InvalidArgumentError (see above for traceback)
Opened this issue · 2 comments
CJJ2923 commented
In the example code: Speaks=13, but in the data, we are processing: Speaks=30, how should we modify the code: dialogue 35 (text)+max turn of dialogue 35 (financial expression)+speaker number 13+movement categories 7. I hope to get your reply, thank you.
XL2248 commented
Hi，
we mainly maintain 12 speakers who speak most utterances and use a special token to indicate all unknown speakers (a total of 13 speakers).
Hope it helps.
CJJ2923 commented
Thank you very much!
…------------------ 原始邮件 ------------------
发件人: ***@***.***>;
发送时间: 2022年10月13日(星期四) 下午2:19
收件人: ***@***.***>;
抄送: ***@***.***>; ***@***.***>;
主题: Re: [XL2248/HGNN] InvalidArgumentError (see above for traceback) (Issue #9)
Hi，
we mainly maintain 12 speakers who speak most utterances and use a special token to indicate all unknown speakers (a total of 13 speakers).
Hope it helps.
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you authored the thread.Message ID: ***@***.***>