2022 Study Group Anyone?

Part 1 2022
103

Hi @Mukeshraj sorry for the delay in putting up the link. I’ll put the link up shortly for next Tuesday’s study group

1 Like
104

@Mukeshraj I won’t have time to meet up again before next Tuesday, but here’s my notebook for the tabular data playground in case you want to look:

1 Like
105

Ok @willsa14 we will meet up next week…

106

Hi all,

I would also like to join the study group. I’m studying through fastbook and have completed 12 chapters.

107

Sure @sadivam, you can join us at the next study group on Tuesday @ 8pm EST. I will post the google meet shortly.

~Bryon

108

Hi Everyone, So for last week and this coming week, the study group is trying on our ML application muscles a bit and we have diverted from the lectures and the readings to see how much we have been able to retain thus far (next week we will go back to the lectures FYI).

Though we are all in different places in our studies, we thought this would be a good method for retaining what we have learned so far.

Specifically, we are using the following Kaggle Competition as put of our study.

In the next study group, we will be going over our approach with ML with tabular data from this project and using the practices we learned in this FastAI course. As such, if you would like to just us for the next study group, please review the competition and come prepared with questions, Eurka and suggested approaches to solving the current ML use case.

FastAI Study Group 2022
Tuesday, May 17 · 8:00 – 9:00pm
Google Meet joining info
Video call link: https://meet.google.com/rki-bzee-mpu

1 Like
109

I’m going to be ten minutes late today. See you soon!

1 Like
110

Here’s my notebook from today:

111

So I found this working in the tabular tutorial

Note : Sometimes with tabular data, your y 's may be encoded (such as 0 and 1). In such a case you should explicitly pass y_block = CategoryBlock in your constructor so fastai won’t presume you are doing regression.

1 Like
112

So working through the Kaggle 2022 Tabular Playground has been a great opportunity to begin applying the methods we have so far learned during the FastAI course. Ironically, the content covered in Lecture 7 provides some great concepts to apply to this competition.

In the next study group, we continue covering the folowing topics:

-leveraging Kaggles APIs for puling in and opening targeted datasets
-exploring methods such as latent factors
-managing categorical and continuous variables for training
-incorporating datablocks with tabular data formats

Hope you can join us!

2022 FastAI Study Group
Tuesday, May 24 · 8:00 – 9:00pm
Google Meet joining info
Video call link: https://meet.google.com/sqk-shwt-tub

113

Hi guys,

1 Like
114

Here’s a cool thing for learning about LightGBM parameters:

115

Hi! Interested in joining the next session of the study group (going through chap 6 atm). Has the next session been scheduled @bbrown ? Thanks! :pray:

116

We will continue to dig into leveraging ML techniques with tabular data formats via the Kaggle ML Project.

FastAI Study Group 2022
Tuesday, May 31 · 8:00 – 9:00pm
Google Meet joining info
Video call link: https://meet.google.com/pnf-gvpp-xts

117

Feel free to join us at our next session tomorrow!

118

I’m interested in joining , hope i get some answers as no one seems to reply to the questions I ask on this forum maybe a new group will be open up for discussions.

119

It’s on @Mukeshraj ! 0.03% AUC difference :laughing:

image
image1445×315 36.5 KB

Also nice to see your kernel featured in some discussions.

1 Like
120

hello, couldn’t make it to the last one, but interested in joining the next one. Could someone pls post the date and time for it along with the time zone? Thanks :pray:

121

Bryon Meetup?

122

Yeah, i think that’s the only one i’ve seen in this thread that’s active.