Lesson 3 - Official Topic

ilovescience · April 1, 2020, 2:11am

Actually I think if you use the experimental instructions here you can get it to work:

(see here)

mario_carrillo · April 1, 2020, 2:11am

So viola runs locally it seems?

sgugger · April 1, 2020, 2:11am

It’s in the fastai2 repo.

gamino · April 1, 2020, 2:12am

Can you host voila it in github?

vijaysai · April 1, 2020, 2:12am

Streamlit is also really cool in getting things done fast.

0tist · April 1, 2020, 2:12am

how can we make voila work on the paperspace?
As in, it wont really work by changing the URL of the paperspace hosted notebook.

sgugger · April 1, 2020, 2:13am

We will see the deployment in a minute.

maya · April 1, 2020, 2:13am

Anything about security issues using iPyWidgets / Voila on production ?

geoffpidcock · April 1, 2020, 2:13am

I have, and can testify that it’s easy and quick. This article offers a walk through for a simple dashboard hosted using heroku. https://towardsdatascience.com/quickly-build-and-deploy-an-application-with-streamlit-988ca08c7e83

jcatanza · April 1, 2020, 2:13am

Sure you could, but how many images do you think it would take to capture “not-bearness”? Interesting experiment to try. My bet would be that it would be a deal-breakingly large number…

ilovescience · April 1, 2020, 2:13am

You will have to set up Colab differently

Albertotono · April 1, 2020, 2:14am

This is awesome, I also didn’t know the connection with vue.js and jupyter

JPKab · April 1, 2020, 2:14am

Definitely. Streamlit, so far, is pretty solid.

username_not_found · April 1, 2020, 2:15am

If I am having difficulty running one of the .nbs, should I start a new post or post to an existing thread? I previously could run fastai last year on this machine, but it says now it doesn’t have enough memory on the GPU for 01_intro this year. Basically just firefox and the docker are running. nvidia-smi shows basically nothing else running.

Running in LXD docker natively with GPU passthrough to a GTX 1080 with ubuntu 18 lts host.

nchukaobah · April 1, 2020, 2:15am

Thanks

ilovescience · April 1, 2020, 2:15am

Yep I am not sure either…

sgugger · April 1, 2020, 2:16am

Use a new thread please. You should adjust your batch size as necessary if you have memory issues.

JPKab · April 1, 2020, 2:16am

Additional note on CPU vs. GPU for deployment:

You might want to build and deploy a desktop application in an enterprise environment. Basically a big use case believe it or not, especially for my customers.

Dilbert and friends don’t have GPUs on their PCs.

ilovescience · April 1, 2020, 2:17am

If you do need to deploy directly on a phone, you can use PyTorch Mobile as well.

maya · April 1, 2020, 2:18am

Yes, this is not easy and it’s the drawback with Classification in DL today. Hence my question.