About the Part 1 (2020) category

More specifically, I believe it’s the one with the lowest wordnet synset id, so often appears first.


Wow. It is great to see all these familiar faces. It has truly been a long time.

1 Like

Great. I recently started to watch walk fastai v2 code walk-thru.The design concept and flexibility of v2 is cool. v2 software engineering is very attractive to me. Thanks :star_struck:

1 Like

The fish looks like a banjo banjo fish based on my “google reasearch” :wink:

Me too just watching :slight_smile:

Wonder are there any suggestions for prep to be ready for new course ?

1 Like

I gonna go through this:
And thinking about projects I want to do.

re-read @init_27 How not to do Fast.ai (or any ML MOOC)

from his interview with Christine Mcleavey:

one of my favorite pieces of advice was from FastAI’s Jeremy Howard. He told me to focus on making one really great project that showed off my skills. He says that so many people have many interesting but half-finished projects on github, and that it’s much better to have a single one that is really polished.


I was lucky to interview her again, where we discussed this in a lot of depth:


This new paper might be good prep too :wink: https://arxiv.org/abs/2002.04688! Great work Jeremy and Sylvain! And here too.


I have a quick question. Is it allowed to watch live/recordings in the online meetup? :slightly_smiling_face:


Yes, We’re allowed to do it, please the last paragraph check here: https://forums.fast.ai/t/official-part-1-2020-updates-and-resources-thread/63376/3

1 Like

Thank you @init_27 . I just got hesitant cause it was mostly referring to offline meetings. So can I share the link to the videos or should I open it myself and share with others as a screen share?

Given the paragraph mentions the following, I’d assume that it’d be preferred to not share the links after 24 hours (at which point, Jeremy usually uploads an edited version), but instead watch it live/ almost live with your friends and/or colleagues.

(no more than 24 hours after the live stream, please, so you stay up to date with the discussions)

I might haven’t understood your question, I think it’s okay to watch it live with friends or play it back at a suitable time for a group-but maybe broadcast sharing the link with folks won’t be the best approach.

1 Like

@shahnoza thanks for the great question. I did indeed consciously not include online meetups in the message - basically I’m wanting to make sure that anyone who is organizing a study group is doing it with a level of care a consideration commensurate with the care and consideration we’re putting into the course, such that the overall experience of all in the fast.ai Live group will benefit from their involvement. It’s certainly possible to do that with an online study group too - for instance see TWIMLAI: https://twimlai.com/twiml-x-fast-ai/

Please don’t share the video links at all. But if you wish to re-stream to a study group that you are carefully managing, and where people have to sign up for access (so not just anyone can connect any time), then you may do so.


Thank you @jeremy for the detailed reply, sure, I will be re-streaming without sharing links and with sign up process. Thanks for the opportunity, I am sure the group members will be excited :slightly_smiling_face:

1 Like

I love that the fastai machine learning course is to be incorporated in this one!


I think that we have to be disciplined in using the advanced section for in depth keeping the normal one open to everyone for any kind of questions

Thank you so much, Jeremy! Amsterdam study group is very thankful!

This is good news, I’ve been planning to revisit the ML course for a long time now. Sounds like a step towards presenting FastAI as a general-purpose tunable-computing framework.

1 Like

I’ve been away from the forums and fastai for a bit now. I got back to it because of your fastaiv2 paper and blog post. Thanks for the invite, I’ll try and keep active! :slight_smile: