GitHub - tony2n2/Machine-learning-neural-network-exercises

Branches Tags

Name		Name	Last commit message	Last commit date
Latest commit History 1 Commit
.gitignore		.gitignore
README		README
checkNNGradients.m		checkNNGradients.m
computeNumericalGradient.m		computeNumericalGradient.m
debugInitializeWeights.m		debugInitializeWeights.m
displayData.m		displayData.m
ex4-2.m		ex4-2.m
ex4.m		ex4.m
ex4data1.mat		ex4data1.mat
ex4weights.mat		ex4weights.mat
fmincg.m		fmincg.m
nnCostFunction.m		nnCostFunction.m
predict.m		predict.m
randInitializeWeights.m		randInitializeWeights.m
sigmoid.m		sigmoid.m
sigmoidGradient.m		sigmoidGradient.m
submit.m		submit.m
submitWeb.m		submitWeb.m
test.csv		test.csv
test.m		test.m
test.txt		test.txt
test2.csv		test2.csv
train.csv		train.csv

Repository files navigation

Machine learning neural network exercise

Solution to exercises of an online course. The neural network is capable of recognizing handwritten digits with approx. 93% accuracy.

Implemented with GNU Octave