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

Adjust namespace of mlpack::Log

Browse Source 
Signed-off-by: Omar Shrit <omar@shrit.me>
This commit is contained in:
Omar Shrit
2022-04-27 17:57:33 -04:00
committed by Ryan Curtin
parent 37367c7ec5
commit 8fffbd55fa
3 changed files with 29 additions and 35 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/mlpack/core/math/lin_alg_impl.hpp
-2
View File
@@ -89,8 +89,6 @@ inline void WhitenUsingSVD(const arma::mat& x,
*/
inline void RandVector(arma::vec& v)
{
v.zeros();
for (size_t i = 0; i + 1 < v.n_elem; i += 2)
{
double a = math::Random();
src/mlpack/methods/det/dtree_impl.hpp
+19 -23
View File
@@ -93,7 +93,7 @@ void ExtractSplits(std::vector<std::pair<ElemType, size_t>>& splitVec,
const size_t minLeafSize)
{
// It's common sense, but we also use it in a check later.
mlpack::Log::Assert(minLeafSize > 0);
Log::Assert(minLeafSize > 0);
typedef std::pair<ElemType, size_t> SplitItem;
const size_t n_elem = end - start;
@@ -116,7 +116,7 @@ void ExtractSplits(std::vector<std::pair<ElemType, size_t>>& splitVec,
const ElemType newVal = valsVec[i];
if (lastVal < ElemType(0) && newVal > ElemType(0) && zeroes > 0)
{
mlpack::Log::Assert(padding == 0); // We should arrive here once!
Log::Assert(padding == 0); // We should arrive here once!
// The minLeafSize > 0 also guarantees we're not entering right at the
// start.
@@ -148,9 +148,8 @@ void ExtractSplits(std::vector<std::pair<ElemType, size_t>>& splitVec,
} // namespace details
template<typename MatType, typename TagType>
mlpack::det::DTree<MatType, TagType>::DTree() :
start(0),
end(0),
DTree<MatType, TagType>::DTree() :
end(0),
splitDim(size_t(-1)),
splitValue(std::numeric_limits<ElemType>::max()),
logNegError(-DBL_MAX),
@@ -188,10 +187,8 @@ DTree<MatType, TagType>::DTree(const DTree& obj) :
}
template<typename MatType, typename TagType>
DTree<MatType, TagType>& mlpack::det::DTree<MatType, TagType>::operator=(
const mlpack::det::DTree<MatType, TagType>& obj)
{
if (this == &obj)
DTree<MatType, TagType>& DTree<MatType, TagType>::operator=(
const DTree<MatType, TagT== &obj)
return *this;
// Copy the values from the other tree.
@@ -332,9 +329,8 @@ DTree<MatType, TagType>::DTree(const StatType& maxVals,
{ /* Nothing to do. */ }
template<typename MatType, typename TagType>
mlpack::det::DTree<MatType, TagType>::DTree(MatType & data) :
start(0),
end(data.n_cols),
DTree<MatType, TagType>::DTree(MatType & data) :
end(data.n_cols),
maxVals(arma::max(data, 1)),
minVals(arma::min(data, 1)),
splitDim(size_t(-1)),
@@ -443,8 +439,8 @@ bool DTree<MatType, TagType>::FindSplit(const MatType& data,
// Ensure the dimensionality of the data is the same as the dimensionality of
// the bounding rectangle.
mlpack::Log::Assert(data.n_rows == maxVals.n_elem);
mlpack::Log::Assert(data.n_rows == minVals.n_elem);
Log::Assert(data.n_rows == maxVals.n_elem);
Log::Assert(data.n_rows == minVals.n_elem);
const size_t points = end - start;
@@ -503,7 +499,7 @@ bool DTree<MatType, TagType>::FindSplit(const MatType& data,
{
// Ensure that the right node will have at least the minimum number of
// points.
mlpack::Log::Assert((points - position) >= minLeafSize);
Log::Assert((points - position) >= minLeafSize);
// Now we have to see if the error will be reduced. Simple manipulation
// of the error function gives us the condition we must satisfy:
@@ -592,8 +588,8 @@ double DTree<MatType, TagType>::Grow(MatType& data,
const size_t maxLeafSize,
const size_t minLeafSize)
{
mlpack::Log::Assert(data.n_rows == maxVals.n_elem);
mlpack::Log::Assert(data.n_rows == minVals.n_elem);
Log::Assert(data.n_rows == maxVals.n_elem);
Log::Assert(data.n_rows == minVals.n_elem);
double leftG, rightG;
@@ -667,7 +663,7 @@ double DTree<MatType, TagType>::Grow(MatType& data,
else
{
// We can make this a leaf node.
mlpack::Log::Assert((size_t) (end - start) >= minLeafSize);
Log::Assert((size_t) (end - start) >= minLeafSize);
subtreeLeaves = 1;
subtreeLeavesLogNegError = logNegError;
}
@@ -821,7 +817,7 @@ double DTree<MatType, TagType>::PruneAndUpdate(const double oldAlpha,
gT = alphaUpper - std::log((double) (subtreeLeaves - 1));
}
mlpack::Log::Assert(gT < std::numeric_limits<double>::max());
Log::Assert(gT < std::numeric_limits<double>::max());
return std::min((double) gT, std::min(leftG, rightG));
}
@@ -863,7 +859,7 @@ bool DTree<MatType, TagType>::WithinRange(const VecType& query) const
template<typename MatType, typename TagType>
double DTree<MatType, TagType>::ComputeValue(const VecType& query) const
{
mlpack::Log::Assert(query.n_elem == maxVals.n_elem);
Log::Assert(query.n_elem == maxVals.n_elem);
if (root == 1) // If we are the root...
{
@@ -908,7 +904,7 @@ TagType DTree<MatType, TagType>::TagTree(const TagType& tag, bool every)
template<typename MatType, typename TagType>
TagType DTree<MatType, TagType>::FindBucket(const VecType& query) const
{
mlpack::Log::Assert(query.n_elem == maxVals.n_elem);
Log::Assert(query.n_elem == maxVals.n_elem);
if (root == 1) // If we are the root...
{
@@ -985,8 +981,8 @@ void DTree<MatType, TagType>::FillMinMax(const StatType& mins,
template <typename MatType, typename TagType>
template <typename Archive>
void mlpack::det::DTree<MatType, TagType>::serialize(Archive& ar,
const uint32_t /* version */)
void DTree<MatType, TagType>::serialize(Archive& ar,
const uint32_t /* version */)
{
ar(CEREAL_NVP(start));
ar(CEREAL_NVP(end));
src/mlpack/methods/reinforcement_learning/q_networks/categorical_dqn.hpp
+10 -10
View File
@@ -81,21 +81,21 @@ class CategoricalDQN
vMax(config.VMax()),
isNoisy(isNoisy)
{
network.Add(new mlpack::ann::Linear<>(inputDim, h1));
network.Add(new mlpack::ann::ReLULayer<>());
network.Add(new ann::Linear<>(inputDim, h1));
network.Add(new ann::ReLULayer<>());
if (isNoisy)
{
noisyLayerIndex.push_back(network.Model().size());
network.Add(new mlpack::ann::NoisyLinear<>(h1, h2));
network.Add(new mlpack::ann::ReLULayer<>());
network.Add(new ann::NoisyLinear<>(h1, h2));
network.Add(new ann::ReLULayer<>());
noisyLayerIndex.push_back(network.Model().size());
network.Add(new mlpack::ann::NoisyLinear<>(h2, outputDim * atomSize));
network.Add(new ann::NoisyLinear<>(h2, outputDim * atomSize));
}
else
{
network.Add(new mlpack::ann::Linear<>(h1, h2));
network.Add(new mlpack::ann::ReLULayer<>());
network.Add(new mlpack::ann::Linear<>(h2, outputDim * atomSize));
network.Add(new ann::Linear<>(h1, h2));
network.Add(new ann::ReLULayer<>());
network.Add(new ann::Linear<>(h2, outputDim * atomSize));
}
}
@@ -181,7 +181,7 @@ class CategoricalDQN
{
for (size_t i = 0; i < noisyLayerIndex.size(); ++i)
{
boost::get<mlpack::ann::NoisyLinear<>*>
boost::get<ann::NoisyLinear<>*>
(network.Model()[noisyLayerIndex[i]])->ResetNoise();
}
}
@@ -234,7 +234,7 @@ class CategoricalDQN
std::vector<size_t> noisyLayerIndex;
//! Locally-stored softmax activation function.
mlpack::ann::Softmax<> softMax;
ann::Softmax<> softMax;
//! Locally-stored activations from softMax.
arma::mat activations;