search

sergeywong / cp-vton

Reimplemented code for "Toward Characteristic-Preserving Image-based Virtual Try-On Network"
MIT License
474 stars 182 forks source link

Restart training from checkpoint #29

Open bohzio opened 4 years ago

bohzio commented 4 years ago

Hi, there is a way to restart the training from a specific checkpoint? I'm trying this command but it starts always from step 0 python train.py --name gmm_train_new --stage GMM --workers 0 --save_count 5000 --checkpoint "checkpoints/gmm_train_new/step_065000.pth"

KangsanKim07 commented 4 years ago

@bohzio , Did you solve that problem? I'm suffering from same problem.

Amazingren commented 3 years ago

Hi, there is a way to restart the training from a specific checkpoint? I'm trying this command but it starts always from step 0 python train.py --name gmm_train_new --stage GMM --workers 0 --save_count 5000 --checkpoint "checkpoints/gmm_train_new/step_065000.pth"

In train.py line 56,
change "for step in range(opt.keep_step + opt.decay_step)" to "for step in range(064999, opt.keep_step + opt.decay_step)"

Then it works for me!