p.238 in NNFS

include/modules/neural_networks/activation_functions/Activation_Softmax_Loss_CategoricalCrossentropy.h

@@ -27,41 +27,45 @@ namespace neural_networks{
 			utils::Matrix<Td> outputs;
 			utils::Matrix<Td> dinputs;
 
-	        utils::Vector<Td> forward(const utils::Matrix<Td>& inputs, const utils::Matrix<Ti>& y_true){
+	        Td forward(const utils::Matrix<Td>& inputs, const utils::Matrix<Ti>& y_true){
 				
 				// Output layer's activation function
 				activation.forward(inputs);
 	        	// Set the output
 			    outputs = activation.outputs;
 			    // Calculate and return loss value
-			    Td data_loss = loss.calculate(inputs, y_true);
+			    Td data_loss = loss.calculate(outputs, y_true);
 			    return data_loss;
 	        }
 
 	        void backward(const utils::Matrix<Td>& dvalues, const utils::Matrix<Ti>& y_true){
 
-	        	// Number of samples
-	        	const uint64_t samples = y_true.rows();
+	        	const uint64_t samples 	= dvalues.rows();
+	        	const uint64_t cols 	= dvalues.cols();
 
-	        	// If the labels are one-hot encoded,
-	        	// turn them into discrete values
+	        	dinputs = dvalues; // Copy
 
-	        	const uint64_t rows = dvalues.rows();
-	        	const uint64_t cols = dvalues.cols();
 
-	        	if ((dinputs.rows() != rows) || dinputs.cols() != cols){
-	        		dinputs.resize(rows, cols);
-	        	}
-
-	        	for (uint64_t i = 0; i < rows; ++i){
-	        		Td dot = Td{0};
-	        		for (uint64_t j = 0; j < cols; ++j){
-	        			dot += outputs(i,j) * dvalues(i,j);
+	        	if (y_true.cols() == 1){
+	        		for (uint64_t i = 0; i < samples; ++i){
+	        			uint64_t class_idx = static_cast<uint64_t>(y_true(i, 0));
+	        			dinputs(i, class_idx) -= Td{1};
 	        		}
-	        		for (uint64_t j = 0; j < cols; ++j){
-	        			dinputs(i,j) = outputs(i,j) * (dvalues(i,j) - dot);
+	        	} else{
+	        		// one-hot: dinputs = dvalues - y_true
+	        		for (uint64_t i = 0; i < samples; ++i){
+	        			for (uint64_t j = 0; j < cols; ++j){
+	        				dinputs(i,j) -= static_cast<Td>(y_true(i,j));
+	        			}
 	        		}
 	        	}
+
+	        	// divide by samples
+        		for (uint64_t i = 0; i < samples; ++i){
+        			for (uint64_t j = 0; j < cols; ++j){
+        				dinputs(i,j) /= static_cast<Td>(samples);
+        			}
+        		}
 	        }
 	};
 

include/modules/neural_networks/loss/Loss_CategoricalCrossentrophy.h

@@ -54,6 +54,7 @@ namespace neural_networks{
 		
 		void backward(const utils::Matrix<Td>& dvalues, const utils::Matrix<Ti>& y_true) override{
 
+
 			// Number of samples
 			const Td samples = static_cast<Td> (y_true.rows());
 			// Number of labels in every sample