Bitcoin Oracle

Uses historical data to train a neural network to predict bitcoin price.

This repository doesn't have updated historical data but you can download it from bitcoincharts.com API: http://api.bitcoincharts.com/v1/csv/bitstampUSD.csv.gz

Name		Name	Last commit message	Last commit date
Latest commit History 12 Commits
.gitignore		.gitignore
BCHARTS-BITSTAMPUSD.csv		BCHARTS-BITSTAMPUSD.csv
README.md		README.md
a.mat		a.mat
ab.mat		ab.mat
bestfit.mat		bestfit.mat
bitstamp-6hours-open-high-low-close-volumeBtc-volumeUsd-weightedPrice.csv		bitstamp-6hours-open-high-low-close-volumeBtc-volumeUsd-weightedPrice.csv
bitstampUSD-sample.csv		bitstampUSD-sample.csv
bitstampUSD-test.csv		bitstampUSD-test.csv
checkNNGradients.m		checkNNGradients.m
computeNumericalGradient.m		computeNumericalGradient.m
debugInitializeWeights.m		debugInitializeWeights.m
denormalize.m		denormalize.m
denormalizeBetween.m		denormalizeBetween.m
ex4data1.mat		ex4data1.mat
fmincg.m		fmincg.m
getDeltas.m		getDeltas.m
hola.mat		hola.mat
nnCostFunction.m		nnCostFunction.m
nnCostFunctionVectorized.m		nnCostFunctionVectorized.m
normalize.m		normalize.m
normalizeBetween.m		normalizeBetween.m
normalizePrediction.m		normalizePrediction.m
octave-workspace		octave-workspace
parseData.m		parseData.m
parseData2.m		parseData2.m
plotPrediction.m		plotPrediction.m
predict.m		predict.m
predict2.m		predict2.m
predictLinear.m		predictLinear.m
predictNormalized.m		predictNormalized.m
randInitializeWeights.m		randInitializeWeights.m
sigmoid.m		sigmoid.m
sigmoidGradient.m		sigmoidGradient.m
trainLinearNN.m		trainLinearNN.m
trainLogisticNN.m		trainLogisticNN.m

Provide feedback