December 09, 2019

Thoughts on the 1st lecture session

Follow-up to Installing your own TensorFlow/Kears + FastAI on Ubuntu from Rudo's blog

Dear all,

[1] today in the lecture, we saw that it was not entirely simple to install your own ML machine with Tensorflow/Keras and FastAI. I have now made a blog entry on how to do this. You can find it here.

[2] In addition, when running the FastAI code, we saw that it might take a long time even for just the pets categorization. I just ran the pets notebook (please download a new version) and have made the pets-stage-1 and pets-stage-1-50 trained models available on the download area. You need to move these models into where FastAI keeps them which in my set-up is

/home/phsht/.fastai/data/oxford-iiit-pet/images/models

Just copy the twp .pth files there and you should be able to run the pets code after the

learn.load('pets-stage-1')

fastai commands in the notebook.

[3] When you "play" with the notebook called "FastAI-lesson2-download.ipynb", note that the file data.zip contains the cats/teddys/wolves that I managed to download and use for the exercise.

[4] Some of you argued that your own machines may not be able to do meaningful ML as you do not have a GPU. Have a look at what the FastAI guys have collected in terms of possible free GPU instances, i.e. Using a GPU. I have not yet used any of these services, but would be keen to learn from your experiences!

OK, keep checking this blog for up-to-date information before we meet again next week.


- 2 comments by 1 or more people Not publicly viewable

  1. Juan

    Regarding possible alternatives to benefit form a GPU, I have personally used Kaggle, by Google, its a free service and you get access to 30 hours of GPU run time per week, which I reckon should be enough. I managed to run the whole lesson1-pets notebook in under 40 minutes there with no need to install anything before hand, just loading the notebook and clicking play.

    09 Dec 2019, 18:37

  2. Rudo Roemer

    That’s cool. Perhaps you want to show people during class next week how to run on Kaggle?

    09 Dec 2019, 19:30


Add a comment

You are not allowed to comment on this entry as it has restricted commenting permissions.

December 2019

Mo Tu We Th Fr Sa Su
Nov |  Today  | Jan
                  1
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31               

Search this blog

Tags

Galleries

Most recent comments

  • Please note that this link to download is now obsolete. Attendees for the 2021 MPAGS/CYU course shou… by Rudo Roemer on this entry
  • To load Tensorflow and PyTorch (and Keras) modules on ORAC may you may have to module load a few mor… by Daniel Paget on this entry
  • That's cool. Perhaps you want to show people during class next week how to run on Kaggle? by Rudo Roemer on this entry
  • Regarding possible alternatives to benefit form a GPU, I have personally used Kaggle, by Google, its… by Juan on this entry
  • Hello to you all, I was just trying to make the programs run in my VirtualBox with Ubuntu 18.04 inst… by Malaquias Correa on this entry

Blog archive

Loading…
Not signed in
Sign in

Powered by BlogBuilder
© MMXXI