How to start training from last epoch where model got saved ?

andabi / deep-voice-conversion

Deep neural networks for voice conversion (voice style transfer) in Tensorflow

MIT License

3.92k stars 843 forks source link

How to start training from last epoch where model got saved ? #94

Open YashBangera7 opened 5 years ago

YashBangera7 commented 5 years ago

Can someone help me ? My computer restarted on 7000 epoch of train 2 . how should i continue training from this epoch number . Everytime I train , it starts from epoch 1 . Please help . It is urgent

linlinsongyun commented 5 years ago

give the command python train2 -ckpt which-model

YashBangera7 commented 5 years ago

The model number when mentioned as -ckpt gives assertion error. Can u help. I tried using -ckpt checkpoint (i.e., the checkpoint file that was created at the last saved epoch). This also didnt work.

YashBangera7 commented 5 years ago

Nevermind, I figured it out. Thanks for the help . I was trying to run a temporary model. Seems like you need to run the last saved model .