Visualizing Neural Network Internals

[@nktthegreat]

I bought your NNFS book a couple of years ago. By far the best book on Neural networking, probably in the entire ML domain. Absolute gold.

[@musicdev]

Also bought it a couple of years ago and completely agree!

[@prateekyadav9811]

निर्मित, क्या तुमने यह पाठ्यपुस्तक पूरी कर ली?

[@digital_down]

I’d recommend more visualization tutorials like this… fascinating

[@isaacandrewdixon]

Seconded. Very interesting to visual learners and people who don't know programming.

[@franklydoodle350]

I was literally just thinking about this. A great visual tool called PerceptiLabs recently was sunsetted. It's unfortunate as it was one of the last tools I know of that tried to do what you're showing in this video. Hopefully something similar gets picked up by the community as I think we're in dire need of some visuals.

[@stackingwins]

This is the first visualization I have seen. You are before you time sir, this will be the norm for models very soon. Excellent work!

[@asksearchknock]

Very cool - and surprisingly useful when pitching AI to investors who love to see the flashing patterns and force directed graphs (sadly I’m being serious)

[@pebre79]

I appreciate all your work. These visualizations confirm this that deep learning is a black box

[@davidanalyst671]

"visualizations confirm that deep learning is a black box" for stupid people

[@busterfranken9105]

Friggin awesome. Nostalgia never misses

[@ferdyg3520]

'which better be pytorch' gotta love it

[@dineshasavsani154]

Very good video keep it up .... love from india ❤

[@muhammadaburayan7471]

Great Presentation, very helpful.

[@hola-kx1gn]

Very cool tool for introductory ML courses at unis, etc.

[@piotrzaborszczyk483]

sentdex asking everybody what is going on but not actually caring about the answer

[@TheFarmacySeedsNetwork]

Freaking excellent dude. Thank you!

[@prateekyadav9811]

Harrison, please inform us if you plan to continue the tutorial series for NNFS. It's a brilliant piece of work. Please do tell us either way. You been super quiet on this for long :P Mighty thanks for your content!

[@lucasb.2410]

likin' the great old sentdex style ! :p

[@TurnLeftTurnRight2]

good content tanks man aways

[@kotcraftchannelukraine6118]

I watched your video of GameGAN (GTA-5 running only on a neural network). This is mind blowing! I try to recreate it with Tux Racer game

[@BarkofthePhoenix]

Very cool, I'd love to use that for a group project I'm working on right now, is there an ETA for the full code?

[@sentdex]

Whoops, forgot to upload the notebooks to that URL in desc. Meant to have it up already. Should be there now, enjoy!

[@BarkofthePhoenix]

@@sentdex Cheers!

[@yuanfongsu]

Very useful! Thank you for sharing. One question, since we could know which neuron is more important for a specific class. Can we find out what filters (or kernels) in the three layers makes it important to the specific class?

[@carnap355]

In PyTorch this can be automated using register_forward_hooks. It even works on custom layers I make. Only issue is how do I visualize convolutional layer weights when there are 128 64x2x2 kernels, so I think I show first 9 kernels in a grid and take first 3 channels

[@deenadayalan03]

Great video

[@sguha7850]

Is it possible to visualize the network architecture directly from the code I build?

[@bhaskarmondal7461]

hei just a simple question to you all, Is it possible to make an web app like streamlit out of it ?

[@thesmallestatom]

the concept of “hidden layers” might be the biggest crime. they are NOT hidden, and they can be introspected.

[@serta5727]

Very cool 😎

[@lolwtf229]

Hey dude, I gotta tell you something, I owe you my career.🙏🏻🙏🏻🙏🏻 keep it up plz🙏🏻👏🏼

[@7dedlysins193]

what do you do?

[@JaysenTC]

I miss the old white idle coding days

[@sentdex]

True. Not old style til that lol. Don't think I can stare at the white bg though.

[@Lavamar]

Will there be more NNFS videos?

[@sentdex]

yes

[@kanishkkala16]

I wanna end up as a data scientist, I have done basics of python well, doing dsa rn, on binary search (array medium questions done) , I should continue my dsa till 3-400 ques on leetcode right, then stats, calcs, probability, Algaberic exp, then machine learning , python libs etc can someone review this roadmap I have in mind

[@sutsuj6437]

fourth?

[@erentas7391]

second

[@orangejuicegang8233]

5th

[@danieldutu2635]

first

[@ShubhamShubhra]

Third

[@SravanTechTalks]

third

[@acebasinnation88888]

And your asking money for this.... ok

[@davidanalyst671]

we dont want visualizations. we want a GUI tool to adjust the network in order to create the hotspots we want. Sure, shirt and coat are confused, but not if we create two different hotspots based on the thickness of the arms of the product. It wouldn't even be close. make a tool to use gui crap to adjust a NN and make it conceptual

[@chrischoir3594]

this is incoherent

[@chrischoir3594]

Python is the worst

[@BerndausderPfalz]

Third