2022 Study Group Anyone?

Hi @bbrown may I join, please?
Thanks :blush:

Hey! I would love to join too!
Btw, why isn’t there a new forum for 2022 on this website?

Also on EST here, would love to join the next call @bbrown !
Cheers

@Danrohn I believe the topic groups (Part 1 2020, Part 1 2019, etc.) are created each time there’s a new course. I think the most recent course is Part 1 2020. The “2022” in this study group thread is just to say the study group is happening this year, but we’re going through the Part 1 2020 course. (That’s the course at https://course.fast.ai)

1 Like

Hello there! I also would love to join next call. Kindly share with me time. I am also on EST.
Best regards,
Andrea

Hi Guys! please add me to the study group also

HI @bbrown I’d be really interested to join this study group please add to it! looking forward to learning and contributing

@LLisa Hi! which Udacity course is it? and is it any good?

I just realized the Discord only allows 10 users to video chat at a time and in an event that everyone shows up, we would have more than 10 users.

So let’s just have the next study group via Google Meet. See info below:

2022 Fastai Study Group Session #2
Tuesday, January 25 · 8:00 – 9:00pm Eastern Standard Time Zone
Google Meet joining info
Video call link: https://meet.google.com/uvh-sccq-wsz

We will also go over the below agenda.

  1. Introduction of new members
  2. Showcase Image Classifier
    a. Any challenges you faced that you needed to create a custom code not included in the lecture? If so, please explain.
    b. What concepts are still unclear to you?
    c. Any Eureka moments?
  3. Share resources
  4. Identify homework for the next study group session.

See you then!

2 Likes

Hi everyone,

This week @willsa14 and I met in the study group where we had some really meaningful conversations around the type of projects we hope to work on with the tools we learn in this course. We also realized that the dialog we have in the study group is where the real value of the weekly meet is.

For those who were not able to attend here are the notes that we covered.

  • Showcased the models we were working on.

Resources

  • Pixplot – PixPlot is open-source software being developed within the Yale Digital Humanities Lab (DHLab). It uses a convolutional neural network and Uniform Manifold Approximation and Projection (UMAP) in order to cluster visually similar images near one another in a web browser.
  • JavisCloud is a good alternative to paperspace which also integrates with Jupyter notebooks.

Next week’s agenda will include:

  • Discussion of lecture 3 – Ethics and app deployment
    • What protocols can we put in place to mitigate bias?
    • Was there something that we learned that surprised us?
      • How can we avoid the same pitfalls?
    • Can someone showcase using volia use case?
  • Discussion on Lecture 4 – Mlist
    • Pixel Similarity
      • Image conversion and manipulation in a tensor or array
      • Mean absolute difference (L1 Norm) and root mean squared error (RMSE or L2 norm). Why and how to use them
    • SGD and why it works for machine learning
      • Steps for SGD
      • Understanding of the concept of derivatives

Fastai Study Group (lessons 1-4)
Tuesday, February 8 · 8:00 – 9:00pm
Google Meet joining info
Video call link: https://meet.google.com/yho-nhdr-yew

See you next week!

Hey mate, can we discuss possible alternate timings for the study group? 6:30 AM is turning out to be intractable for me to show up :frowning:

1 Like

Hi I’d love to join as well. I’ve done Coursera’s Deep Learning CNN course and started this course recently. The video call link doesn’t seem to be working for me though.

1 Like

Can I join too. If i am not too late😅

Sure you can join us!

HI @bbrown ,
I am siva from India. I have been learning AIML from Great lakes and have just started to learn fastai_V4.
I am interested in participating in your study group. can you lend me a spot??!!

Hi @aviz and @willsa14. I had a hard week with childcare issues so I am canceling this week’s study group (won’t happen again). Feel free to meet up via Discord if you want to go forward with today. will share next weeks google meet link later tonight.

Best,
~Bryon

@bbrown Thanks for letting me know, hang in there! @Aviz I’ll be on Discord in ten minutes just in case you do want to meet up.

Here’s the Discord link again in case anyone wants to join this week

Never mind, message me if you want the Discord link! Maybe not good to share the link in a page searchable from Google or by bots…

hi @willsa14
sorry was not able to attend the study. will be joining for sure in next weeks schedule

1 Like

I am bit late, but hope that its not closed. I’m interested. How may I attend?