Thank you, Sir.
Thank you for the invite, Jeremy!
Looking forward to continuing my ML journey and contributing back to this great community!
Thank you Jermey! Looking forward to grow on ML with FastAI
Delighted to be able to participate!!
Given that there is transcription and translation work to be done I assume weāll have access to a recording? I ask as the live stream starts at 2.30am Dublin time which generally isnāt my most productive time of the day 
would love to be able to watch it the following morning when Iām fresh and wide awake if thatās possible?
1 Like
Wow thanks a lot ! Iāll start looking into fastai2 so that I can help as much as possible.
Thank you @jeremy for the invitation! Thatās awesome!
Wow! Iām now part of a Deep Learning Gang how about that !
Knowing nothing about coding 18 month from now this feels pretty good.
While I havenāt yet had the pleasure to play with fastai2, the main issue I have with the previous issues of the course was sound quality!
I donāt know what microphone was used but the result is pretty poor considering the quality of the verbalized content. Most of the best material is said by Jeremy and the notebooks are merely a support to understand the underlying mechanics. Without Jeremyās explainations the tricky latent mechanics of DL such as backrprop, debiasing, normalizations techniques, training mechanics and else wouldnāt be half as easy to understand. Also sometimes, words or pieces of sentences gets just eaten by the āmixed-precisionā recording quality 
, even listening them multiple time doesnāt help.
So maybe itās worth considering a USB Lavalier microphone or something wireless.
No need to buy a top notch expensive microphone while you can reduce as much as possible the distance between the source and the mic.
Also be careful of the audio encoding quality in the recording software.
Itās too bad the course starts March 17th, Iāll be in SF the second half of February, I hopped I might be able to attend to at least one of these great courses.
1 Like
Thank you @jeremy excited for the course and to test out fastai2 and the new lib nbdev!
Thank you!! Iām excited about the course and the opportunity 
When I was a kid I used to wake up in the middle of the night to watch NBA games live (Poland is 9 hours ahead of PST), feeling the same energy now to tune in to the fast.ai v4 live sessions
4 Likes
Thanks a lot. Iāve been quiet for some time and I hope will be honoring the invitation.
Sure Daniel,
(I havenāt owned a box in 12-ish years, so please do consider my lack of expertise here)
The parts havenāt arrived yet, but Iāll def. try to post the build process once I get my hands on it.
I built my purchase list primarily based on this post. https://timdettmers.com/2018/12/16/deep-learning-hardware-guide/
This is what my list looks like, trying to keep things to a mininum, and a possibility for 1 more GPU. https://pcpartpicker.com/list/VmPvq3
Sidenote : I first opted for AMD cpus(Ryzen 5 series) CPU, as they have better computation/money, but after reading up several posts about how MKL is crippled in AMD, I decide to stick with Intel. Not happy about buying old 14nm tech, but I donāt want to spend much time yak shaving the AMD/MKL speedup-hacks either.
https://sites.google.com/a/uci.edu/mingru-yang/programming/mkl-has-bad-performance-on-an-amd-cpu
4 Likes
Thanks a lot for the invite to participate in the course again! Itās a privilidge to be part of this community.
Look forward to learning more about DL and the new library, and will do my best to contribute as much as I can.
Thank you very much for the invitation! I am learning more and more with each run, starting a year with your course is one of the best traditions Iāve acquired over the years
Want to purchase a laptop. Any suggestion is greatly appreciated.
1 Like
Thanks a lot for the invitation Jeremy. I will try to contribute something to the community.
1 Like
Thanks for invitation. For last year, I was not able to keep up with of pace of FASTAI with fastai2 development. I had feeling that my foundation for deep learning was so brittle that I could not move ahead with it. So I started reading old papers, going over old fastai lectures and reading http://neuralnetworksanddeeplearning.com/ and https://www.deeplearningbook.org along with working on small code snippets to start from scratch. Learning process is still incomplete but overall feeling more confident. With this course, I hope I can absorb more material and can restart contributing towards this community again.
1 Like
Humbled to be a part of this, really looking forward to the course!
I havenāt really looked much into this. My naive assumption is that getting a laptop of comparative GPU/cuda performance would def. be more expensive compared to building a box.
With a quick search, hereās a recently updated post I found on GPU enabled laptops that might be helpful as a starting point for further investigation.
NOTE : I have not tested any of these, and you want to get laptops ONLY with NVIDIA GPUs (cuda cores)
Perhaps, somebody with more hands-on experience in laptops well suited for Deep Learning (and Fast.ai) can help you further.
1 Like
Wow 6:30pm PST is 3:30 am European Central Timeā¦ this will be hard to manage ā¦
It definitely is more expensive, relatively speaking. If itās viable, a pc thatās connected to the internet 24x7 is a much more flexible and economical option. You can ssh into it from anywhere, anytime with an internet connection, and also leave notebooks running without leaving your laptop on.
You wonāt have to worry about thermal throttling, gain from desktop GPU performance, and most importantly, have the option to upgrade your hardware over time.
I was considering buying a laptop for mobility myself, but with a desktop setup as described above, I find it more flexible in terms of mobility too
2 Likes