Welcome to conv-emotion Discussions! #48
Replies: 3 comments 1 reply
-
|
In this dilogueGCN model on IEMOCAP using your model, I am getting 55% F1 score. epoch: 60, train_loss: 0.588, train_acc: 80.64, train_fscore: 80.52, In your paper, you are claiming 65.25 |
Beta Was this translation helpful? Give feedback.
-
|
Can you share the full command that you are using for training? You can also use the alternate implementation in the |
Beta Was this translation helpful? Give feedback.
-
|
Sir
Downloaded the code, along with the IEMOCAP features provided.
Just run the train.py file.
and after 60 Epochs :
epoch: 60, train_loss: 0.5595, train_acc: 82.86, train_fscore: 82.76,
*valid_loss: nan, valid_acc: nan, valid_fscore: nan, *
test_loss: 1.0062, test_acc: 61.49, test_fscore: 60.7,
time: 2.97 sec
Test performance..
F-Score: 60.79
Why valid data is nan, is it because nothing is split into this data
partition?
changing the seed giving me this accuracy and its changing somewhat
everytime.
Is there anything i need to tune for your accuracy sir?
Regards
…On Sat, Jul 30, 2022 at 11:32 AM Deepanway ***@***.***> wrote:
Can you share the full command that you are using for training?
You can also use the alternate implementation in the DialogueGCN-mianzhang
directory.
—
Reply to this email directly, view it on GitHub
<#48 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/A2ERTZMTEPH2NFNUQ7E2RHLVWTAORANCNFSM5W6E36OQ>
.
You are receiving this because you commented.Message ID:
***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
|
Hi Sir, What's the difference between the two versions of implementation of DialogueGCN? Thanks. |
Beta Was this translation helpful? Give feedback.
-
👋 Welcome!
We’re using Discussions as a place to connect with other members of our community. We hope that you:
build together 💪.
To get started, comment below with an introduction of yourself and tell us about what you do with this community.
Beta Was this translation helpful? Give feedback.
All reactions