Aonnghus

Hi, Might be a bit too late for you, but I found the solution to this problem with c++ errors. Do what @ftorregrossa said, but when it comes to `cython...

Hi, I'm currently using TF 1.8.0 and Keras 2.0.8 and everything's working well. Those versions are a bit old, but due to API changes, the code won't work with newer...

Hello, Since January you surely found a solution to your problem, but if that's not the case and you're still wondering, or at least for other people that might come...

Hello @VkoHov. The images I'm tracking are directly created with `aim.Image` from torch tensors, with options `format="png"` and `optimize=True`. Their size is 256x256. I cannot send you the real data...

Hello @alberttorosyan Yes, the same thing is happening with versions newer than 3.12.0. Actually, I noticed the issue on the latest version to date (3.13.1), and I downgraded Aim until...

I tried on another machine, launching time was very short. This part might be hardware-related (but it would be weird, I have 64 Gb of RAM and a Xeon CPU)....

A bit more details about my setup: I am connecting to a remote machine through SSH, where aim is installed in a virtual env. Home directory is mounted from another...

Hello @mihran113. That was it! Forcing the use of FileLock solved my performance issues. Thank you so much!

Yes such warnings appeared in stdout

`mount` command gives the following output concerning the home directory: `: on /home type nfs (rw,nosuid,nodev,relatime,vers=3,rsize=65536,wsize=65536,namlen=255,hard,proto=tcp,timeo=600,retrans=2,sec=sys,local_lock=none,addr=)`