Tinker with a Neural Network in Your Browser
playground.tensorflow.org
Tinker with a Neural Network in Your Browser
1–10 of 122 posts
Re: Tinker with a Neural Network in Your Browser
#2When it says "right here in your browser," it's not joking. On my desktop (Safari), the window becomes unresponsive after a few iterations. Does not happen in Chrome.
On my phone (Safari/iOS 9.3), the default neural nework doesn't converge at all even after 300 iterations while it does on the desktop, which is legit weird: https://i.imgur.com/KNaXeHH.png
Re: Tinker with a Neural Network in Your Browser
#3In case you are an idiot like me, you have to train your neural network by pressing "play".
Re: Tinker with a Neural Network in Your Browser
#4This is a lot of fun. The default dataset is too easy, though, try out the Swiss Roll one!
Re: Tinker with a Neural Network in Your Browser
#5This is a lot of fun. The default dataset is too easy, though, try out the Swiss Roll one!
There is a reason why sin(X) is an input property. :p
Re: Tinker with a Neural Network in Your Browser
#6Neat stuff, fun to play with. I wasn't able to get a net to classify the swiss roll. Last time I was playing around with this stuff I found the single biggest factor in the success was the optimizer used. Is this just using a simple gradient descent? I would like to see a drop down for different optimizers.
Re: Tinker with a Neural Network in Your Browser
#7This is a lot of fun. The default dataset is too easy, though, try out the Swiss Roll one!
There is a reason why sin(X) is an input property. :p
Using sin(x) or the other input features like x^2 goes back to making it too easy, though. So far the best I can do is 7 layers of 7 which gets a loss of 0.02. 3x7 is almost cracking the Swiss Roll but can't quite finish it off and gets stuck at 0.05: https://imgur.com/Z3f2ECc ... Surprisingly, 2x8 can do it, as long as I have noise or regularization on, but 8/7 then seriously struggles. Is 16 neurons a critical limit here?
Re: Tinker with a Neural Network in Your Browser
#8Earlier quoted context omitted.
There is a reason why sin(X) is an input property. :p
Using sin(x) or the other input features like x^2 goes back to making it too easy, though. So far the best I can do is 7 layers of 7 which gets a loss of 0.02. 3x7 is almost cracking the Swiss Roll but can't quite finish it off and gets stuck at 0.05: https://imgur.com/Z3f2ECc ... Surprisingly, 2x8 can do it, as long as I have noise or regularization on, but 8/7 then seriously struggles. Is 16 neurons a critical limi…
I managed to get to 0.01 loss from only x1/x2, using 3 hidden layers, L1 regularization, a bit of added noise, and some patience: http://i.imgur.com/Y3zKpJF.png
Re: Tinker with a Neural Network in Your Browser
#9Earlier quoted context omitted.
Using sin(x) or the other input features like x^2 goes back to making it too easy, though. So far the best I can do is 7 layers of 7 which gets a loss of 0.02. 3x7 is almost cracking the Swiss Roll but can't quite finish it off and gets stuck at 0.05: https://imgur.com/Z3f2ECc ... Surprisingly, 2x8 can do it, as long as I have noise or regularization on, but 8/7 then seriously struggles. Is 16 neurons a critical limi…
I managed to get to 0.01 loss from only x1/x2, using 3 hidden layers, L1 regularization, a bit of added noise, and some patience: http://i.imgur.com/Y3zKpJF.png
Yes, noise & regularization seem to be key here. I've gotten a 2-layer with 7/8 neurons down to 0.06 and dropping but only with noise & l1: http://playground.tensorflow.org/#activation=relu®ulariza... Final loss of 0.051. Interestingly, increasing noise from 10 to 15 destroys performance, loss of 0.47.
Re: Tinker with a Neural Network in Your Browser
#10This is a lot of fun. The default dataset is too easy, though, try out the Swiss Roll one!
All the patterns except the swiss roll work best without any hidden layers at all