Hello, this is my first post on the forums!
Ok so, I installed fastai and fastbook via Anaconda Prompt. And I am using the localhost to try and run the first notebook with Jupyter Notebook (on my own PC, Windows 10). I followed the Windows 10 installation steps from Jeremy (2018), but I do not understand what the problems are. When I use Google Colab, it doesn’t have this particular error, but it seems to take at least an hour to finish training, and, of course, it will timeout because I am AFK by then, and it doesn’t show the entire content of the notebooks, so I do not ever complete them. So, I am trying to run this on my computer with Anaconda/Jupyter instead of Google Colab. I realize I might need to do this on Linux distro to utilize GPU, but I also wasn’t sure if that was really still the case. Let me know if you guys have any tips or tricks for me. THANKS
Colab should not take a hour during training. Make sure you have gpu enabled by going to runtime - change runtime type - gpu.
Try adding this before you install fastbook
pip install --no-cache-dir torch==1.6.0+cu101 torchvision==0.7.0+cu101 -f https://download.pytorch.org/whl/torch_stable.html
THANKS. Yeah, I’m seeing that was even mentioned on fast.ai, I just totally assumed it was already running for some reason. I was using some kind of notebook at first that might’ve had it run by default.
Here’s the output using your suggestion on my end:
Hello, did you find a solution to this issue? I am having the same error.
This should work as is now. Just tested it and the gist can be found here
from fastai.test_utils import show_install
=== Software ===
python : 3.7.10
fastai : 2.2.7
fastcore : 1.3.19
fastprogress : 0.2.7
torch : 1.7.1
nvidia driver : 460.32
torch cuda : 10.2 / is available
torch cudnn : 7605 / is enabled
=== Hardware ===
nvidia gpus : 1
torch devices : 1
- gpu0 : Tesla K80
=== Environment ===
platform : Linux-4.19.112+-x86_64-with-Ubuntu-18.04-bionic
distro : #1 SMP Thu Jul 23 08:00:38 PDT 2020
conda env : Unknown
python : /usr/bin/python3