PWCNet-tf2 icon indicating copy to clipboard operation
PWCNet-tf2 copied to clipboard

Unable to run inference.py

Open mpslxz opened this issue 5 years ago • 7 comments

I get the following error when running inference.py:

tensorflow/core/framework/op_kernel.cc:1763] OP_REQUIRES failed at save_restore_v2_ops.cc:205 : Out of range: Read less bytes than requested Seems to be something wrong with the checkpoint files?

mpslxz avatar Jan 13 '21 18:01 mpslxz

Hi,

I am seeing the same issue. Were you able to solve it. If yes, please provide steps. Thank you.

Ajaz-Ahmad avatar Mar 19 '21 11:03 Ajaz-Ahmad

No, didn’t work for me.

On Mar 19, 2021, at 4:10 AM, Ajaz Ahmad @.***> wrote:

 Hi,

I am seeing the same issue. Were you able to solve it. If yes, please provide steps. Thank you.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

mpslxz avatar Mar 21 '21 04:03 mpslxz

For me the issues is solved by downloading the weights separately from the checkpoints directory. When I downloaded the checkpoints with downloading the whole project, the checkpoint were not downloaded correctly. Maybe because the checkpoints are stored in GitLFS.

smehdia avatar Nov 03 '21 23:11 smehdia

Hi @smehdia, could you teach me how to download the weights separately from the checkpoints directory? I got the message that says the repository is over its data quota, Account for LFS bandwidth should purchase more data packs to restore access.

Thanks in advance!

wenxiaojie avatar Apr 08 '22 02:04 wenxiaojie

@wenxiaojie I have re-uploaded the weights in here: PWCNet-tf2-weights

smehdia avatar Apr 08 '22 05:04 smehdia

It works now. Thank you very much! @smehdia

wenxiaojie avatar Apr 08 '22 10:04 wenxiaojie

Do u know if there is a way to run the inference in real time on a webcam ?

ilinii avatar May 02 '22 10:05 ilinii