search

minar09 / cp-vton-plus

Official implementation for "CP-VTON+: Clothing Shape and Texture Preserving Image-Based Virtual Try-On", CVPRW 2020
https://minar09.github.io/cpvtonplus/
MIT License
346 stars 120 forks source link

Restart training from checkpoint #20

Closed Bouncer51 closed 4 years ago

Bouncer51 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 --stage GMM --workers 8 --save_count 500 --checkpoint "checkpoints/gmm_train_new/step_000500.pth"

thaithanhtuan 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 --stage GMM --workers 8 --save_count 500 --checkpoint "checkpoints/gmm_train_new/step_000500.pth"

If you do so, It actually continues to train from step 500, but if you want the output print from 500 (not step 0), from train.py line 70, you can change from: for step in range(opt.keep_step + opt.decay_step): to start_step = 500

for step in range(start_step, opt.keep_step + opt.decay_step):