Error in Writing model during runtime saving checkpoint

Support
1

Hi all, I am getting this error during runtime.

Saving checkpoint to ‘Abstractdemo-Abs-model_epoch5_34.24.t7’…
/root/torch/install/bin/luajit: /root/torch/install/share/lua/5.1/torch/File.lua:210: write error: wrote 10542800 blocks instead of 41668004 at /tmp/luarocks_torch-scm-1-4168/torch7/lib/TH/THDiskFile.c:356

I am running the torch version model with 4 encoders and 4 decoders. I dont think GPU memory or system memory is an issue as I am using GTX1080Ti (11 GB) and my system memory is 32 GB. Please help me

2

Hi @vishnmt - are you sure it is not simply a diskspace issue?

3

@jean.senellart yes there is no problem with diskspace

4

Could you post the output of the command df -h?