Refactor neural network prediction function
Created a general forward propagation function that works with any
neural network models. The old code only works with specific number of
layers and nodes.
Change-Id: I5edb8aa09479c8d9c105da4b0acc89325c45092e
diff --git a/av1/encoder/ml.c b/av1/encoder/ml.c
new file mode 100644
index 0000000..3a27e58
--- /dev/null
+++ b/av1/encoder/ml.c
@@ -0,0 +1,57 @@
+/*
+ * Copyright (c) 2016, Alliance for Open Media. All rights reserved
+ *
+ * This source code is subject to the terms of the BSD 2 Clause License and
+ * the Alliance for Open Media Patent License 1.0. If the BSD 2 Clause License
+ * was not distributed with this source code in the LICENSE file, you can
+ * obtain it at www.aomedia.org/license/software. If the Alliance for Open
+ * Media Patent License 1.0 was not distributed with this source code in the
+ * PATENTS file, you can obtain it at www.aomedia.org/license/patent.
+ */
+
+#include <assert.h>
+
+#include "av1/encoder/ml.h"
+
+void av1_nn_predict(const float *features, const NN_CONFIG *nn_config,
+ float *output) {
+ int num_input_nodes = nn_config->num_inputs;
+ int buf_index = 0;
+ float buf[2][NN_MAX_NODES_PER_LAYER];
+ const float *input_nodes = features;
+
+ // Propagate hidden layers.
+ const int num_layers = nn_config->num_hidden_layers;
+ assert(num_layers <= NN_MAX_HIDDEN_LAYERS);
+ for (int layer = 0; layer < num_layers; ++layer) {
+ const float *weights = nn_config->weights[layer];
+ const float *bias = nn_config->bias[layer];
+ float *output_nodes = buf[buf_index];
+ const int num_output_nodes = nn_config->num_hidden_nodes[layer];
+ assert(num_output_nodes < NN_MAX_NODES_PER_LAYER);
+ for (int node = 0; node < num_output_nodes; ++node) {
+ float val = 0.0f;
+ for (int i = 0; i < num_input_nodes; ++i)
+ val += weights[i] * input_nodes[i];
+ val += bias[node];
+ // ReLU as activation function.
+ val = val > 0.0f ? val : 0.0f; // Could use AOMMAX().
+ output_nodes[node] = val;
+ weights += num_input_nodes;
+ }
+ num_input_nodes = num_output_nodes;
+ input_nodes = output_nodes;
+ buf_index = 1 - buf_index;
+ }
+
+ // Final output layer.
+ const float *weights = nn_config->weights[num_layers];
+ for (int node = 0; node < nn_config->num_outputs; ++node) {
+ const float *bias = nn_config->bias[num_layers];
+ float val = 0.0f;
+ for (int i = 0; i < num_input_nodes; ++i)
+ val += weights[i] * input_nodes[i];
+ output[node] = val + bias[node];
+ weights += num_input_nodes;
+ }
+}
