dreamPoet

Thx. BYW, I find during the training there is no log file created. Are there any codes I can uncomment directly to produce log files with tensorboardX?

> The performance on LIP and Pascal Context (COCO stuff and ADE) is stable. But on Cityscapes, the performance is not very stable. Shall I use the same environment setting...

BTW, I am using python version

Hi @angeladai ， is there anyway to run reader.py for all .sens files scattered in different scene directories? I get the error `Is a directory: '../../ScanNet/scans/' `when I use `--filename...

> [#34 (comment)](https://github.com/nv-tlabs/GSCNN/issues/34#issuecomment-564370469) > You can try it thx it's a good idea...but how about the parameters sync. when training on multi GPU? Will it cause problem?

Have u done the training process successfully now? In fact I meet problem when I try evaluating it...the codes stuck when import encoding and did not return any error message...

> > Have u done the training process successfully now? In fact I meet problem when I try evaluating it...the codes stuck when import encoding and did not return any...

I clone the whole repo but found the same problem...