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

advantage and value networks work!

Browse Source
This commit is contained in:
nishantkr18
2020-05-25 19:34:26 +05:30
parent 7dc424359e
commit 31294089bf
2 changed files with 21 additions and 27 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/mlpack/methods/ann/layer/concat.hpp
-8
View File
@@ -138,14 +138,6 @@ class Concat
arma::Mat<eT>& gradient,
const size_t index);
/*
* Add a new module to the model.
*
* @param layer The Layer to be added to the model.
*/
template<typename LayerType>
void Add(const LayerType& layer) { network.push_back(new LayerType(layer)); }
/*
* Add a new module to the model.
*
src/mlpack/methods/reinforcement_learning/q_networks/dueling_dqn.hpp
+21 -19
View File
@@ -29,8 +29,8 @@ using namespace mlpack::ann;
*/
template <
typename FeatureNetworkType = FFN<EmptyLoss<>, GaussianInitialization>,
typename AdvantageNetworkType = Sequential<>,
typename ValueNetworkType = Sequential<>
typename AdvantageNetworkType = Sequential<>*,
typename ValueNetworkType = Sequential<>*
>
class DuelingDQN
{
@@ -57,21 +57,24 @@ class DuelingDQN
valueNetwork(),
advantageNetwork()
{
valueNetwork.Add(new Linear<>(h1, h2));
valueNetwork.Add(new ReLULayer<>());
valueNetwork.Add(new Linear<>(h2, 1));
valueNetwork = new Sequential<>();
valueNetwork->Add(new Linear<>(h1, h2));
valueNetwork->Add(new ReLULayer<>());
valueNetwork->Add(new Linear<>(h2, 1));
advantageNetwork.Add(new Linear<>(h1, h2));
advantageNetwork.Add(new ReLULayer<>());
advantageNetwork.Add(new Linear<>(h1, outputDim));
advantageNetwork = new Sequential<>();
advantageNetwork->Add(new Linear<>(h1, h2));
advantageNetwork->Add(new ReLULayer<>());
advantageNetwork->Add(new Linear<>(h2, outputDim));
Concat<> concat = new Concat<>();
concat.Add<Sequential<>>(valueNetwork);
concat.Add<Sequential<>>(advantageNetwork);
Concat<>* concat = new Concat<>(true);
concat->Add(valueNetwork);
concat->Add(advantageNetwork);
featureNetwork.Add(new Linear<>(inputDim, h1));
featureNetwork.Add(new ReLULayer<>());
featureNetwork.Add(concat);
this->ResetParameters();
}
DuelingDQN(FeatureNetworkType featureNetwork,
@@ -81,10 +84,11 @@ class DuelingDQN
advantageNetwork(std::move(advantageNetwork)),
valueNetwork(std::move(valueNetwork))
{
Concat<> concat = new Concat<>();
concat.Add<Sequential<>>(valueNetwork);
concat.Add<Sequential<>>(advantageNetwork);
Concat<>* concat = new Concat<>(true);
concat->Add(valueNetwork);
concat->Add(advantageNetwork);
featureNetwork.Add(concat);
this->ResetParameters();
}
/**
@@ -114,12 +118,12 @@ class DuelingDQN
*/
void Forward(const arma::mat state, arma::mat& output)
{
// arma::mat output, advantage, value;
// featureNetwork.Forward(state, output);
arma::mat advantage, value;
featureNetwork.Forward(state, output);
// actionValue = advantage.each_row() +
// (value - arma::mean(arma::mean(advantage)));
// networkOutput = output;
networkOutput = output;
}
/**
@@ -151,8 +155,6 @@ class DuelingDQN
void ResetParameters()
{
featureNetwork.ResetParameters();
advantageNetwork.ResetParameters();
valueNetwork.ResetParameters();
}
//! Return the Parameters.