Skip to content
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

How to train the model? #1

Open
Guodong0712 opened this issue Dec 1, 2019 · 5 comments
Open

How to train the model? #1

Guodong0712 opened this issue Dec 1, 2019 · 5 comments

Comments

@Guodong0712
Copy link

Hi
Thank you in advance,
I am trying to implement the X-vector using your TDNN model, but when I try to train it and the loss can not down and the rate of right is so low.
Thank you, could you give me some suggestions?
my input: 24-dim MFCC

                                                                                                                                    Yours sincerely
                                                                                                                                             Ma
@qinxiaoyi
Copy link

Hi
Thank you in advance,
I am trying to implement the X-vector using your TDNN model, but when I try to train it and the loss can not down and the rate of right is so low.
Thank you, could you give me some suggestions?
my input: 24-dim MFCC

                                                                                                                                    Yours sincerely
                                                                                                                                             Ma

i have the same question, when i train the model, the loss always nan, than i exchange the relu and batchnorm, but the eer is very high

@Guodong0712
Copy link
Author

Thanks for your reply, I am a freshman and have not to compute the eer because the final exam.

@Guodong0712
Copy link
Author

Hi
Thank you in advance,
I am trying to implement the X-vector using your TDNN model, but when I try to train it and the loss can not down and the rate of right is so low.
Thank you, could you give me some suggestions?
my input: 24-dim MFCC

                                                                                                                                    Yours sincerely
                                                                                                                                             Ma

i have the same question, when i train the model, the loss always nan, than i exchange the relu and batchnorm, but the eer is very high

Hi,thank you first, I make the score with cosine, and about 0.998 between the same speaker utt and
0.993 between the different, so I predict that eer will be high, can you give me some suggestions?

Ma

@HappyGallery
Copy link

Hi
Thank you in advance,
I am trying to implement the X-vector using your TDNN model, but when I try to train it and the loss can not down and the rate of right is so low.
Thank you, could you give me some suggestions?
my input: 24-dim MFCC

                                                                                                                                    Yours sincerely
                                                                                                                                             Ma

i have the same question, when i train the model, the loss always nan, than i exchange the relu and batchnorm, but the eer is very high

Hi ,if you have solved the question ,loss always nan .I have meet the problem too , but i have no method to solve . if you know how please teach me . Thanks!

@HappyGallery
Copy link

@qinxiaoyi

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants