
Champo
@champonation
Followers
55
Following
145
Media
137
Statuses
947
Joined December 2015
I tweaked @welchlabs 's code by adding my own gradient descent implementation and made it learn the MNIST dataset made by @ylecun . All done from scratch other than numpy #ai #NeuralNetworks #deeplearning #NN #MachineLearning #python #DataScience
2
0
5
Congratulations to @quasimondo for developing an artwork that basically has a really time painter constantly mixing things up.
Sold! Mario Klingemann's pioneering Artificial Intelligence artwork sells to an online bidder for £40,000 in the artist's auction debut 🤖 #SothebysContemporary @quasimondo
0
0
2
@KellerRinaudo Do you think having a larger plane carrying more blood that is able to reach all the hospitals in one long route is more efficient that many smaller planes? Just curious. Again amazing company you built!.
1
0
1
This is a Venezuelan poodle moth. I just recently learned about them (not much is known about them), but they do look cool! #bugs #insects #nature #climatechange .
0
0
0
For my last problem (z = sin(x)sin(y)) , hopefully, you ended up with a graph like this. This isn't much to explain here, you would just need to understand how the trig functions behave. #graphdaily
1
0
0
Hopefully you should have gotten a graph that looks something like this: This is an interesting equation because it infinitely extending downwards while narrowing, BUT doesn't touch the line where x and y both equal 0. #graphdaily #math
0
0
0
Awesome video by @_brohrer_ to learn more about how a network "understand" the training data to make predictions.#ai #deeplearing #NeuralNetworks.
www.youtube.com
Part of the End-to-End Machine Learning Course 193, How Neural Networks Work at http://e2eml.school/193Blog post: https://brohrer.github.io/what_nns_learn.ht...
0
1
8
RT @dennybritz: The only master of optimization we know is nature. So it probably makes sense to look for more “nature-inspired” algorithms….
0
38
0
Extended origami flasher made by #robertlang, extended by myself. #Origami #mathematics #tessellation
1
0
0
These are the images I processed in my research: Their large pixel dimensions proved to be a computationally hard task. But thanks to #opencv, #tensorflow and @astropy, I was able to achieve results which were on par with hand classified surveys! Those libraries are amazing! #ai.
1
1
3
Just curious on why are steps broken up in computational graphs, especially, why are derivatives broken up via the chain rule? Ex: the logistic regression example in @AndrewYNg 's neural network tutorial (nothing against the series, it's amazing!) #ai #deeplearning #python #ML.
0
0
2