Restarting random forest notebook results in Import error with relative paths

I had restarted the lesson1-rf notebook when I got this ModuleNotFoundError while executing this cell

I did not see any path information related to fastai when I printed sys.path. Is this ok?

My directory structure:

~
└── workspace 
           └── fastai 

My conjecture was that this was a sys.path issue, so I went ahead and inserted the fastai location in the previous cell. So, here my <path-to-fastai-repo> is ~/workspace.

import sys
print(sys.path)
sys.path.insert(0, `<path-to-fastai-repo>`)
print(sys.path)

But the issue still seems to persist. Not sure what I am missing here.

Maybe the isoweek module is not installed?

Just do ‘pip install isoweek’ in terminal. Do that for the other packages if the error pops up with some other package name.

Thanks @anandsaha and @ar_ai : Fair points , although the notebook ran just fine the first time. It was only when I restarted it, this issue has cropped up.

I went ahead and did the following, installed all packages from requirements.txt, pandas-summary and torch and tested it out as well:

pip3 install -qr requirements.txt
pip3 install pandas-summary
git clone https://github.com/torch/distro.git ~/torch --recursive
cd ~/torch; bash install-deps;
./install.sh
source ~/.zshrc
$th

  ______             __   |  Torch7
 /_  __/__  ________/ /   |  Scientific computing for Lua.
  / / / _ \/ __/ __/ _ \  |  Type ? for help
 /_/  \___/_/  \__/_//_/  |  https://github.com/torch
                          |  http://torch.ch

th>

I am starting to sense that I should not be doing this exercise in my local computer.

You installed Torch for Lua. You need to install pytorch. Clone from this link https://github.com/pytorch/pytorch#from-source. You can find binaries for your OS from the pytorch homepage.

1 Like

Best to use anaconda, and ‘conda env update’