Faceswap Forum

adam_macchiato

i fix it by model restore already , thnak you

adam_macchiato

i just chane the training source folder , cannot train and show these error , 05/23/2023 13:54:05 MainProcess _training config _check_exists DEBUG Config file exists: 'C:\Users\adama\faceswap\config\train.ini' 05/23/2023 13:54:05 MainProcess _training config _load_config VERBOSE Loading config: 'C:\...

adam_macchiato

is there any setting can set batchsize of components and extended ? 05/11/2023 19:18:14 INFO Initializing Components (Mask)... 05/11/2023 19:18:14 INFO Initialized Components (Mask) with batchsize of 1 05/11/2023 19:18:14 INFO Initializing Extended (Mask)... 05/11/2023 19:18:14 INFO Initialized Exte...

adam_macchiato

i just saw the message :
05/10/2023 23:43:41 INFO Sorting by identity similarity...
05/10/2023 23:43:41 INFO VGG Face2 Embeddings are being written to the image header. Sorting by this method will be quicker next time

actually how to use that ?> thank you

adam_macchiato

Hi , what is the main problem is ? some video cannot extract , thank you :) 04/03/2023 20:02:20 MainProcess MainThread config _load_defaults_from_module DEBUG Importing defaults module: plugins.extract.recognition.vgg_face2_defaults 04/03/2023 20:02:20 MainProcess MainThread config add_section DEBUG...

adam_macchiato

dont know why , i have to click " learn mask" , can train normal , if i dont click " learn mask " , always show the error message as before ,

adam_macchiato

Thank you , Fixed . Just want to say V2_L is really really amazing , i just use your new sym384 preset with rtx 3080 , the result is totally cannot compare with v2_L . even train 400k +200k(no warp) still a lot of blur. if i use v2_L with stojo 256px(need 3090) , 400K + 400K(nowarp) , the result ver...

adam_macchiato

Hi , just update the Faceswap and restart , cant open it anymore , the cash report as below , thank you : 0/12/2022 23:57:55 MainProcess MainThread logger log_setup INFO Log level set to: INFO 10/12/2022 23:57:56 MainProcess MainThread tpu_cluster_resolver <module> DEBUG Falling back to TensorFlow c...

adam_macchiato

So do u ready for support rtx 4090 ? i cant wait to buy it

adam_macchiato

Got it , i tried situation as below : 1) Use original mode and set eyes/ mouth to 0 , start a new model - Its work 2) Use original mode and reset eyes/ mouth to default , start a new model - Its work 3) Use Phaze-A mode and eyes/ mouth is default , V2_l with Stojo preset, start a new model - Its wor...

adam_macchiato

Ｉtried uninstall CUDA , and faceswap , and reinstall faceswap only , still not work . :cry: 09/18/2022 20:08:02 MainProcess _training api converted_call DEBUG Applying mask from channel 3 09/18/2022 20:08:02 MainProcess _training api converted_call DEBUG Processing loss function: (func: <tensorflow....

adam_macchiato

hi , this is the report , thank you for your help . 09/15/2022 10:09:59 MainProcess _training api converted_call DEBUG Applying mask from channel 3 09/15/2022 10:09:59 MainProcess _training api converted_call DEBUG Processing loss function: (func: <tensorflow.python.keras.engine.compile_utils.Losses...

adam_macchiato

Hi , I search the forum already and try to roll back , deleted all conda , cuda , and reinstall faceswap , still can't fix this problem , now all model cannot continue traning , but no problem to convert , what can i do ? thanks you. 09/15/2022 10:11:47 ERROR Caught exception in thread: '_training' ...

adam_macchiato

Just want to share my case , and hope any suggestion , I ｈave 2 PC , 1 with 3090 other with 3080 , if output 256px , 3090 can use Stojo model with V2_l , quality very good and impression 3080 can use stojo with V2_s because for 1080p+ video , some close-up shot face go blur , so i start to try 512px...

adam_macchiato

Thank you for your advise , Would you please suggest some setting about 1080p video for using V2 with rtx 3080 n 3090 ? because for now , v2-s with stojo present , 1080p video , some close-up shot the face go blur , but v2-l with stojo present much better . suggestion like : extraction px phaze-A ou...

adam_macchiato

Hi , I am using Phaze-A for traning with stojo present , its has been trained for 4days but the loss still keep on 0.017 0.016 this 48hrs , still cant go down , here is my setting : RTX3080 Phaze-A Stojo mode ( efficientnet v2_s) Learning rate : 4.5e , EE :-5 Mixed Precision and Nan on , Face A and ...

Faceswap Forum

Search found 16 matches

Re: suddently error when training

suddently error when training

Re: A stupid question , how to use VGG face2 ?

A stupid question , how to use VGG face2 ?

Some video cannot extract

Re: Caught exception in thread: '_training'

Re: cant open Faceswap after update ...

cant open Faceswap after update ...

Re: Caught exception in thread: '_training'

Re: Caught exception in thread: '_training'

Re: Caught exception in thread: '_training'

Re: Caught exception in thread: '_training'

Caught exception in thread: '_training'

Output 512px with V2 is really a difficult Task

Re: Loss cant go down spent over 48 hrs ?

Loss cant go down spent over 48 hrs ?