For people watching now, you don’t need to be set up or have the notebooks running. Watch the material now and set up later.
22 Likes
yes, p4000 will work.
You’ll have to build pytorch 1.0 yourself. Although people have been successful it is a difficult process.
1 Like
If you don’t have admin privilege, it’s because you didn’t login using jupyter@… on your first login. I’m afraid you’ll need a new instance.
1 Like
Yes it can!
ah, right! thanks sgugger!
are these notebooks discussed already available in git hub FAI ?
1 Like
you can, but you will have to build pytorch from source. Search the forums for details.
1 Like
The GitHub repo link is in the FAQ/Welcome post now. But:
git clone https://github.com/fastai/course-v3
Are the slides available online?
3 Likes
or you can use one of the cloud GPU options that Jeremy outlined.
I have 1.0.5, how can I update
The live stream is stopped. Like here if you have the same experience.
That’s not true. See: fastai/README.md at master · fastai/fastai · GitHub
Please save installation/setup questions for later. Let’s keep this thread focused on questions related to Jeremy’s lecture.
10 Likes
Hi, the EC2 link is not working. Anyone else have the same problem.
I keep seeing this idiom inside the notebook
path/‘filename.csv’
What exactly is this? Some type of string interpolation?
try reloading the youtube page. Or switch to a different browser. It stopped working for me on firefox, switched to chrome - it’s fine there.
i install it on local machine. just ran through the lesson 1 note book.
AttributeError Traceback (most recent call last)
in
----> 1 learn = ConvLearner(data, models.resnet34, metrics=error_rate)
AttributeError: module ‘fastai.vision.models’ has no attribute ‘resnet34’
got this error. any help
1 Like