I think that running the code in 01_intro.ipynb may be using up memory and want to clean up my local hard drive to get back some space, but I can’t figure out where the downloaded image files are saved in Running Your First Notebook. When I run:
print(path)
import os
print(os.path.abspath("HOME"))
I don’t see a storage directory. Where do I look for it?
I found a reply on Saved Location for learn.save by marcmuc on Oct 18 that indicates pretty much what I found, but there are only two directories under Notebooks: bears and images and those I created in 02_production.ipynb so that I could easily find them to see what the algorithm was using.
I think that I’ll just create my own path like I did with the bears lesson. I would still like to know where the notebook code saved the pet files though.
When you open the notebook (from Paperspace for instance) don’t you see 3 directories (datasets, fastbook, and storage) listed?
Then when you enter the fastbook directory that’s where you find the lessons etc., that is also where most stuff (images, bears, etc.) get created. If not (unlikely), maybe you can go check in the storage directory.
Alright, I feel really stupid now. I thought that /HOME was a system variable in Python, but it is just a Windows directory name!
C:\HOME\.fastai\data\oxford-iiit-pet\images
It took Windows quite a while to find oxford-iiit-pet, but it did. The cat pictures are there. FYI, the /oxford-iiit-pet and its subdirectories hold 783 MB of data, in case anyone is wondering how much your local HD will have to provide if you want to run this locally.
