kafaichoi · July 2, 2017 00:16
diff --git a/gistfile1.txt b/gistfile1.txt
 It's consisted of encoder and decoder.
 Encoders(Autoencoders) are networks(unsuperprised learning, feed-forward), which try to reconstruct their own input
 (reproduct input as output). You construct the network so that it reduces the input size by using one or more hidden layers,
 until it reaches a reasonably small hidden layer in the middle. As a result your data has been compressed (encoded) into
 a few variables. From this hidden representation the network tries to reconstruct (decode) the input again.

 In order to do a good job at reconstructing the input the network has to learn a good representation of the data in
 the middle hidden layer. This can be useful for dimensionality reduction, or for generating new “synthetic” data
 from a given hidden representation.
	It's consisted of encoder and decoder.
	Encoders(Autoencoders) are networks(unsuperprised learning, feed-forward), which try to reconstruct their own input
	(reproduct input as output). You construct the network so that it reduces the input size by using one or more hidden layers,
	until it reaches a reasonably small hidden layer in the middle. As a result your data has been compressed (encoded) into
	a few variables. From this hidden representation the network tries to reconstruct (decode) the input again.

	In order to do a good job at reconstructing the input the network has to learn a good representation of the data in
	the middle hidden layer. This can be useful for dimensionality reduction, or for generating new “synthetic” data
	from a given hidden representation.