[GH-ISSUE #278] Does rvc support training with cpu? #136

Author

Owner

@kalomaze commented on GitHub (May 13, 2023):

CPU inference of most AI models is already pretty slow as is. I don't think there is any popular machine learning software that has CPU training as an option. It would probably take weeks to get a decent result.

@kalomaze commented on GitHub (May 13, 2023): CPU inference of most AI models is already pretty slow as is. I don't think there is any popular machine learning software that has CPU training as an option. It would probably take weeks to get a decent result.

bsakaguchi commented

Author

Owner

@fumiama commented on GitHub (May 14, 2023):

Maybe add later.

@fumiama commented on GitHub (May 14, 2023): Maybe add later.

bsakaguchi commented

Author

Owner

@justinjohn0306 commented on GitHub (May 14, 2023):

@c469591 use google colab

@justinjohn0306 commented on GitHub (May 14, 2023): @c469591 use google colab

bsakaguchi commented

Author

Owner

@fumiama commented on GitHub (May 15, 2023):

Yes, that's also a solution.

@fumiama commented on GitHub (May 15, 2023): Yes, that's also a solution.

bsakaguchi commented

Author

Owner

@c469591 commented on GitHub (May 15, 2023):

I still hope this will be added to the new version.
Thanks!

@c469591 commented on GitHub (May 15, 2023): I still hope this will be added to the new version. Thanks!

bsakaguchi commented

Author

Owner

@Sovenok-Hacker commented on GitHub (Jul 12, 2023):

I still hope this will be added to the new version. Thanks!

Yes. I have the same problem. I want to train a model but I don`t have a GPU.

@Sovenok-Hacker commented on GitHub (Jul 12, 2023): > I still hope this will be added to the new version. Thanks! Yes. I have the same problem. I want to train a model but I don`t have a GPU.

bsakaguchi commented

Author

Owner

@RonFusaishi commented on GitHub (Sep 2, 2023):

considering google now banning RVC code, i think something like this is better than nothing.

@RonFusaishi commented on GitHub (Sep 2, 2023): considering google now banning RVC code, i think something like this is better than nothing.

bsakaguchi commented

Author

Owner

@Sovenok-Hacker commented on GitHub (Sep 2, 2023):

considering google now banning RVC code, i think something like this is better than nothing.

Google is nuts

@Sovenok-Hacker commented on GitHub (Sep 2, 2023): > considering google now banning RVC code, i think something like this is better than nothing. Google is nuts

bsakaguchi commented

Author

Owner

@RonFusaishi commented on GitHub (Sep 2, 2023):

Google is nuts

google and greed are one of the same thing unfortunately, you can still run it but only if you have a colab pro.

I personally blame the influx of zoomers abusing the damn thing just to spam youtube with their degenerate garbage, i saw this coming from a mile away.

@RonFusaishi commented on GitHub (Sep 2, 2023): > Google is nuts google and greed are one of the same thing unfortunately, you can still run it but only if you have a colab pro. I personally blame the influx of zoomers abusing the damn thing just to spam youtube with their degenerate garbage, i saw this coming from a mile away.

bsakaguchi commented

Author

Owner

@yapvoonyee commented on GitHub (Sep 8, 2023):

I would like a cpu only version because I have an intel GPU.

@yapvoonyee commented on GitHub (Sep 8, 2023): I would like a cpu only version because I have an intel GPU.

bsakaguchi commented

Author

Owner

@fumiama commented on GitHub (Sep 8, 2023):

I would like a cpu only version because I have an intel GPU.

You can try the dml or wait #278

@fumiama commented on GitHub (Sep 8, 2023): > I would like a cpu only version because I have an intel GPU. You can try the `dml` or wait #278

bsakaguchi commented

Author

Owner

@loboere commented on GitHub (Sep 22, 2023):

I have a GPU and I want to infer with a CPU, because my GPU gives cuda errors, is there any parameter so that RVC does not detect the GPU?

@loboere commented on GitHub (Sep 22, 2023): I have a GPU and I want to infer with a CPU, because my GPU gives cuda errors, is there any parameter so that RVC does not detect the GPU?

bsakaguchi commented

Author

Owner

@LoloCodes commented on GitHub (Oct 5, 2023):

so can we use cpus to train or not?

@LoloCodes commented on GitHub (Oct 5, 2023): so can we use cpus to train or not?

bsakaguchi commented

Author

Owner

@treya-lin commented on GitHub (Nov 9, 2023):

It can. At least worked when tested on my linux server.
My server actually has gpu on it. To disable gpu, I added os.environ["CUDA_VISIBLE_DEVICES"] = "-1" before importing torch in infer/modules/train/train.py

from infer.lib.train import utils

hps = utils.get_hparams()
# os.environ["CUDA_VISIBLE_DEVICES"] = hps.gpus.replace("-", ",")
os.environ["CUDA_VISIBLE_DEVICES"] = "-1" # set this to "-1" if you wanna test cpu training on a server with gpu. Training with cpu does work, but it is extremely slow.

As for the speed, according to my very brief experience, for an epoch which costs 20s on a 1080 Ti GPU, it took about 2min on cpu.

@treya-lin commented on GitHub (Nov 9, 2023): It can. At least worked when tested on my linux server. My server actually has gpu on it. To disable gpu, I added `os.environ["CUDA_VISIBLE_DEVICES"] = "-1"` before importing torch in `infer/modules/train/train.py` ``` from infer.lib.train import utils hps = utils.get_hparams() # os.environ["CUDA_VISIBLE_DEVICES"] = hps.gpus.replace("-", ",") os.environ["CUDA_VISIBLE_DEVICES"] = "-1" # set this to "-1" if you wanna test cpu training on a server with gpu. Training with cpu does work, but it is extremely slow. ``` As for the speed, according to my very brief experience, for an epoch which costs 20s on a 1080 Ti GPU, it took about 2min on cpu.

bsakaguchi commented

Author

Owner

@PurpBatBoi commented on GitHub (Jan 25, 2024):

It can. At least worked when tested on my linux server. My server actually has gpu on it. To disable gpu, I added os.environ["CUDA_VISIBLE_DEVICES"] = "-1" before importing torch in infer/modules/train/train.py
from infer.lib.train import utils

hps = utils.get_hparams()
# os.environ["CUDA_VISIBLE_DEVICES"] = hps.gpus.replace("-", ",")
os.environ["CUDA_VISIBLE_DEVICES"] = "-1" # set this to "-1" if you wanna test cpu training on a server with gpu. Training with cpu does work, but it is extremely slow.
As for the speed, according to my very brief experience, for an epoch which costs 20s on a 1080 Ti GPU, it took about 2min on cpu.

slow but still useful to ppl who wanna make models

@PurpBatBoi commented on GitHub (Jan 25, 2024): > It can. At least worked when tested on my linux server. My server actually has gpu on it. To disable gpu, I added `os.environ["CUDA_VISIBLE_DEVICES"] = "-1"` before importing torch in `infer/modules/train/train.py` > > ``` > from infer.lib.train import utils > > hps = utils.get_hparams() > # os.environ["CUDA_VISIBLE_DEVICES"] = hps.gpus.replace("-", ",") > os.environ["CUDA_VISIBLE_DEVICES"] = "-1" # set this to "-1" if you wanna test cpu training on a server with gpu. Training with cpu does work, but it is extremely slow. > ``` > > As for the speed, according to my very brief experience, for an epoch which costs 20s on a 1080 Ti GPU, it took about 2min on cpu. slow but still useful to ppl who wanna make models

bsakaguchi commented

Author

Owner

@PurpBatBoi commented on GitHub (Jan 25, 2024):

I would like a cpu only version because I have an intel GPU.

You can try the dml or wait #278

what is 'dml'?

@PurpBatBoi commented on GitHub (Jan 25, 2024): > > I would like a cpu only version because I have an intel GPU. > > You can try the `dml` or wait #278 what is 'dml'?

bsakaguchi commented

Author

Owner

@thefellaguy commented on GitHub (May 14, 2024):

I'm know I'm late to the game, but I've trained on CPU only with no issues. I let it run on all night on 6 cores and it gets the job done in 8-10 hours on harvest. Converting audio is about a 1:2 ratio. a 10 audio clip tales about 20 seconds to convert.

@thefellaguy commented on GitHub (May 14, 2024): I'm know I'm late to the game, but I've trained on CPU only with no issues. I let it run on all night on 6 cores and it gets the job done in 8-10 hours on harvest. Converting audio is about a 1:2 ratio. a 10 audio clip tales about 20 seconds to convert.

bsakaguchi commented

Author

Owner

@GOSFO commented on GitHub (Aug 23, 2024):

I'm know I'm late to the game, but I've trained on CPU only with no issues. I let it run on all night on 6 cores and it gets the job done in 8-10 hours on harvest. Converting audio is about a 1:2 ratio. a 10 audio clip tales about 20 seconds to convert.

How did you configure your rvc?

@GOSFO commented on GitHub (Aug 23, 2024): > I'm know I'm late to the game, but I've trained on CPU only with no issues. I let it run on all night on 6 cores and it gets the job done in 8-10 hours on harvest. Converting audio is about a 1:2 ratio. a 10 audio clip tales about 20 seconds to convert. How did you configure your rvc?

bsakaguchi commented