[Guide] Linux Install Guide: Installer Method

[torzdf]

Did anybody successfully install faceswap without conda?

Yes, I have. But it means managing your Cuda/cuDNN dependencies yourself, and we don't provide support for that, sadly.

[jiapei100]

Thank you ... @torzdf.
Oh... That's you, who replied my message from github?

Yeah, I've been managing my cuda and cudnn.

And successfully ran python setup.py build, which clearly allows me to build faceswap without conda. I mean, there is options in it.
But I cannot build a wheel out...

[torzdf]

Ah, that's because it's not a pypi package, so there is no wheel

Setup.py is just a convenience wrapper for the requirements.txt files and other things which can get installed through Conda.

If running in normal python (needs to be 3.10) then I would create a python venv virtual environment and just run

Code: Select all

pip install -r ./requirements/requirements_nvidia.txt

[jiapei100]

All runned already...

Just have NO idea why python faceswap.py gui won't pop up anything....

I found a file gui.py. It seems there is a chance to run python gui.py directly??
Still testing...

[jiapei100]

Code: Select all

➜  faceswap git:(master) ✗ python gui.py
Setting Faceswap backend to NVIDIA
~/.local/lib/python3.10/site-packages/numpy/core/getlimits.py:549: UserWarning: The value of the smallest subnormal for <class 'numpy.float64'> type is zero.
  setattr(self, word, getattr(machar, word).flat[0])
~/.local/lib/python3.10/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for <class 'numpy.float64'> type is zero.
  return self._float_to_str(self.smallest_subnormal)
~/.local/lib/python3.10/site-packages/numpy/core/getlimits.py:549: UserWarning: The value of the smallest subnormal for <class 'numpy.float32'> type is zero.
  setattr(self, word, getattr(machar, word).flat[0])
~/.local/lib/python3.10/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for <class 'numpy.float32'> type is zero.
  return self._float_to_str(self.smallest_subnormal)
2023-08-01 15:16:09.657119: E tensorflow/compiler/xla/stream_executor/cuda/cuda_dnn.cc:9342] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2023-08-01 15:16:09.657157: E tensorflow/compiler/xla/stream_executor/cuda/cuda_fft.cc:609] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2023-08-01 15:16:09.657167: E tensorflow/compiler/xla/stream_executor/cuda/cuda_blas.cc:1518] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2023-08-01 15:16:09.662649: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: SSE3 SSE4.1 SSE4.2 AVX AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
WARNING:tensorflow:From ~/.local/lib/python3.10/site-packages/tensorflow/python/ops/distributions/distribution.py:259: ReparameterizationType.__init__ (from tensorflow.python.ops.distributions.distribution) is deprecated and will be removed after 2019-01-01.
Instructions for updating:
The TensorFlow Distributions library has moved to TensorFlow Probability (https://github.com/tensorflow/probability). You should update all references to use `tfp.distributions` instead of `tf.distributions`.
WARNING:tensorflow:From ~/.local/lib/python3.10/site-packages/tensorflow/python/ops/distributions/bernoulli.py:165: RegisterKL.__init__ (from tensorflow.python.ops.distributions.kullback_leibler) is deprecated and will be removed after 2019-01-01.
Instructions for updating:
The TensorFlow Distributions library has moved to TensorFlow Probability (https://github.com/tensorflow/probability). You should update all references to use `tfp.distributions` instead of `tf.distributions`.

That is all what I obtained so far...

[torzdf]

No. Launching as you are is the correct way that is:

Code: Select all

python faceswap.py gui

Any other way simply will not work because that isn't how the program is built.

Ultimately, this is exactly why we standardise an install through Conda. Sure, it isn't for everyone, and has some major issues, but it standardises the install making these kinds of problems far easier to solve.

As it is, I have no idea about your individual Python setup, what's in your base environment, what site-packages you have, and which are available in the virtual environment, how your Cuda is installed, etc. etc. etc.

So, unfortunately, we have to take the approach that you are entirely entitled (and indeed capable) to install Faceswap whichever way you please, but at that point you are, unfortunately, on your own.

You are welcome to open a thread re: manual installs to see if anyone can help, but that's the best I can offer. As it is, this thread is for installing on Linux through the installer, so isn't the best place for this dicussion.

[jiapei100]

@torzdf

Done now... I‘ve got a 2nd questions . Check this: https://forum.faceswap.dev/viewtopic.php?t=2763

Thank you ...

[SoCalFoxx]

I have read all the forum post, I even googled to see if I was missing anything. I can not get it load. I tried moving to the ~/Downloads dir then add the code. I tried to add the ~/Download dir with the code at the same time and nothing works.

See error:

Code: Select all

chris@chris-MacBookPro:~$ ~/Downloads$ bash ./faceswap_setup_x64.sh
bash: /home/chris/Downloads$: No such file or directory
chris@chris-MacBookPro:~$ ~/Downloads
bash: /home/chris/Downloads: Is a directory
chris@chris-MacBookPro:~$ bash ./faceswap_setup_x64.sh
bash: ./faceswap_setup_x64.sh: No such file or directory
chris@chris-MacBookPro:~$

[torzdf]

Do you have the correct installer? This is the Linux thread, you need the macOS version:
viewtopic.php?t=2748

Then it's as simple as entering your Downloads folder in terminal and entering:

Code: Select all

bash faceswap_setup_macos.sh