Hi, anyone getting this error in lesson 1-pets notebook
from pathlib import Path
path = Path(’/storage/pradeep/data/oxford_iiit_pet/’)
path_anno = path/‘annotations’
path_img = path/‘images’
fnames = get_image_files(path_img)
np.random.seed(2)
pat = r’/([^/]+)_\d+.jpg$
data = ImageDataBunch.from_name_re(path_img, fnames, pat, ds_tfms=get_transforms(), size=224)
data.normalize(imagenet_stats)
Error:
ValueError: padding_mode needs to be ‘zeros’ or ‘border’, but got reflection
Checkout your pytorch version. This link might be helpful - Errors and corrections in 005_dogs_cats.ipynb
Yes, updating the pytorch version solved the issue.
Has anyone of you built a deep learning box. I want to build one. Please share your experience.
Below are the links for the fastai blog and forum discussion related to building own box
https://www.fast.ai/2017/11/16/what-you-need/
Hi Guys, it’s been a great journey so far in Part1 of the course and now the course is ending today. We should share our learnings, thoughts.
Learning can be from this course or any other place.
Thoughts can be related to course or impact AI/ML done on you or what will be the future for you and AI/ML or anything under the sun.
We should also discuss a better way of organising the meet so all of us stay focused.
Let’s meet up this Sunday 15 dec before all of us take a leave for Xmas & new year.
Hi Guys,
I see that there is no activity post Dec. Would you guys be willing to help locally in Pune?
Looking for some help to get through initial teething trouble. And also to make friends I can exchange knowledge locally with a good cup of coffee… that never fails 
or any other drink of your choice!
Sounds great! I am interested.
Hoping its the coffee which turned you around and not the offer for “drink of your choice”
No, Its Pune and exchange knowledge that did the trick but yes the other drink of choice sounds cool 
1 Like
have sent you a DM
Hello, I’ve just launched a discussion thread on fastai study groups to gather feedbacks from the organizers and participants in order to list the best practices and also to avoid some gaps.
More information in this post. Thank you if you have a few minutes to participate in the discussion
you can use GCP as it provide 1 GPU quota free. and if you have $300 credits it will charge your from it only. clouderizer.com having a video on it, you can check it in there youtube channel. I am using it for my fast.ai and till now it’s good.