Re: [Guide] Using Faceswap on Nvidia RTX 30xx cards
Posted: Wed Jan 06, 2021 5:59 pm
Are you running command prompt as ADMIN?
Faceswap Forum
The place to discuss Faceswap and Deepfakes
https://forum.faceswap.dev/
[Guide] Using Faceswap on Nvidia RTX 30xx cards
Page 2 of 7
Re: [Guide] Using Faceswap on Nvidia RTX 30xx cards
Posted: Wed Jan 06, 2021 6:00 pm
DO NOT DO THIS! DO NOT INSTALL OR RUN FACESWAP AS ADMIN!
It is not necessary and can lead to problems.
Re: [Guide] Using Faceswap on Nvidia RTX 30xx cards
Posted: Thu Jan 07, 2021 4:43 am
No, not tried it in Admin.
When I do, Bryan beats me with a old videocard in a sock.
Re: [Guide] Using Faceswap on Nvidia RTX 30xx cards
Posted: Thu Jan 07, 2021 8:02 pm
Re: [Guide] Using Faceswap on Nvidia RTX 30xx cards
Posted: Thu Jan 07, 2021 10:06 pm
Re: [Guide] Using Faceswap on Nvidia RTX 30xx cards
Posted: Fri Jan 08, 2021 4:00 pm
Hi Sirs,
I followed the instruction to install it. but I encountered some problems.
It seems like issue happened and it caused the installation not successful.
I am not familiar with Conda environment.
Could you help to take a look on this ? Thanks !! 
My graphic card : RTX3070
Nvidia driver: 457.51
CUDA version : 11.0
CuDNN version : 8.0.4
When I typing command which is "conda activate faceswap" in Anaconda Prompt.
It shows following information.
code C:\Users\chris>conda activate faceswap
Could not find conda environment: faceswap
You can list all discoverable environments withconda info --envs.[/code]When I typing command which is "conda remove tensorflow" in Anaconda Prompt.
code C:\Users\chris>conda remove tensorflow
Collecting package metadata (repodata.json): done
Solving environment: failed
PackagesNotFoundError: The following packages are missing from the target environment:
tensorflow[/code]
When I typing command which is "pip install tensorflow-gpu==2.4" in Anaconda Prompt.
code C:\Users\chris>pip install tensorflow-gpu==2.4
Requirement already satisfied: tensorflow-gpu==2.4 in c:\users\chris\anaconda3\lib\site-packages (2.4.0)
Requirement already satisfied: numpy~=1.19.2 in c:\users\chris\anaconda3\lib\site-packages (from tensorflow-gpu==2.4) (1.19.2)
Requirement already satisfied: six~=1.15.0 in c:\users\chris\anaconda3\lib\site-packages (from tensorflow-gpu==2.4) (1.15.0)
Requirement already satisfied: h5py~=2.10.0 in c:\users\chris\anaconda3\lib\site-packages (from tensorflow-gpu==2.4) (2.10.0)
Requirement already satisfied: grpcio~=1.32.0 in c:\users\chris\anaconda3\lib\site-packages (from tensorflow-gpu==2.4) (1.32.0)
Requirement already satisfied: wrapt~=1.12.1 in c:\users\chris\anaconda3\lib\site-packages (from tensorflow-gpu==2.4) (1.12.1)
Requirement already satisfied: absl-py~=0.10 in c:\users\chris\anaconda3\lib\site-packages (from tensorflow-gpu==2.4) (0.11.0)
Requirement already satisfied: opt-einsum~=3.3.0 in c:\users\chris\anaconda3\lib\site-packages (from tensorflow-gpu==2.4) (3.3.0)
Requirement already satisfied: wheel~=0.35 in c:\users\chris\anaconda3\lib\site-packages (from tensorflow-gpu==2.4) (0.35.1)
Requirement already satisfied: tensorflow-estimator<2.5.0,>=2.4.0rc0 in c:\users\chris\anaconda3\lib\site-packages (from tensorflow-gpu==2.4) (2.4.0)
Requirement already satisfied: keras-preprocessing~=1.1.2 in c:\users\chris\anaconda3\lib\site-packages (from tensorflow-gpu==2.4) (1.1.2)
Requirement already satisfied: typing-extensions~=3.7.4 in c:\users\chris\anaconda3\lib\site-packages (from tensorflow-gpu==2.4) (3.7.4.3)
Requirement already satisfied: flatbuffers~=1.12.0 in c:\users\chris\anaconda3\lib\site-packages (from tensorflow-gpu==2.4) (1.12)
Requirement already satisfied: protobuf>=3.9.2 in c:\users\chris\anaconda3\lib\site-packages (from tensorflow-gpu==2.4) (3.14.0)
Requirement already satisfied: gast==0.3.3 in c:\users\chris\anaconda3\lib\site-packages (from tensorflow-gpu==2.4) (0.3.3)
Requirement already satisfied: tensorboard~=2.4 in c:\users\chris\anaconda3\lib\site-packages (from tensorflow-gpu==2.4) (2.4.0)
Requirement already satisfied: termcolor~=1.1.0 in c:\users\chris\anaconda3\lib\site-packages (from tensorflow-gpu==2.4) (1.1.0)
Requirement already satisfied: google-pasta~=0.2 in c:\users\chris\anaconda3\lib\site-packages (from tensorflow-gpu==2.4) (0.2.0)
Requirement already satisfied: astunparse~=1.6.3 in c:\users\chris\anaconda3\lib\site-packages (from tensorflow-gpu==2.4) (1.6.3)
Requirement already satisfied: requests<3,>=2.21.0 in c:\users\chris\anaconda3\lib\site-packages (from tensorboard~=2.4->tensorflow-gpu==2.4) (2.24.0)
Requirement already satisfied: google-auth<2,>=1.6.3 in c:\users\chris\anaconda3\lib\site-packages (from tensorboard~=2.4->tensorflow-gpu==2.4) (1.24.0)
Requirement already satisfied: tensorboard-plugin-wit>=1.6.0 in c:\users\chris\anaconda3\lib\site-packages (from tensorboard~=2.4->tensorflow-gpu==2.4) (1.7.0)
Requirement already satisfied: markdown>=2.6.8 in c:\users\chris\anaconda3\lib\site-packages (from tensorboard~=2.4->tensorflow-gpu==2.4) (3.3.3)
Requirement already satisfied: werkzeug>=0.11.15 in c:\users\chris\anaconda3\lib\site-packages (from tensorboard~=2.4->tensorflow-gpu==2.4) (1.0.1)
Requirement already satisfied: google-auth-oauthlib<0.5,>=0.4.1 in c:\users\chris\anaconda3\lib\site-packages (from tensorboard~=2.4->tensorflow-gpu==2.4) (0.4.2)
Requirement already satisfied: setuptools>=41.0.0 in c:\users\chris\anaconda3\lib\site-packages (from tensorboard~=2.4->tensorflow-gpu==2.4) (50.3.1.post20201107)
Requirement already satisfied: chardet<4,>=3.0.2 in c:\users\chris\anaconda3\lib\site-packages (from requests<3,>=2.21.0->tensorboard~=2.4->tensorflow-gpu==2.4) (3.0.4)
Requirement already satisfied: idna<3,>=2.5 in c:\users\chris\anaconda3\lib\site-packages (from requests<3,>=2.21.0->tensorboard~=2.4->tensorflow-gpu==2.4) (2.10)
Requirement already satisfied: certifi>=2017.4.17 in c:\users\chris\anaconda3\lib\site-packages (from requests<3,>=2.21.0->tensorboard~=2.4->tensorflow-gpu==2.4) (2020.6.20)
Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in c:\users\chris\anaconda3\lib\site-packages (from requests<3,>=2.21.0->tensorboard~=2.4->tensorflow-gpu==2.4) (1.25.11)
Requirement already satisfied: rsa<5,>=3.1.4; python_version >= "3.6" in c:\users\chris\anaconda3\lib\site-packages (from google-auth<2,>=1.6.3->tensorboard~=2.4->tensorflow-gpu==2.4) (4.6)
Requirement already satisfied: cachetools<5.0,>=2.0.0 in c:\users\chris\anaconda3\lib\site-packages (from google-auth<2,>=1.6.3->tensorboard~=2.4->tensorflow-gpu==2.4) (4.2.0)
Requirement already satisfied: pyasn1-modules>=0.2.1 in c:\users\chris\anaconda3\lib\site-packages (from google-auth<2,>=1.6.3->tensorboard~=2.4->tensorflow-gpu==2.4) (0.2.8)
Requirement already satisfied: requests-oauthlib>=0.7.0 in c:\users\chris\anaconda3\lib\site-packages (from google-auth-oauthlib<0.5,>=0.4.1->tensorboard~=2.4->tensorflow-gpu==2.4) (1.3.0)
Requirement already satisfied: pyasn1>=0.1.3 in c:\users\chris\anaconda3\lib\site-packages (from rsa<5,>=3.1.4; python_version >= "3.6"->google-auth<2,>=1.6.3->tensorboard~=2.4->tensorflow-gpu==2.4) (0.4.8)
Requirement already satisfied: oauthlib>=3.0.0 in c:\users\chris\anaconda3\lib\site-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<0.5,>=0.4.1->tensorboard~=2.4->tensorflow-gpu==2.4) (3.1.0)
[/code]I can open the faceswap software , it seems like running in CPU mode.
Even the information shows setting backend to NVIDIA. but when I tried to run the extract/train mode , the GPU loading rate only 5% , CPU loading rate is 100%.
Code: Select all
Loading...
Setting Faceswap backend to NVIDIA
01/08/2021 23:55:49 INFO Log level set to: INFO
01/08/2021 23:55:51 INFO Model A Directory: E:\videodata\1
01/08/2021 23:55:51 INFO Model B Directory: E:\videodata\2
01/08/2021 23:55:51 INFO Training data directory: E:\videodata\4
01/08/2021 23:55:51 INFO ===================================================
01/08/2021 23:55:51 INFO Starting
01/08/2021 23:55:51 INFO Press 'Stop' to save and quit
01/08/2021 23:55:51 INFO ===================================================
01/08/2021 23:55:52 INFO Loading data, this may take a while...
01/08/2021 23:55:52 INFO Loading Model from Original plugin...
01/08/2021 23:55:52 INFO Using configuration saved in state file
01/08/2021 23:55:53 INFO Loaded model from disk: 'E:\videodata\4\original.h5'
01/08/2021 23:55:53 INFO Loading Trainer from Original plugin...
Reading training images (A): 0%| | 0/368 [00:00<?, ?it/s]
Reading training images (A): 55%|█████▌ | 203/368 [00:00<00:00, 1931.61it/s]
Reading training images (B): 0%| | 0/388 [00:00<?, ?it/s]
Reading training images (B): 17%|█▋ | 65/388 [00:00<00:00, 624.44it/s]
Reading training images (B): 75%|███████▌ | 292/388 [00:00<00:00, 1566.48it/s]01/08/2021 23:55:54 INFO Reading alignments from: 'E:\videodata\1\8_alignments.fsa'
01/08/2021 23:55:54 INFO Reading alignments from: 'E:\videodata\2\BBB_alignments.fsa'Re: [Guide] Using Faceswap on Nvidia RTX 30xx cards
Posted: Fri Jan 08, 2021 4:27 pm
Looks as though you skipped one of the first steps.
.
Looks as though the conda environment wasn't created , or you named the environment something other than "faceswap"
Install FaceSwap first, per the instructions.
That will create the conda environment... Then continue with the instructions
Re: [Guide] Using Faceswap on Nvidia RTX 30xx cards
Posted: Fri Jan 08, 2021 11:42 pm
Ok, followed this process. I noticed extraction is significantly more "zoomed out" than it was before. I loaded a previous project file and kept the settings the same. See below for before / after.
Is this expected behavior?
Before
After
Re: [Guide] Using Faceswap on Nvidia RTX 30xx cards
Posted: Sat Jan 09, 2021 2:16 am
Hi abigflea ,
Thanks for your help. Now I already finished installation again and solve the ""conda environment".
But I tried to run the extract/train mode , the GPU loading rate still only 5% , CPU loading rate is 100%.
I put some faceswap information as below , could you help to take a look on this ?
Output system information from faceswap
Code: Select all
============ System Information ============
encoding: cp950
git_branch: master
git_commits: 163b8fe sysinfo/setup - Standardize cuda/cudnn checks
gpu_cuda: 11.0
gpu_cudnn: 8.0.5
gpu_devices: GPU_0: GeForce RTX 3070
gpu_devices_active: GPU_0
gpu_driver: 457.51
gpu_vram: GPU_0: 8192MB
os_machine: AMD64
os_platform: Windows-10-10.0.18362-SP0
os_release: 10
py_command: C:\Users\chris\faceswap/faceswap.py gui
py_conda_version: conda 4.9.2
py_implementation: CPython
py_version: 3.8.5
py_virtual_env: True
sys_cores: 16
sys_processor: AMD64 Family 23 Model 8 Stepping 2, AuthenticAMD
sys_ram: Total: 32715MB, Available: 19321MB, Used: 13394MB, Free: 19321MB
=============== Pip Packages ===============
absl-py==0.11.0
astunparse==1.6.3
cachetools==4.2.0
certifi==2020.12.5
chardet==4.0.0
cycler==0.10.0
fastcluster==1.1.26
ffmpy==0.2.3
flatbuffers==1.12
gast==0.3.3
google-auth==1.24.0
google-auth-oauthlib==0.4.2
google-pasta==0.2.0
grpcio==1.32.0
h5py==2.10.0
idna==2.10
imageio @ file:///tmp/build/80754af9/imageio_1594161405741/work
imageio-ffmpeg @ file:///home/conda/feedstock_root/build_artifacts/imageio-ffmpeg_1609799311556/work
joblib @ file:///tmp/build/80754af9/joblib_1607970656719/work
Keras-Preprocessing==1.1.2
kiwisolver @ file:///C:/ci/kiwisolver_1604014703538/work
Markdown==3.3.3
matplotlib @ file:///C:/ci/matplotlib-base_1592837548929/work
mkl-fft==1.2.0
mkl-random==1.1.1
mkl-service==2.3.0
numpy @ file:///C:/ci/numpy_and_numpy_base_1603466732592/work
nvidia-ml-py3 @ git+https://github.com/deepfakes/nvidia-ml-py3.git@6fc29ac84b32bad877f078cb4a777c1548a00bf6
oauthlib==3.1.0
olefile==0.46
opencv-python==4.5.1.48
opt-einsum==3.3.0
pathlib==1.0.1
Pillow @ file:///C:/ci/pillow_1609786840597/work
protobuf==3.14.0
psutil @ file:///C:/ci/psutil_1598370330503/work
pyasn1==0.4.8
pyasn1-modules==0.2.8
pyparsing==2.4.7
python-dateutil==2.8.1
pywin32==227
requests==2.25.1
requests-oauthlib==1.3.0
rsa==4.6
scikit-learn @ file:///C:/ci/scikit-learn_1598377018496/work
scipy @ file:///C:/ci/scipy_1604596260408/work
sip==4.19.13
six @ file:///C:/ci/six_1605187374963/work
tensorboard==2.4.0
tensorboard-plugin-wit==1.7.0
tensorflow-estimator==2.4.0
tensorflow-gpu==2.4.0
termcolor==1.1.0
threadpoolctl @ file:///tmp/tmp9twdgx9k/threadpoolctl-2.1.0-py3-none-any.whl
tornado @ file:///C:/ci/tornado_1606942392901/work
tqdm @ file:///tmp/build/80754af9/tqdm_1609788246169/work
typing-extensions==3.7.4.3
urllib3==1.26.2
Werkzeug==1.0.1
wincertstore==0.2
wrapt==1.12.1
============== Conda Packages ==============
# packages in environment at C:\Users\chris\anaconda3\envs\faceswap:
#
# Name Version Build Channel
absl-py 0.11.0 pypi_0 pypi
astunparse 1.6.3 pypi_0 pypi
blas 1.0 mkl
ca-certificates 2020.12.8 haa95532_0
cachetools 4.2.0 pypi_0 pypi
certifi 2020.12.5 py38haa95532_0
chardet 4.0.0 pypi_0 pypi
cycler 0.10.0 py38_0
fastcluster 1.1.26 py38h251f6bf_2 conda-forge
ffmpeg 4.2.2 he774522_0
ffmpy 0.2.3 pypi_0 pypi
flatbuffers 1.12 pypi_0 pypi
freetype 2.10.4 hd328e21_0
gast 0.3.3 pypi_0 pypi
git 2.23.0 h6bb4b03_0
google-auth 1.24.0 pypi_0 pypi
google-auth-oauthlib 0.4.2 pypi_0 pypi
google-pasta 0.2.0 pypi_0 pypi
grpcio 1.32.0 pypi_0 pypi
h5py 2.10.0 pypi_0 pypi
icc_rt 2019.0.0 h0cc432a_1
icu 58.2 ha925a31_3
idna 2.10 pypi_0 pypi
imageio 2.9.0 py_0
imageio-ffmpeg 0.4.3 pyhd8ed1ab_0 conda-forge
intel-openmp 2020.2 254
joblib 1.0.0 pyhd3eb1b0_0
jpeg 9b hb83a4c4_2
keras-preprocessing 1.1.2 pypi_0 pypi
kiwisolver 1.3.0 py38hd77b12b_0
libpng 1.6.37 h2a8f88b_0
libtiff 4.1.0 h56a325e_1
lz4-c 1.9.2 hf4a77e7_3
markdown 3.3.3 pypi_0 pypi
matplotlib 3.2.2 0
matplotlib-base 3.2.2 py38h64f37c6_0
mkl 2020.2 256
mkl-service 2.3.0 py38h196d8e1_0
mkl_fft 1.2.0 py38h45dec08_0
mkl_random 1.1.1 py38h47e9c7a_0
numpy 1.19.2 py38hadc3359_0
numpy-base 1.19.2 py38ha3acd2a_0
nvidia-ml-py3 7.352.1 pypi_0 pypi
oauthlib 3.1.0 pypi_0 pypi
olefile 0.46 py_0
opencv-python 4.5.1.48 pypi_0 pypi
openssl 1.1.1i h2bbff1b_0
opt-einsum 3.3.0 pypi_0 pypi
pathlib 1.0.1 py_1
pillow 8.1.0 py38h4fa10fc_0
pip 20.3.3 py38haa95532_0
protobuf 3.14.0 pypi_0 pypi
psutil 5.7.2 py38he774522_0
pyasn1 0.4.8 pypi_0 pypi
pyasn1-modules 0.2.8 pypi_0 pypi
pyparsing 2.4.7 py_0
pyqt 5.9.2 py38ha925a31_4
python 3.8.5 h5fd99cc_1
python-dateutil 2.8.1 py_0
python_abi 3.8 1_cp38 conda-forge
pywin32 227 py38he774522_1
qt 5.9.7 vc14h73c81de_0
requests 2.25.1 pypi_0 pypi
requests-oauthlib 1.3.0 pypi_0 pypi
rsa 4.6 pypi_0 pypi
scikit-learn 0.23.2 py38h47e9c7a_0
scipy 1.5.2 py38h14eb087_0
setuptools 51.0.0 py38haa95532_2
sip 4.19.13 py38ha925a31_0
six 1.15.0 py38haa95532_0
sqlite 3.33.0 h2a8f88b_0
tensorboard 2.4.0 pypi_0 pypi
tensorboard-plugin-wit 1.7.0 pypi_0 pypi
tensorflow-estimator 2.4.0 pypi_0 pypi
tensorflow-gpu 2.4.0 pypi_0 pypi
termcolor 1.1.0 pypi_0 pypi
threadpoolctl 2.1.0 pyh5ca1d4c_0
tk 8.6.10 he774522_0
tornado 6.1 py38h2bbff1b_0
tqdm 4.55.1 pyhd3eb1b0_0
typing-extensions 3.7.4.3 pypi_0 pypi
urllib3 1.26.2 pypi_0 pypi
vc 14.2 h21ff451_1
vs2015_runtime 14.27.29016 h5e58377_2
werkzeug 1.0.1 pypi_0 pypi
wheel 0.36.2 pyhd3eb1b0_0
wincertstore 0.2 py38_0
wrapt 1.12.1 pypi_0 pypi
xz 5.2.5 h62dcd97_0
zlib 1.2.11 h62dcd97_4
zstd 1.4.5 h04227a9_0
================= Configs ==================
--------- .faceswap ---------
backend: nvidia
--------- convert.ini ---------
[color.color_transfer]
clip: True
preserve_paper: True
[color.manual_balance]
colorspace: HSV
balance_1: 0.0
balance_2: 0.0
balance_3: 0.0
contrast: 0.0
brightness: 0.0
[color.match_hist]
threshold: 99.0
[mask.box_blend]
type: gaussian
distance: 11.0
radius: 5.0
passes: 1
[mask.mask_blend]
type: normalized
kernel_size: 3
passes: 4
threshold: 4
erosion: 0.0
[scaling.sharpen]
method: none
amount: 150
radius: 0.3
threshold: 5.0
[writer.ffmpeg]
container: mp4
codec: libx264
crf: 23
preset: medium
tune: none
profile: auto
level: auto
skip_mux: False
[writer.gif]
fps: 25
loop: 0
palettesize: 256
subrectangles: False
[writer.opencv]
format: png
draw_transparent: False
jpg_quality: 75
png_compress_level: 3
[writer.pillow]
format: png
draw_transparent: False
optimize: False
gif_interlace: True
jpg_quality: 75
png_compress_level: 3
tif_compression: tiff_deflate
--------- extract.ini ---------
[global]
allow_growth: False
[align.fan]
batch-size: 12
[detect.cv2_dnn]
confidence: 50
[detect.mtcnn]
minsize: 20
threshold_1: 0.6
threshold_2: 0.7
threshold_3: 0.7
scalefactor: 0.709
batch-size: 8
[detect.s3fd]
confidence: 70
batch-size: 4
[mask.unet_dfl]
batch-size: 8
[mask.vgg_clear]
batch-size: 6
[mask.vgg_obstructed]
batch-size: 2
--------- gui.ini ---------
[global]
fullscreen: False
tab: extract
options_panel_width: 30
console_panel_height: 20
icon_size: 14
font: default
font_size: 9
autosave_last_session: prompt
timeout: 120
auto_load_model_stats: True
--------- train.ini ---------
[global]
centering: face
coverage: 68.75
icnr_init: False
conv_aware_init: True
optimizer: adam
learning_rate: 5e-05
reflect_padding: False
allow_growth: True
mixed_precision: False
convert_batchsize: 16
[global.loss]
loss_function: ssim
mask_loss_function: mse
l2_reg_term: 100
eye_multiplier: 3
mouth_multiplier: 2
penalized_mask_loss: True
mask_type: extended
mask_blur_kernel: 3
mask_threshold: 4
learn_mask: False
[model.dfaker]
output_size: 128
[model.dfl_h128]
lowmem: False
[model.dfl_sae]
input_size: 128
clipnorm: True
architecture: df
autoencoder_dims: 0
encoder_dims: 42
decoder_dims: 21
multiscale_decoder: False
[model.dlight]
features: best
details: good
output_size: 256
[model.original]
lowmem: False
[model.realface]
input_size: 64
output_size: 128
dense_nodes: 1536
complexity_encoder: 128
complexity_decoder: 512
[model.unbalanced]
input_size: 128
lowmem: False
clipnorm: True
nodes: 1024
complexity_encoder: 128
complexity_decoder_a: 384
complexity_decoder_b: 512
[model.villain]
lowmem: False
[trainer.original]
preview_images: 14
zoom_amount: 5
rotation_range: 10
shift_range: 5
flip_chance: 50
disable_warp: False
color_lightness: 30
color_ab: 8
color_clahe_chance: 50
color_clahe_max_size: 4
The following information shows up when I am executing "Train" function ,
Code: Select all
Loading...
Setting Faceswap backend to NVIDIA
01/09/2021 10:12:49 INFO Log level set to: INFO
01/09/2021 10:12:51 INFO Model A Directory: E:\videodata\1
01/09/2021 10:12:51 INFO Model B Directory: E:\videodata\2
01/09/2021 10:12:51 INFO Training data directory: E:\videodata\4
01/09/2021 10:12:51 INFO ===================================================
01/09/2021 10:12:51 INFO Starting
01/09/2021 10:12:51 INFO Press 'Stop' to save and quit
01/09/2021 10:12:51 INFO ===================================================
01/09/2021 10:12:52 INFO Loading data, this may take a while...
01/09/2021 10:12:52 INFO Loading Model from Dfl_H128 plugin...
01/09/2021 10:12:52 INFO Using configuration saved in state file
01/09/2021 10:12:52 INFO Setting allow growth for GPU: PhysicalDevice(name='/physical_device:GPU:0', device_type='GPU')
01/09/2021 10:12:54 INFO Loaded model from disk: 'E:\videodata\4\dfl_h128.h5'
01/09/2021 10:12:54 INFO Loading Trainer from Original plugin...
Reading training images (A): 0%| | 0/368 [00:00<?, ?it/s]
Reading training images (A): 52%|█████▏ | 190/368 [00:00<00:00, 1825.29it/s]
Reading training images (B): 0%| | 0/388 [00:00<?, ?it/s]
Reading training images (B): 54%|█████▍ | 210/388 [00:00<00:00, 2056.98it/s]
01/09/2021 10:12:55 INFO Reading alignments from: 'E:\videodata\1\8_alignments.fsa'
01/09/2021 10:12:55 INFO Reading alignments from: 'E:\videodata\2\BBB_alignments.fsa'Re: [Guide] Using Faceswap on Nvidia RTX 30xx cards
Posted: Sat Jan 09, 2021 2:24 am
This is correct. See here:
Re: [Guide] Using Faceswap on Nvidia RTX 30xx cards
Posted: Sat Jan 09, 2021 2:25 am
See here:
app.php/faqpage#f0r3
Re: [Guide] Using Faceswap on Nvidia RTX 30xx cards
Posted: Sat Jan 09, 2021 10:44 pm
[mention]elon[/mention] How did you fix the freezing issues?
I have a 3070 and my Training freezes randomly on Dfl-H128, reducing Batch Size doesn't seem to help either.
Re: [Guide] Using Faceswap on Nvidia RTX 30xx cards
Posted: Sun Jan 10, 2021 2:14 am
Using Win10
Nvidia 457.51 (Do not use 460.xx - bad juju)
Only : 3060Ti + 2070
Followed post exactly.
Kept getting the "brotli" error, no matter how many drivers/software I uninstalled, environments removed.
Google search gave me:
Code: Select all
conda install -c anaconda urllib3and it worked.
3060Ti now running.
Although with both cards I'm getting a strangely low possible batch and only a few 1000 iteration before a OOM crash.
Getting a batch of 3-5 with either card.
Normally with Villian get 1214 batch.
Still 2070 works fantastic with a normal Faceswap install.
At a batch of 3, I can tell the 3060Ti is at minimum 12% faster than a 2070.
Code: Select all
============ System Information ============
encoding: cp1252
git_branch: master
git_commits: 163b8fe sysinfo/setup - Standardize cuda/cudnn checks
gpu_cuda: 11.0
gpu_cudnn: 8.0.5
gpu_devices: GPU_0: GeForce RTX 3060 Ti, GPU_1: GeForce RTX 2070
gpu_devices_active: GPU_0, GPU_1
gpu_driver: 457.51
gpu_vram: GPU_0: 8192MB, GPU_1: 8192MB
os_machine: AMD64
os_platform: Windows-10-10.0.19041-SP0
os_release: 10
py_command: C:\Users\abigf\faceswap/faceswap.py gui
py_conda_version: conda 4.9.2
py_implementation: CPython
py_version: 3.8.5
py_virtual_env: True
sys_cores: 12
sys_processor: AMD64 Family 23 Model 8 Stepping 2, AuthenticAMD
sys_ram: Total: 32712MB, Available: 26730MB, Used: 5982MB, Free: 26730MB
=============== Pip Packages ===============
absl-py==0.11.0
astunparse==1.6.3
brotlipy==0.7.0
cachetools==4.2.0
certifi==2020.6.20
cffi @ file:///C:/ci/cffi_1600699246375/work
chardet==4.0.0
cryptography @ file:///C:/ci/cryptography_1601046905460/work
cycler==0.10.0
fastcluster==1.1.26
ffmpy==0.2.3
flatbuffers==1.12
gast==0.3.3
google-auth==1.24.0
google-auth-oauthlib==0.4.2
google-pasta==0.2.0
grpcio==1.32.0
h5py==2.10.0
idna @ file:///tmp/build/80754af9/idna_1593446292537/work
imageio @ file:///tmp/build/80754af9/imageio_1594161405741/work
imageio-ffmpeg @ file:///home/conda/feedstock_root/build_artifacts/imageio-ffmpeg_1609799311556/work
joblib @ file:///tmp/build/80754af9/joblib_1607970656719/work
Keras-Preprocessing==1.1.2
kiwisolver @ file:///C:/ci/kiwisolver_1604014703538/work
Markdown==3.3.3
matplotlib @ file:///C:/ci/matplotlib-base_1592837548929/work
mkl-fft==1.2.0
mkl-random==1.1.1
mkl-service==2.3.0
numpy @ file:///C:/ci/numpy_and_numpy_base_1603466732592/work
nvidia-ml-py3 @ git+https://github.com/deepfakes/nvidia-ml-py3.git@6fc29ac84b32bad877f078cb4a777c1548a00bf6
oauthlib==3.1.0
olefile==0.46
opencv-python==4.5.1.48
opt-einsum==3.3.0
pathlib==1.0.1
Pillow @ file:///C:/ci/pillow_1609786840597/work
protobuf==3.14.0
psutil @ file:///C:/ci/psutil_1598370330503/work
pyasn1==0.4.8
pyasn1-modules==0.2.8
pycparser @ file:///tmp/build/80754af9/pycparser_1594388511720/work
pyOpenSSL @ file:///tmp/build/80754af9/pyopenssl_1594392929924/work
pyparsing==2.4.7
PySocks==1.7.1
python-dateutil==2.8.1
pywin32==227
requests==2.25.1
requests-oauthlib==1.3.0
rsa==4.6
scikit-learn @ file:///C:/ci/scikit-learn_1598377018496/work
scipy @ file:///C:/ci/scipy_1604596260408/work
sip==4.19.13
six @ file:///C:/ci/six_1605187374963/work
tensorboard==2.4.0
tensorboard-plugin-wit==1.7.0
tensorflow-estimator==2.4.0
tensorflow-gpu==2.4.0
termcolor==1.1.0
threadpoolctl @ file:///tmp/tmp9twdgx9k/threadpoolctl-2.1.0-py3-none-any.whl
tornado @ file:///C:/ci/tornado_1606942392901/work
tqdm @ file:///tmp/build/80754af9/tqdm_1609788246169/work
typing-extensions==3.7.4.3
urllib3 @ file:///tmp/build/80754af9/urllib3_1603305693037/work
Werkzeug==1.0.1
win-inet-pton==1.1.0
wincertstore==0.2
wrapt==1.12.1
============== Conda Packages ==============
# packages in environment at C:\Users\abigf\MiniConda3\envs\faceswap:
#
# Name Version Build Channel
absl-py 0.11.0 pypi_0 pypi
astunparse 1.6.3 pypi_0 pypi
blas 1.0 mkl
brotli 1.0.9 ha925a31_2
brotlipy 0.7.0 py38he774522_1000 anaconda
ca-certificates 2020.10.14 0 anaconda
cachetools 4.2.0 pypi_0 pypi
certifi 2020.6.20 py38_0 anaconda
cffi 1.14.3 py38h7a1dbc1_0 anaconda
chardet 4.0.0 pypi_0 pypi
cryptography 3.1.1 py38h7a1dbc1_0 anaconda
cycler 0.10.0 py38_0
fastcluster 1.1.26 py38h251f6bf_2 conda-forge
ffmpeg 4.2.2 he774522_0
ffmpy 0.2.3 pypi_0 pypi
flatbuffers 1.12 pypi_0 pypi
freetype 2.10.4 hd328e21_0
gast 0.3.3 pypi_0 pypi
git 2.23.0 h6bb4b03_0
google-auth 1.24.0 pypi_0 pypi
google-auth-oauthlib 0.4.2 pypi_0 pypi
google-pasta 0.2.0 pypi_0 pypi
grpcio 1.32.0 pypi_0 pypi
h5py 2.10.0 pypi_0 pypi
icc_rt 2019.0.0 h0cc432a_1
icu 58.2 ha925a31_3
idna 2.10 py_0 anaconda
imageio 2.9.0 py_0
imageio-ffmpeg 0.4.3 pyhd8ed1ab_0 conda-forge
intel-openmp 2020.2 254
joblib 1.0.0 pyhd3eb1b0_0
jpeg 9b hb83a4c4_2
keras-preprocessing 1.1.2 pypi_0 pypi
kiwisolver 1.3.0 py38hd77b12b_0
libpng 1.6.37 h2a8f88b_0
libtiff 4.1.0 h56a325e_1
lz4-c 1.9.2 hf4a77e7_3
markdown 3.3.3 pypi_0 pypi
matplotlib 3.2.2 0
matplotlib-base 3.2.2 py38h64f37c6_0
mkl 2020.2 256
mkl-service 2.3.0 py38h196d8e1_0
mkl_fft 1.2.0 py38h45dec08_0
mkl_random 1.1.1 py38h47e9c7a_0
numpy 1.19.2 py38hadc3359_0
numpy-base 1.19.2 py38ha3acd2a_0
nvidia-ml-py3 7.352.1 pypi_0 pypi
oauthlib 3.1.0 pypi_0 pypi
olefile 0.46 py_0
opencv-python 4.5.1.48 pypi_0 pypi
openssl 1.1.1h he774522_0 anaconda
opt-einsum 3.3.0 pypi_0 pypi
pathlib 1.0.1 py_1
pillow 8.1.0 py38h4fa10fc_0
pip 20.3.3 py38haa95532_0
protobuf 3.14.0 pypi_0 pypi
psutil 5.7.2 py38he774522_0
pyasn1 0.4.8 pypi_0 pypi
pyasn1-modules 0.2.8 pypi_0 pypi
pycparser 2.20 py_2 anaconda
pyopenssl 19.1.0 py_1 anaconda
pyparsing 2.4.7 py_0
pyqt 5.9.2 py38ha925a31_4
pysocks 1.7.1 py38_0 anaconda
python 3.8.5 h5fd99cc_1
python-dateutil 2.8.1 py_0
python_abi 3.8 1_cp38 conda-forge
pywin32 227 py38he774522_1
qt 5.9.7 vc14h73c81de_0
requests 2.25.1 pypi_0 pypi
requests-oauthlib 1.3.0 pypi_0 pypi
rsa 4.6 pypi_0 pypi
scikit-learn 0.23.2 py38h47e9c7a_0
scipy 1.5.2 py38h14eb087_0
setuptools 51.0.0 py38haa95532_2
sip 4.19.13 py38ha925a31_0
six 1.15.0 py38haa95532_0
sqlite 3.33.0 h2a8f88b_0
tensorboard 2.4.0 pypi_0 pypi
tensorboard-plugin-wit 1.7.0 pypi_0 pypi
tensorflow-estimator 2.4.0 pypi_0 pypi
tensorflow-gpu 2.4.0 pypi_0 pypi
termcolor 1.1.0 pypi_0 pypi
threadpoolctl 2.1.0 pyh5ca1d4c_0
tk 8.6.10 he774522_0
tornado 6.1 py38h2bbff1b_0
tqdm 4.55.1 pyhd3eb1b0_0
typing-extensions 3.7.4.3 pypi_0 pypi
urllib3 1.25.11 py_0 anaconda
vc 14.2 h21ff451_1
vs2015_runtime 14.27.29016 h5e58377_2
werkzeug 1.0.1 pypi_0 pypi
wheel 0.36.2 pyhd3eb1b0_0
win_inet_pton 1.1.0 py38_0 anaconda
wincertstore 0.2 py38_0
wrapt 1.12.1 pypi_0 pypi
xz 5.2.5 h62dcd97_0
zlib 1.2.11 h62dcd97_4
zstd 1.4.5 h04227a9_0
================= Configs ==================
--------- .faceswap ---------
backend: NVIDIA
--------- convert.ini ---------
[color.color_transfer]
clip: True
preserve_paper: True
[color.manual_balance]
colorspace: HSV
balance_1: 0.0
balance_2: 0.0
balance_3: 0.0
contrast: 0.0
brightness: 0.0
[color.match_hist]
threshold: 99.0
[mask.box_blend]
type: gaussian
distance: 11.0
radius: 5.0
passes: 1
[mask.mask_blend]
type: normalized
kernel_size: 3
passes: 4
threshold: 4
erosion: 0.0
[scaling.sharpen]
method: none
amount: 150
radius: 0.3
threshold: 5.0
[writer.ffmpeg]
container: mp4
codec: libx264
crf: 23
preset: medium
tune: none
profile: auto
level: auto
skip_mux: False
[writer.gif]
fps: 25
loop: 0
palettesize: 256
subrectangles: False
[writer.opencv]
format: png
draw_transparent: False
jpg_quality: 75
png_compress_level: 3
[writer.pillow]
format: png
draw_transparent: False
optimize: False
gif_interlace: True
jpg_quality: 75
png_compress_level: 3
tif_compression: tiff_deflate
--------- extract.ini ---------
[global]
allow_growth: False
[align.fan]
batch-size: 12
[detect.cv2_dnn]
confidence: 50
[detect.mtcnn]
minsize: 20
threshold_1: 0.6
threshold_2: 0.7
threshold_3: 0.7
scalefactor: 0.709
batch-size: 8
[detect.s3fd]
confidence: 70
batch-size: 4
[mask.unet_dfl]
batch-size: 8
[mask.vgg_clear]
batch-size: 6
[mask.vgg_obstructed]
batch-size: 2
--------- gui.ini ---------
[global]
fullscreen: False
tab: extract
options_panel_width: 30
console_panel_height: 20
icon_size: 14
font: default
font_size: 9
autosave_last_session: prompt
timeout: 120
auto_load_model_stats: True
--------- train.ini ---------
[global]
centering: face
coverage: 69.09
icnr_init: False
conv_aware_init: False
optimizer: adam
learning_rate: 5e-05
reflect_padding: False
allow_growth: True
mixed_precision: True
convert_batchsize: 16
[global.loss]
loss_function: ssim
mask_loss_function: mse
l2_reg_term: 100
eye_multiplier: 3
mouth_multiplier: 2
penalized_mask_loss: True
mask_type: extended
mask_blur_kernel: 3
mask_threshold: 4
learn_mask: False
[model.dfaker]
output_size: 128
[model.dfl_h128]
lowmem: False
[model.dfl_sae]
input_size: 128
clipnorm: True
architecture: df
autoencoder_dims: 0
encoder_dims: 42
decoder_dims: 21
multiscale_decoder: False
[model.dlight]
features: best
details: good
output_size: 256
[model.original]
lowmem: False
[model.realface]
input_size: 64
output_size: 128
dense_nodes: 1536
complexity_encoder: 128
complexity_decoder: 512
[model.unbalanced]
input_size: 128
lowmem: False
clipnorm: True
nodes: 1024
complexity_encoder: 128
complexity_decoder_a: 384
complexity_decoder_b: 512
[model.villain]
lowmem: False
[trainer.original]
preview_images: 14
zoom_amount: 5
rotation_range: 10
shift_range: 5
flip_chance: 50
disable_warp: False
color_lightness: 30
color_ab: 8
color_clahe_chance: 50
color_clahe_max_size: 4Slow extraction
Posted: Sun Jan 10, 2021 6:49 pm
Hello
i have 3080 gpu and face extraction is really slow only 2,38 it/s
followed all instructions here app.php/faqpage#f1r1 and here viewtopic.php?t=1226 but it is still slow.
thank you
Re: [Guide] Using Faceswap on Nvidia RTX 30xx cards
Posted: Mon Jan 11, 2021 2:10 am
Thank you for the guide! it works for me in the following configuration:
Cuda 11.1.1
cuDNN 8.0.5
TensorFlow 2.4
Nvidia driver 461.09 (latest)
I haven't done any extensive testing but it starts to train and I can see the preview and I can see that it's going through iterations. 
Re: [Guide] Using Faceswap on Nvidia RTX 30xx cards
Posted: Mon Jan 11, 2021 7:51 pm
ok so the extraction does work around 25 it/s it also ends with extracted images
but i have some errors
Code: Select all
2021-01-11 20:49:03.683501: E tensorflow/stream_executor/stream.cc:5011] Internal: Failed to enqueue async memset operation: CUDA_ERROR_INVALID_VALUE: invalid argument
2021-01-11 20:49:03.684211: E tensorflow/stream_executor/cuda/cuda_driver.cc:1051] failed to enqueue async memcpy from device to host: CUDA_ERROR_INVALID_VALUE: invalid argument; host dst: 0x65dfb3c9e0; GPU src: (nil); size: 8=0x8
2021-01-11 20:49:03.684536: E tensorflow/stream_executor/cuda/cuda_blas.cc:226] failed to create cublas handle: CUBLAS_STATUS_ALLOC_FAILED
2021-01-11 20:49:03.686348: F tensorflow/core/common_runtime/gpu/gpu_util.cc:291] GPU->CPU Memcpy failedor
Code: Select all
2021-01-11 20:52:44.133832: E tensorflow/stream_executor/cuda/cuda_blas.cc:226] failed to create cublas handle: CUBLAS_STATUS_ALLOC_FAILED
2021-01-11 20:52:44.446660: E tensorflow/stream_executor/cuda/cuda_blas.cc:226] failed to create cublas handle: CUBLAS_STATUS_ALLOC_FAILED
2021-01-11 20:52:44.761302: E tensorflow/stream_executor/cuda/cuda_blas.cc:226] failed to create cublas handle: CUBLAS_STATUS_ALLOC_FAILEDRe: [Guide] Using Faceswap on Nvidia RTX 30xx cards
Posted: Thu Jan 21, 2021 10:16 pm
Any updates on 3080/3090 support in W10?
Is faceswap still unsupported/untested for these new cards or are they more stable now?
Re: [Guide] Using Faceswap on Nvidia RTX 30xx cards
Posted: Fri Jan 22, 2021 11:47 am
Until Anaconda Cloud supports Tensorflow 2.4, this is out of our hands beyond the instructions detailed in the first post:
Re: [Guide] Using Faceswap on Nvidia RTX 30xx cards
Posted: Wed Jan 27, 2021 5:21 am
for Win10 Gforce game Ready Driver 461.40 seems to not break anything new with current manual install described in the first post.
Re: [Guide] Using Faceswap on Nvidia RTX 30xx cards
Posted: Sat Jan 30, 2021 2:53 am
RTX 3090 / 461.40 drivers
I needed the brotli fix: conda install -c anaconda urllib3
Then installed with Cuda Toolkit 11.1 and the newer cuDNN. After that no errors.
However, training is extremely slow. Like 1 iteration every 10 seconds regardless of batch size.
I am getting this warning from previously extracted faces (on a 1070 card)
You are using legacy extracted faces but have selected 'face' centering which is incompatible. Switching centering to 'legacy'
Could that be the reason for the extremely slow performance?