The kernel for ch01.ipynb appears to have died. It will restart automatically

Part 1 2022
1

Hi, I hope you will be able to help me with this issue.

I have recently bought RTX 2060 GPU and installed Deep Learning Libraries in the below sequence:

  1. Mambaforge using WSL
  2. PyTorch ==> conda install pytorch torchvision torchaudio pytorch-cuda=11.6 -c pytorch -c nvidia
  3. FastAi ==> conda install -c fastchan fastai

And now if I try to run FastAI Vision exercise from the first chapter I’m getting this error “The kernel for ch01.ipynb appears to have died. It will restart automatically”

What I have tried:

  1. Deleted my entire installation and then reinstalled the downgraded PyTorch using ==> conda install pytorch==1.12.1 torchvision==0.13.1 torchaudio==0.12.1 cudatoolkit=11.6 -c pytorch -c conda-forge

  2. Updated Numpy

  3. Added this piece of code import os os.environ["KMP_DUPLICATE_LIB_OK"]="TRUE"

But none of them have worked so far.

Note: I’m facing this issue only when I’m using Vision class…I could run Text learner without any issues (by using GPU).

2

I think I found a solution to this problem. The route cause is probably something to do with vision learner compatibility with the latest PyTorch.

So we need to use the below command to install a pytorch compatible version before installing FastAI

conda install pytorch==1.11.0 torchvision==0.12.0 torchaudio==0.11.0 cudatoolkit=11.3 -c pytorch

Here is a link( Previous PyTorch Versions | PyTorch) Pytorch website.

A similar thread The kernel appears to have died. It will restart automatically. v2 - Part 1 2022 - fast.ai Course Forums

1 Like