tried to re-train the e4e on a dataset from production. In doing so, I found that the model is able to reproduce the dimensions of my images...

chrisgaig

scripts/train.py --dataset_type cars_encode --exp_dir directory --use_w_pool --w_discriminator_lambda 0.1 --progressive_start 20000 --id_lambda 0.5 --val_interval 10000 --start_from_latent_avg --max_steps 200000 --stylegan_size 512 --stylegan_weights a/stylegan2-ffhq-config-f.pt --workers 8 --batch_size 8 --test_batch_size 4 --test_workers 4 {'batch_size':...

softalter

How do I train an encoder with a resolution of 256*256 on the FFHQ dataset?

1

How do I train an encoder with a resolution of 256*256 on the FFHQ dataset?

li-car-fei

For e4e pre-training weights, the input size is 256 and the output is 1024. For a series of indicators such as FID, when calculating the indicators, are the original images...

achangzx