New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
ERROR #1
Comments
This problem only arises in the val phase, since the sampleling() requires the cond parameter, but self.cond is not defined in the val phase, is this a bug? |
Hi @biqigubafan , sorry but I cannot reproduce this error. In my running, there was no issue on val sampling. Could you please make sure that you used the latest version of the code? There have been some updates in the repo without leaving the commit logs (I initialized all things as it's just born). |
Hello! I downloaded your code on Sunday. I found that in the val stage of “shallow”, called diffusion.sampling(), but sampling() needs to use “cond”. Since self.cond is defined in forward, at this time diffusion.sampling () does not run forward, so “cond” will show undefined, which is my guess, and I don't understand it very well, but I added the cond parameter to sampling() and changed the relevant parts, it can run normally. |
I see. Did you follow the steps in README? You're supposed to train 'shallow' model after training 'aux' model, so that the 'shallow' model can leverage the pre-trained 'aux' model during its training. |
I follow the steps try again,thanks! |
|
hi , i also met the problem when i train naive, may you tell me how to modify the code? 3q |
Have you solved it? Can you share the code you modified? This is really important to me. As a novice, it's really difficult for me |
您解决了吗,怎么解决的可以分享一下吗 |
时间太过久远我忘记了,不过确实成功运行了,好像是一层层把这个参数改了还是去掉,我再看看 |
非常感谢您!我刚开始接触这些,很多东西没有人问和交流,这对我来说实在困难 |
请加我的联系方式,qq97557509
…---Original---
From: ***@***.***>
Date: Thu, Mar 7, 2024 22:21 PM
To: ***@***.***>;
Cc: ***@***.***>;"State ***@***.***>;
Subject: Re: [keonlee9420/DiffGAN-TTS] ERROR (Issue #1)
你好!我在星期天下载了你的代码。我发现在“shallow”的 val 阶段,称为 diffusion.sampling(),但 sampling() 需要使用“cond”。由于self.cond是向前定义的,此时diffusion.sampleing()不会向前运行,所以“cond”会显示undefined,这是我的猜测,我不太了解,但是我在sampling()中添加了cond参数并更改了相关部分,它可以正常运行。也许我遇到了一些非常特殊的错误......无论如何,非常感谢您的工作!这个代码对我很有帮助,我想改进基线,帮助我毕业!
嗨,我在训练幼稚的时候也遇到了问题,你能告诉我如何修改代码吗?3Q
您解决了吗,怎么解决的可以分享一下吗
时间太过久远我忘记了,不过确实成功运行了,好像是一层层把这个参数改了还是去掉,我再看看
非常感谢您!我刚开始接触这些,很多东西没有人问和交流,这对我来说实在困难
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you modified the open/close state.Message ID: ***@***.***>
|
File "train.py", line 320, in 3.24s/it]
main(args, configs)
File "train.py", line 196, in main
figs, wav_reconstruction, wav_prediction, tag = synth_one_sample(
File "/data/workspace/liukaiyang/TTS/DiffGAN-TTS-main/utils/tools.py", line 227, in synth_one_sample
mels = [mel_pred[0, :mel_len].float().detach().transpose(0, 1) for mel_pred in diffusion.sampling()]
File "/root/anaconda3/envs/LKYBase/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 28, in decorate_context
return func(*args, **kwargs)
File "/data/workspace/liukaiyang/TTS/DiffGAN-TTS-main/model/diffusion.py", line 157, in sampling
b, *_, device = *self.cond.shape, self.cond.device
File "/root/anaconda3/envs/LKYBase/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1177, in getattr
raise AttributeError("'{}' object has no attribute '{}'".format(
AttributeError: 'GaussianDiffusion' object has no attribute 'cond'
Thank for your work!I seem to get a Error……
The text was updated successfully, but these errors were encountered: