DaGAN Windows Installation Tutorial

My main video:

Installation GitHub:

This video is supported by the kind Patrons & YouTube Members:
🙏Andrew Lescelius, Chris LeDoux, Dan Kennedy, 正毅 水野, Shawn77134, Panther Modern, Jake Disco, Demilson Quintao, Tony Jimenez, Shuhong Chen, Hongbo Men, happi nyuu nyaa, Carol Lo, Mose Sakashita, Miguel, Bandera, Gennaro Schiano, gunwoo, Ravid Freedman, Mert Seftali, Mrityunjay

[giffinr]

Thanks, this has been hugely helpful. Being able to take an AI generated character and bring them to live with this approach is so much fun.

[Mobay]

Definitely going to try this out, thanks for the Windows tutorial!

[SyntheticVoices]

Awesome - got it working a few reinstalls!

[daffertube]

ERROR: Ignored the following versions that require a different python version: 1.1.0 Requires-Python >=3.8; ect requires python version ect python version ect

python --version shows Python 3.9.13. Anybody have a guess?

[TrggrWarning]

Huh damn cloudhead opens up world of possibilities and gets lazy! Lol j/k we gotta learn to walk too. The folder typo, for some even if seen might not register as problem, happy you kept it in. I’m looking to make 3D animated shortcuts, in “real” space videos this looks promising, perhaps requiring some forethought, need to put it together and experiment. Never seen GitHub used this way, are we compiling or building an instance? Lol feels like stumbling around in the dark to have ai animate in the dark.

[ChristianIce]

I've got this error:

raise RuntimeError('Attempting to deserialize object on a CUDA '
RuntimeError: Attempting to deserialize object on a CUDA device but torch.cuda.is_available() is False. If you are running on a CPU-only machine, please use torch.load with to map your storages to the CPU.

I am using a 2080Ti.
Please help

[reesebenson]

Hey, Really excited to give this a go. I am having an issue though. It doesn't matter if I use the --cpu flag or not I get this error:

UserWarning: Failed to load image Python extension:
warn(f"Failed to load image Python extension: {e}")
UserWarning: Using 'weights' as positional parameter(s) is deprecated since 0.13 and will be removed in 0.15. Please use keyword parameter(s) instead.
f"Using {sequence_to_str(tuple(keyword_only_kwargs.keys()), separate_last='and ')} as positional "
UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and will be removed in 0.15. The current behavior is equivalent to passing `weights=None`.
warnings.warn(msg)
Traceback (most recent call last):
File "demo.py", line 165, in <module>
loaded_dict_enc =
File "C:REDACTED\anaconda3\envs\DaGAN\lib\site-packages\torch\serialization.py", line 712, in load
return _load(opened_zipfile, map_location, pickle_module, **pickle_load_args)
File "C:REDACTED\anaconda3\envs\DaGAN\lib\site-packages\torch\serialization.py", line 1049, in _load
result = unpickler.load()
File "C:REDACTED\anaconda3\envs\DaGAN\lib\site-packages\torch\serialization.py", line 1019, in persistent_load
load_tensor(dtype, nbytes, key,
File "C:REDACTED\anaconda3\envs\DaGAN\lib\site-packages\torch\serialization.py", line 1001, in load_tensor
wrap_storage=restore_location(storage, location),
File "C:REDACTED\anaconda3\envs\DaGAN\lib\site-packages\torch\serialization.py", line 175, in default_restore_location
result = fn(storage, location)
File "C:REDACTED\anaconda3\envs\DaGAN\lib\site-packages\torch\serialization.py", line 152, in _cuda_deserialize
device =
File "C:REDACTED\anaconda3\envs\DaGAN\lib\site-packages\torch\serialization.py", line 136, in validate_cuda_device
raise RuntimeError('Attempting to deserialize object on a CUDA '
RuntimeError: Attempting to deserialize object on a CUDA device but torch.cuda.is_available() is False. If you are running on a CPU-only machine, please use torch.load with to map your storages to the CPU.

Not sure what I am missing. I went into the code and it looks like when the flag is set it should be using torch.load with
Hope someone can help!

[RyanClark-gryb]

Do you need to have a super powerful GPU for this? I mean what do you think is a reasonable computational power level?

[HassanalBolkiahSoyjak]

Finally one that works on windows, updooted

[bryantmorris]

Can you showhow to install animeganv3 and also do you need a gpu to run it?

[mgphotoart]

i got this message, dont know what to do? can anyone help?: RuntimeError: Attempting to deserialize object on a CUDA device but torch.cuda.is_available() is False. If you are running on a CPU-only machine, please use torch.load with to map your storages to the CPU.

[techthinkerstrade]

crop video for face alignment is not outputting command to execute .. is it internally executing and storing the result somewhere ?

[AdrianJensen]

Man I'm following the instructions as close as I can and it just won't work. I can't even get the web demo to work, all I get is errors. I've been trying to make at least one of these things work all day. Any advice?

[BenjaminGulmez]

Thanks for that, do we have a possibility to generate outputs larger than 256x256 pixels?

[treelibrarian]

What do you think the chances/hardware requirements of getting this working in realtime from a webcam are? Thinking it could be an alt method for Vtuber-style streaming...

[TheGamingR]

'pip install torchaudio==0.10.1+rocm4.' isn't working, it seems the file doesn't exists
edit: I was able to install it without the '+rocm4'

[aaronmarkstaller]

Where do I get videos of people talking or singing centered on their face?

my google searches just come up with android apps that try to make conversions

[deathseverance]

- For both normal & spade checkpoints i have the same error :

Traceback (most recent call last):
File "demo.py", line 1, in <module>
import matplotlib
ModuleNotFoundError: No module named 'matplotlib'


idk what to do with it, any idea ?

[aaronmarkstaller]

It doesn't work for me.
First it came up with error "no module named 'matplotlib'"
so I pip installed
then "ERROR: No matching distribution found for yaml"

in the demo.py file

[vikramchary]

iam getting this
UserWarning: Using 'weights' as positional parameter(s) is deprecated since 0.13 and may be removed in the future. Please use keyword parameter(s) instead
f"Using {sequence_to_str(tuple(keyword_only_kwargs.keys()), separate_last='and ')} as positional "
UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=None`.
warnings.warn(msg)
Traceback (most recent call last):
File "demo.py", line 165, in <module>
loaded_dict_enc =
how can i slove