Murray Chen

@Susan19900316 Did you solve this problem? I'm also stuck here :(

@boaz-codota Thanks for the instruction and I have found the status bar. For one thing that I used vscode remote server attached to my remote docker container. - It is...

Hi, @yonip23 @boaz-codota Thank you guys for the quick responses. Unfortunately, I have tried several times following the instructions strictly, but this problem still exists, and the log also recorded...

@yonip23 Yes, I still seeing the log. Still, the replaced version failed. I attach the whole log file of a new trial [tabnine.log](https://github.com/codota/TabNine/files/6900111/tabnine.log). Hope this will help. Thank you.

Again, it failed. And I find that no matter I do the steps as you suggested, or rebuild the remote envrionment, or even reinstall TabNine, it always creates the `engine...

@zhou-frank @feihuzhang Hello, I also found this problem when evaluating by code "evaluate.py" with the provided pre-trained model "sceneflow_epoch_10.pth". So, how to fix this problem in the evaluation?

@oliverxudd The case that all the layers are fixed is actually like the evaluation process, isn't it? As I found, it also costs 6s for a 576x960 image.

I have looked up #47 #53 , which discuss the similar problem of this code about its time-consuming. But I am still looking forward to fixing it. As the author...

mark

+1，不过高并发一般可以试试多实例部署吧