nurture.ai is organising study chapters in various cities across the world and it is free to attend for anyone. They are covering fast.ai lessons as study material and it will be great for fast.ai students to catch up and discuss and continue their deep learning studies.
Thank you, I register to open one in my City Mons, Belgium. Hopefully it will be center enough to attract other cities as well, Brussels, Lille etc
I too have registered for it. It’s a great thing to do.
I try to understand what this initiative is and what it brings to joint it (or even create it in my city).
- Could you send feedbacks ? For example, as coordinator, do you get special access to material ? to documents ? to events ?
- The participants : do they get as well special resources ?
- Everything is free ? What is nurture.ai (association, company, etc.) ?
I know there is a forum online in this site but… there is one here as well … with all the students of the precedent courses taught by @jeremy and @rachel. So, I would like to understand better why to use https://nurture.ai/ai-saturdays and not http://forums.fast.ai/ ?
Thanks a lot !
I’m one of the volunteer mentors in the KL chapter, so here’s my 2 cents.
For communication, we’re using Slack and the our own discourse forum (https://ai6forums.nurture.ai/). Slack is used as more of a broadcast channel for adhoc communications, while the forum contains discussions specific to each chapter.
As far as KL is concerned, from the coordinators pov, we’re rather flexible in terms of how each Saturday programme should be handled, since it very much depends on the feedback that we receive.
Fast.ai viewing and discussion aside, we also discuss relevant research papers that we or the audience seem to be interested in understanding.
Participants are expected to come up with their own resources (in terms GPU availability etc). As far as the Saturday meetups are concerned, things have been free so far.
Nurture.ai itself is a startup that is still in beta phase. It’s currently hosting a lot of research papers and their implementations, along with a platform for discussions. Not too sure on their long terms plans though.
To add to @hafidz’s answer:
We had tried helping with setting up for Fast AI-in our discussions and for Course related discussions, we encourage using the Fast AI forums, since making another mini community would be confusing.
I can’t say about the Meetups, since I haven’t attended one (I’m interning). I try to help around with discussions on the Forums and in slack.
There are no special resources provided as such. We have a community with more experienced geeks-helping new geeks.
Yes, Everything is free.
For GPU Resources, we were trying GCP since they give a 300$ jumpstarter.
This is my understanding from your answers :
online working place : use forums.fast.ai as working place (do not reinvent the wheel in another place).
communication canal : use another online service as communication canal (Slack but I guess that WhatsApp, a Facebook or Google group, or even the Meetup site could be used as well ?).
computer and GPU : each participant has to come with its own computer and a GPU cloud setup (could be the one of Google Cloud Platform (GCP) with a credit of 300$, but could be others like written in the post of @tenoke “Cost effectiveness of the different Cloud/Server renting options” and in the post of @hetelek “Jupyter Notebooks in the cloud w/ GPUs, starting at $0.35/hour!” ?).
courses’ program : stay flexible about the program of each new meeting. Setup it from feedback of participants.
cost : there is no cost for participants.
About AI Saturdays, it is therefore an initiative from a startup (nurture.ai). It provides a forum and a proposition of program, free of cost. As you said, it is better use the forums.fast.ai site for posting questions to the fast.ai community. So, I don’t see the point for using this site. About the AI Saturdays program (3 sessions each saturday), I’m wondering if it is not too much and if it does not even mix up the methodology setup by @jeremy and @rachel ? More, the objectives of each AI Saturdays session are different I think. The session 1 in the morning, using the fast.ai material, is dedicated to people who want to learn by doing. The sessions of the afternoon are more conferences, no ?
Please, I have 3 more questions :
- Do you watch the fast.ai videos during the session or ask participants for watching them before ?
- How many participants are in each session concerning the fast.ai course ?
- How many advanced DL teachers do you have and how long does it take to get ready a session ?
Thanks very much
Sorry for the late reply - it was quite a long holiday here in Malaysia.
I’ll try to answer your question to the best of my knowledge, based on the events happening in the KL chapter. Keep in mind though that I’m not affiliated with the company.
The afternoon session are more conferences?
[A] Not quite. It’s more like a group discussion where they compare each other’s understanding of the paper. The moderators will be reading the papers beforehand so that the session will be more guided.
Do you watch the fast.ai videos during the session or ask participants for watching them before ?
[A] The mentors are supposed to go through the videos prior to the session.
How many participants are in each session concerning the fast.ai course ?
[A] We’ve had a big turnout during the first few weeks (>50). Numbers are getting lesser though as we progress through the lessons.
How many advanced DL teachers do you have and how long does it take to get ready a session ?
[A] I wouldn’t call any of us “advanced DL teachers” by any measure. We’d usually spend a few days prior to get ready for the session. What usually ends up happening during the session is that if none us could come up with a convincing explanation to a problem posed by the audience, we’d usually take those back and answer them at a later period, either via the forum or slack.
a message just to tell you that we finally started our Deep Learning study group in Brasilia
We started with the Fastai part 1.
Thanks again for your feedback more than a month ago.
All the best for you too.
Great to hear that Pierre. Happy learning!!