Use cases for each algorithm

aymericdamien / TensorFlow-Examples

TensorFlow Tutorial and Examples for Beginners (support TF v1 & v2)

Other

43.36k stars 14.95k forks source link

Use cases for each algorithm #35

Closed alexbilbie closed 8 years ago

alexbilbie commented 8 years ago

I'm very new to machine learning and I've found this repository really useful.

I'm wondering if you could please add some notes to either the README or the individual examples detailing some example use cases that are appropriate for each algorithm?

aymericdamien commented 8 years ago

There are many different usage, but the most common:

Convolutional network: learn to classify images.
Recurrent network : learn to classify sequences of data (such as text or audio).
Neural network in general: Any classification or regression task .
AutoEncoder: Encode/Decode data, it is similar to compression process, but you are more likely to lose information when decoding (see example notebook image to compare decoded images with original ones).

jdfweb commented 8 years ago

Thanks for the info @aymericdamien !

jgschmitz commented 7 years ago

It could could also use vision and GPS to 'reserve' a lane for you on the road. Intelligent crowd planned roads and intersections would be safer.