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

can't get performance result on kth #13

Closed
xiaomingdujin opened this issue Jun 14, 2020 · 2 comments
Closed

can't get performance result on kth #13

xiaomingdujin opened this issue Jun 14, 2020 · 2 comments

Comments

@xiaomingdujin
Copy link

Hello, thank you for the paper and releasing the code. I run the test on the kth action dataset uising the downloaded the pre-trained weights, but I can't get the results of the paper which report on the kth task.
Maybe it's my parameters or I am not running it the right way?

pd11
pd12
pd13
pd14
pd15

@wyb15
Copy link
Collaborator

wyb15 commented Jul 6, 2020

Hi Jinming,

You are right. It should return "new_global_memory" in Line 217 in "e3d_lstm/src/layers/rnn_cell.py", instead of "global_memory". It is the reason why you and some others cannot get the performance reported in the paper.

We noticed that this bug only appeared in the code repo, which was revised and reproduced in accordance with company policy after the paper was accepted. This did not exist in our original experiments, so the results reported in the paper are good.

We are fixing the code and will re-upload the pre-trained models. We apologize for the inconvenience and thank you for your patience.

Best,
Yunbo

@wyb15 wyb15 closed this as completed Jul 6, 2020
@roadjiang
Copy link
Contributor

Duplicated issue. Please refer to the following one for future discussions:
#1

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