ScratchData LogoScratchData
Back to cactus3's profile

training a single neuron

CAcactus3•Created June 17, 2017
training a single neuron
27
21
1263 views
View on Scratch

Instructions

uses back propagation!

Description

uses the hyperbolic tangent as an activation function, the 2/(e^2x + x^-2x) is the derivative of the tanh function uses gradient descent with no momentum (because it's not necessary for 1 Neuron :) source: http://galaxy.agh.edu.pl/~vlsi/AI/backp_t_en/backprop.html

Project Details

Project ID166431436
CreatedJune 17, 2017
Last ModifiedNovember 23, 2020
SharedJune 18, 2017
Visibilityvisible
CommentsAllowed