Pospielov Serhii

Hi there, I'm a CompreFace repository contributor and looks like I know how to reproduce it. So real Ivan Kurnosau has such config for git on a local machine: user.email:...

Hi, it looks like you use configuration from the master branch, instead of release. Please follow the instructions from here: https://github.com/exadel-inc/CompreFace/releases

Could you clarify, when you do recognition, does it work? Because if you start CompreFace build for GPU on a machine without GPU, it won't work.

It doesn't support it for now. You can use Python code snippets from Postman as a workaround: https://documenter.getpostman.com/view/17578263/UUxzAnde#28c03009-c546-4c8e-83bb-b2cd55dd68cd

It looks like you have 4 uwsgi processes. Each process has to load the model to RAM. And processes don't share the memory. So yes, it's expected. The question is...

> ../src/nnvm/legacy_json_util.cc:203: Warning: loading symbol saved by MXNet version 20000 with lower version of MXNet v10900. May cause undefined behavior. Please update MXNet if you encounter any issue This is...

Could you run `docker-compose logs` and share the result?

It looks like CompreFace didn't run all database scripts because it was stopped. Is it a new CompreFace installation? If yes, you need to clean the DB: 1. Stop Compreface:...

From the new log, it looks like it didn't start yet. You need to wait 30-60 seconds until you open CompreFace Other option - to run `docker-compose logs -f` and...

in log3 and log4 there is the same error as in log1 So please, delete the volume again and run CompreFace, but wait 60 seconds.