apaam/mlpack
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Fix Tests and Clean Implementation

Browse Source
This commit is contained in:
Shikhar Jaiswal
2018-07-23 16:25:41 +05:30
parent 9c57ab50fa
commit bc7b512db3
3 changed files with 35 additions and 18 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/mlpack/methods/ann/rbm/rbm_impl.hpp
+12 -4
View File
@@ -111,8 +111,11 @@ typename std::enable_if<std::is_same<Policy, BinaryRBM>::value, double>::type
RBM<InitializationRuleType, DataType, PolicyType>::FreeEnergy(
arma::Mat<ElemType>&& input)
{
preActivation = arma::log(1 + arma::trunc_exp((weight * input) + hiddenBias));
return -(arma::accu(preActivation) + arma::dot(input, visibleBias));
preActivation = (weight * input);
preActivation.each_col() += hiddenBias;
preActivation = arma::log(1 + arma::trunc_exp(preActivation));
return -(arma::accu(preActivation) + arma::dot(input, arma::repmat(
visibleBias, 1, input.n_cols)));
}
template<
@@ -204,7 +207,8 @@ typename std::enable_if<std::is_same<Policy, BinaryRBM>::value, void>::type
RBM<InitializationRuleType, DataType, PolicyType>::VisibleMean(DataType&& input,
DataType&& output)
{
output = weight.t() * input + visibleBias;
output = weight.t() * input;
output.each_col() += visibleBias;
LogisticFunction::Fn(output, output);
}
@@ -218,7 +222,8 @@ typename std::enable_if<std::is_same<Policy, BinaryRBM>::value, void>::type
RBM<InitializationRuleType, DataType, PolicyType>::HiddenMean(DataType&& input,
DataType&& output)
{
output = weight * input + hiddenBias;
output = weight * input;
output.each_col() += hiddenBias;
LogisticFunction::Fn(output, output);
}
@@ -281,6 +286,7 @@ void RBM<InitializationRuleType, DataType, PolicyType>::Gradient(
negativeGradient += tempNegativeGradient;
}
gradient = ((negativeGradient / negSteps) - positiveGradient);
}
@@ -313,6 +319,8 @@ void RBM<InitializationRuleType, DataType, PolicyType>::serialize(
ar & BOOST_SERIALIZATION_NVP(negSteps);
ar & BOOST_SERIALIZATION_NVP(persistence);
ar & BOOST_SERIALIZATION_NVP(poolSize);
ar & BOOST_SERIALIZATION_NVP(visibleBias);
ar & BOOST_SERIALIZATION_NVP(hiddenBias);
ar & BOOST_SERIALIZATION_NVP(weight);
ar & BOOST_SERIALIZATION_NVP(weightCube);
ar & BOOST_SERIALIZATION_NVP(spikeBias);
src/mlpack/methods/ann/rbm/spike_slab_rbm_impl.hpp
+21 -12
View File
@@ -41,9 +41,8 @@ RBM<InitializationRuleType, DataType, PolicyType>::Reset()
slabMean.set_size(poolSize, hiddenSize);
// Weight shape D * K * N
weightCube = arma::Cube<ElemType>(parameter.memptr(),
visibleSize, poolSize, hiddenSize,
false, false);
weightCube = arma::Cube<ElemType>(parameter.memptr(), visibleSize, poolSize,
hiddenSize, false, false);
// spike bias shape N * 1
spikeBias = DataType(parameter.memptr() + weight.n_elem, hiddenSize, 1,
false, false);
@@ -70,8 +69,7 @@ typename std::enable_if<std::is_same<Policy, SpikeSlabRBM>::value, double>::type
RBM<InitializationRuleType, DataType, PolicyType>::FreeEnergy(
arma::Mat<ElemType>&& input)
{
ElemType freeEnergy = 0.5 * arma::as_scalar(visiblePenalty(0) * input.t() *
input);
ElemType freeEnergy = 0.5 * visiblePenalty(0) * arma::accu(input.t() * input);
freeEnergy -= 0.5 * hiddenSize * poolSize *
std::log((2.0 * M_PI) / slabPenalty);
@@ -112,11 +110,15 @@ RBM<InitializationRuleType, DataType, PolicyType>::Phase(
SlabMean(std::move(input), std::move(spikeSamples), std::move(slabMean));
for (size_t i = 0 ; i < hiddenSize; i++)
weightGrad.slice(i) = input * slabMean.col(i).t() * spikeMean(i);
{
weightGrad.slice(i) = input * arma::repmat(slabMean.col(i).t(),
input.n_cols, 1) * spikeMean(i);
}
spikeBiasGrad = spikeMean;
visiblePenaltyGrad = -0.5 * input.t() * input;
visiblePenaltyGrad = -0.5 * arma::accu(input.t() * input)
/ std::pow(input.n_cols, 2);
}
template<
@@ -166,7 +168,9 @@ RBM<InitializationRuleType, DataType, PolicyType>::SampleVisible(
output(i) = math::RandNormal(visibleMean(i), 1.0 / visiblePenalty(0));
}
if (arma::norm(output, 2) < radius)
{
break;
}
}
if (k == numMaxTrials)
@@ -197,7 +201,9 @@ RBM<InitializationRuleType, DataType, PolicyType>::VisibleMean(
false);
for (size_t i = 0; i < hiddenSize; i++)
{
output += weightCube.slice(i) * slab.col(i) * spike(i);
}
output = ((1.0 / visiblePenalty(0)) * output);
}
@@ -237,9 +243,9 @@ RBM<InitializationRuleType, DataType, PolicyType>::SpikeMean(
{
for (size_t i = 0; i < hiddenSize; i++)
{
spikeMean(i) = LogisticFunction::Fn(0.5 * (1.0 / slabPenalty) *
arma::as_scalar(visible.t() * weightCube.slice(i) *
weightCube.slice(i).t() * visible) + spikeBias(i));
spikeMean(i) = LogisticFunction::Fn(0.5 * (1.0 / slabPenalty) * arma::accu(
visible.t() * (weightCube.slice(i) * weightCube.slice(i).t()) * visible)
/ std::pow(visible.n_cols, 2) + spikeBias(i));
}
}
@@ -255,7 +261,9 @@ RBM<InitializationRuleType, DataType, PolicyType>::SampleSpike(
DataType&& spike)
{
for (size_t i = 0; i < hiddenSize; i++)
{
spike(i) = math::RandBernoulli(spikeMean(i));
}
}
template<
@@ -272,8 +280,8 @@ RBM<InitializationRuleType, DataType, PolicyType>::SlabMean(
{
for (size_t i = 0; i < hiddenSize; i++)
{
slabMean.col(i) = (1.0 / slabPenalty) * spike(i) *
weightCube.slice(i).t() * visible;
slabMean.col(i) = arma::mean((1.0 / slabPenalty) * spike(i) *
weightCube.slice(i).t() * visible, 1);
}
}
@@ -296,6 +304,7 @@ RBM<InitializationRuleType, DataType, PolicyType>::SampleSlab(
}
}
}
} // namespace ann
} // namespace mlpack
src/mlpack/tests/rbm_network_test.cpp
+2 -2
View File
@@ -38,7 +38,7 @@ using namespace mlpack::regression;
BOOST_AUTO_TEST_SUITE(RBMNetworkTest);
BOOST_AUTO_TEST_CASE(ClassificationTest)
BOOST_AUTO_TEST_CASE(BinaryRBMClassificationTest)
{
// Normalised dataset.
int hiddenLayerSize = 100;
@@ -74,7 +74,7 @@ BOOST_AUTO_TEST_CASE(ClassificationTest)
size_t numRBMIterations = trainData.n_cols * numEpoches;
numRBMIterations /= batchSize;
optimization::StandardSGD msgd(0.06, batchSize, numRBMIterations, 0, true);
optimization::StandardSGD msgd(0.03, batchSize, numRBMIterations, 0, true);
model.Reset();
model.VisibleBias().ones();
model.HiddenBias().ones();