github/DeepFaceLab
1
0
Fork 0
mirror of https://github.com/iperov/DeepFaceLab.git synced 2025-07-07 05:22:06 -07:00
Code Issues Projects Releases Packages Wiki Activity Actions

SAEHD,Quick96:

Browse source 
improved model generalization, overall accuracy and sharpness
by using new 'Learning rate dropout' technique from paper https://arxiv.org/abs/1912.00144
An example of a loss histogram where this function is enabled after the red arrow:
https://i.imgur.com/3olskOd.jpg
This commit is contained in:
Colombo 2019-12-15 15:53:06 +04:00
parent c866448645
commit 71ebf06c89
2 changed files with 5 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

4
models/Model_Quick96/Model.py
View file

@ -143,8 +143,8 @@ class Quick96Model(ModelBase):
self.CA_conv_weights_list += [layer.weights[0]] #- is Conv2D kernel_weights self.CA_conv_weights_list += [layer.weights[0]] #- is Conv2D kernel_weights
if self.is_training_mode: if self.is_training_mode:
self.src_dst_opt = RMSprop(lr=2e-4) self.src_dst_opt = RMSprop(lr=2e-4, lr_dropout=0.3)
self.src_dst_mask_opt = RMSprop(lr=2e-4) self.src_dst_mask_opt = RMSprop(lr=2e-4, lr_dropout=0.3)
target_src_masked = self.model.target_src*self.model.target_srcm target_src_masked = self.model.target_src*self.model.target_srcm
target_dst_masked = self.model.target_dst*self.model.target_dstm target_dst_masked = self.model.target_dst*self.model.target_dstm

6
models/Model_SAEHD/Model.py
View file

@ -452,9 +452,9 @@ class SAEHDModel(ModelBase):
psd_target_dst_anti_masked = self.model.pred_src_dst*(1.0 - target_dstm) psd_target_dst_anti_masked = self.model.pred_src_dst*(1.0 - target_dstm)
if self.is_training_mode: if self.is_training_mode:
self.src_dst_opt = RMSprop(lr=5e-5, clipnorm=1.0 if self.options['clipgrad'] else 0.0, tf_cpu_mode=self.options['optimizer_mode']-1) self.src_dst_opt = RMSprop(lr=5e-5, lr_dropout=0.3, clipnorm=1.0 if self.options['clipgrad'] else 0.0, tf_cpu_mode=self.options['optimizer_mode']-1)
self.src_dst_mask_opt = RMSprop(lr=5e-5, clipnorm=1.0 if self.options['clipgrad'] else 0.0, tf_cpu_mode=self.options['optimizer_mode']-1) self.src_dst_mask_opt = RMSprop(lr=5e-5, lr_dropout=0.3, clipnorm=1.0 if self.options['clipgrad'] else 0.0, tf_cpu_mode=self.options['optimizer_mode']-1)
self.D_opt = RMSprop(lr=5e-5, clipnorm=1.0 if self.options['clipgrad'] else 0.0, tf_cpu_mode=self.options['optimizer_mode']-1) self.D_opt = RMSprop(lr=5e-5, lr_dropout=0.3, clipnorm=1.0 if self.options['clipgrad'] else 0.0, tf_cpu_mode=self.options['optimizer_mode']-1)
src_loss = K.mean ( 10*dssim(kernel_size=int(resolution/11.6),max_value=1.0)( target_src_masked_opt, pred_src_src_masked_opt) ) src_loss = K.mean ( 10*dssim(kernel_size=int(resolution/11.6),max_value=1.0)( target_src_masked_opt, pred_src_src_masked_opt) )
src_loss += K.mean ( 10*K.square( target_src_masked_opt - pred_src_src_masked_opt ) ) src_loss += K.mean ( 10*K.square( target_src_masked_opt - pred_src_src_masked_opt ) )