Merge branch 'fix_optim_dataType' into 'dev'

Fix tensor data type in optimizer (SGD, Adam)

See merge request !42

include/aidge/learning/optimizer/Adam.hpp

@@ -65,15 +65,15 @@ public:
 
         Tensor alpha = Tensor(learningRate() * mSqrtReversedBeta2Power / mReversedBeta1Power);
         alpha.setBackend(mParameters[0]->getImpl()->backend());
+        alpha.setDataType(mParameters[0]->dataType());
 
         Tensor epsilon_hat = Tensor(this->getAttr<AdamAttr::Epsilon>() * mSqrtReversedBeta2Power);
         epsilon_hat.setBackend(mParameters[0]->getImpl()->backend());
+        epsilon_hat.setDataType(mParameters[0]->dataType());
 
         if (mLRScheduler.step() == 0) {
             for (std::size_t i = 0; i < mParameters.size(); ++i) {
-                mMomentum1[i].setDataType(mParameters[i]->grad()->dataType());
                 mMomentum1[i].zeros();
-                mMomentum2[i].setDataType(mParameters[i]->grad()->dataType());
                 mMomentum2[i].zeros();
             }
         }
@@ -94,14 +94,20 @@ public:
         for (std::size_t i = 0; i < parameters.size(); ++i) {
             mMomentum1[i] = Tensor(parameters[i]->dims());
             mMomentum1[i].setBackend(parameters[i]->getImpl()->backend());
+            mMomentum1[i].setDataType(parameters[i]->dataType());
             mMomentum2[i] = Tensor(parameters[i]->dims());
             mMomentum2[i].setBackend(parameters[i]->getImpl()->backend());
+            mMomentum2[i].setDataType(parameters[i]->dataType());
         }
         if (parameters.size() > 0) {
             mBeta1.setBackend(mParameters[0]->getImpl()->backend());
+            mBeta1.setDataType(parameters[0]->dataType());
             mReversedBeta1.setBackend(mParameters[0]->getImpl()->backend());
+            mReversedBeta1.setDataType(parameters[0]->dataType());
             mBeta2.setBackend(mParameters[0]->getImpl()->backend());
+            mBeta2.setDataType(parameters[0]->dataType());
             mReversedBeta2.setBackend(mParameters[0]->getImpl()->backend());
+            mReversedBeta2.setDataType(parameters[0]->dataType());
         }
     }
 };

include/aidge/learning/optimizer/SGD.hpp

@@ -58,7 +58,9 @@ public:
     void update() override final {
         mLR = Tensor(learningRate());
         mLR.setBackend(mParameters[0]->getImpl()->backend());
+        mLR.setDataType(mParameters[0]->dataType());
         mWeightDecay.setBackend(mParameters[0]->getImpl()->backend());
+        mWeightDecay.setDataType(mParameters[0]->dataType());
 
         if (mLRScheduler.step() == 0) {
             for (std::size_t i = 0; i < mParameters.size(); ++i) {
@@ -81,10 +83,13 @@ public:
         for (std::size_t i = 0; i < parameters.size(); ++i) {
             mGradientInertia[i] = Tensor(parameters[i]->dims());
             mGradientInertia[i].setBackend(parameters[i]->backend());
+            mGradientInertia[i].setDataType(parameters[i]->dataType());
         }
         if (parameters.size() > 0) {
             mReversedDampening.setBackend(mParameters[0]->getImpl()->backend());
+            mReversedDampening.setDataType(parameters[0]->dataType());
             mMomentum.setBackend(mParameters[0]->getImpl()->backend());
+            mMomentum.setDataType(parameters[0]->dataType());
         }
     }
 };