diff options
Diffstat (limited to 'examples/graph_deepspeech_v0_4_1.cpp')
| -rw-r--r-- | examples/graph_deepspeech_v0_4_1.cpp | 238 |
1 files changed, 120 insertions, 118 deletions
diff --git a/examples/graph_deepspeech_v0_4_1.cpp b/examples/graph_deepspeech_v0_4_1.cpp index da163b6493..08cd4a47b1 100644 --- a/examples/graph_deepspeech_v0_4_1.cpp +++ b/examples/graph_deepspeech_v0_4_1.cpp @@ -23,6 +23,7 @@ */ #include "arm_compute/graph.h" #include "arm_compute/graph/Types.h" + #include "support/ToolchainSupport.h" #include "utils/CommonGraphOptions.h" #include "utils/GraphUtils.h" @@ -37,8 +38,7 @@ using namespace arm_compute::graph_utils; class GraphDeepSpeechExample : public Example { public: - GraphDeepSpeechExample() - : cmd_parser(), common_opts(cmd_parser), common_params(), graph(0, "DeepSpeech v0.4.1") + GraphDeepSpeechExample() : cmd_parser(), common_opts(cmd_parser), common_params(), graph(0, "DeepSpeech v0.4.1") { } bool do_setup(int argc, char **argv) override @@ -51,7 +51,7 @@ public: common_params = consume_common_graph_parameters(common_opts); // Return when help menu is requested - if(common_params.help) + if (common_params.help) { cmd_parser.print_help(argv[0]); return false; @@ -64,7 +64,7 @@ public: std::string data_path = common_params.data_path; const std::string model_path = "/cnn_data/deepspeech_model/"; - if(!data_path.empty()) + if (!data_path.empty()) { data_path += model_path; } @@ -77,131 +77,131 @@ public: const float cell_clip = 20.f; // Create input descriptor - const TensorShape tensor_shape = permute_shape(TensorShape(26U, 19U, n_steps, 1U), DataLayout::NHWC, common_params.data_layout); - TensorDescriptor input_descriptor = TensorDescriptor(tensor_shape, common_params.data_type).set_layout(common_params.data_layout); + const TensorShape tensor_shape = + permute_shape(TensorShape(26U, 19U, n_steps, 1U), DataLayout::NHWC, common_params.data_layout); + TensorDescriptor input_descriptor = + TensorDescriptor(tensor_shape, common_params.data_type).set_layout(common_params.data_layout); // Set weights trained layout const DataLayout weights_layout = DataLayout::NHWC; - graph << common_params.target - << common_params.fast_math_hint + graph << common_params.target << common_params.fast_math_hint << InputLayer(input_descriptor, - get_weights_accessor(data_path, "input_values_x" + std::to_string(n_steps) + ".npy", weights_layout)) - .set_name("input_node"); + get_weights_accessor(data_path, "input_values_x" + std::to_string(n_steps) + ".npy", + weights_layout)) + .set_name("input_node"); - if(common_params.data_layout == DataLayout::NCHW) + if (common_params.data_layout == DataLayout::NCHW) { graph << PermuteLayer(PermutationVector(2U, 0U, 1U), common_params.data_layout).set_name("permute_to_nhwc"); } graph << ReshapeLayer(TensorShape(494U, n_steps)).set_name("Reshape_input") // Layer 1 - << FullyConnectedLayer( - 2048U, - get_weights_accessor(data_path, "h1_transpose.npy", weights_layout), - get_weights_accessor(data_path, "MatMul_bias.npy")) - .set_name("fc0") + << FullyConnectedLayer(2048U, get_weights_accessor(data_path, "h1_transpose.npy", weights_layout), + get_weights_accessor(data_path, "MatMul_bias.npy")) + .set_name("fc0") << ActivationLayer(ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::BOUNDED_RELU, cell_clip)) - .set_name("Relu") + .set_name("Relu") // Layer 2 - << FullyConnectedLayer( - 2048U, - get_weights_accessor(data_path, "h2_transpose.npy", weights_layout), - get_weights_accessor(data_path, "MatMul_1_bias.npy")) - .set_name("fc1") + << FullyConnectedLayer(2048U, get_weights_accessor(data_path, "h2_transpose.npy", weights_layout), + get_weights_accessor(data_path, "MatMul_1_bias.npy")) + .set_name("fc1") << ActivationLayer(ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::BOUNDED_RELU, cell_clip)) - .set_name("Relu_1") + .set_name("Relu_1") // Layer 3 - << FullyConnectedLayer( - 2048U, - get_weights_accessor(data_path, "h3_transpose.npy", weights_layout), - get_weights_accessor(data_path, "MatMul_2_bias.npy")) - .set_name("fc2") + << FullyConnectedLayer(2048U, get_weights_accessor(data_path, "h3_transpose.npy", weights_layout), + get_weights_accessor(data_path, "MatMul_2_bias.npy")) + .set_name("fc2") << ActivationLayer(ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::BOUNDED_RELU, cell_clip)) - .set_name("Relu_2") + .set_name("Relu_2") // Layer 4 << ReshapeLayer(TensorShape(2048U, 1U, n_steps)).set_name("Reshape_1"); // Unstack Layer (using SplitLayerNode) - NodeParams unstack_params = { "unstack", graph.hints().target_hint }; - NodeID unstack_nid = GraphBuilder::add_split_node(graph.graph(), unstack_params, { graph.tail_node(), 0 }, n_steps, 2); + NodeParams unstack_params = {"unstack", graph.hints().target_hint}; + NodeID unstack_nid = + GraphBuilder::add_split_node(graph.graph(), unstack_params, {graph.tail_node(), 0}, n_steps, 2); // Create input state descriptor - TensorDescriptor state_descriptor = TensorDescriptor(TensorShape(2048U), common_params.data_type).set_layout(common_params.data_layout); - SubStream previous_state(graph); - SubStream add_y(graph); + TensorDescriptor state_descriptor = + TensorDescriptor(TensorShape(2048U), common_params.data_type).set_layout(common_params.data_layout); + SubStream previous_state(graph); + SubStream add_y(graph); // Initial state for LSTM is all zeroes for both state_h and state_c, therefore only one input is created - previous_state << InputLayer(state_descriptor, - get_weights_accessor(data_path, "zeros.npy")) - .set_name("previous_state_c_h"); - add_y << InputLayer(state_descriptor, - get_weights_accessor(data_path, "ones.npy")) - .set_name("add_y"); + previous_state << InputLayer(state_descriptor, get_weights_accessor(data_path, "zeros.npy")) + .set_name("previous_state_c_h"); + add_y << InputLayer(state_descriptor, get_weights_accessor(data_path, "ones.npy")).set_name("add_y"); // Create LSTM Fully Connected weights and bias descriptors - TensorDescriptor lstm_weights_descriptor = TensorDescriptor(TensorShape(4096U, 8192U), common_params.data_type).set_layout(common_params.data_layout); - TensorDescriptor lstm_bias_descriptor = TensorDescriptor(TensorShape(8192U), common_params.data_type).set_layout(common_params.data_layout); - SubStream lstm_fc_weights(graph); - SubStream lstm_fc_bias(graph); - lstm_fc_weights << ConstantLayer(lstm_weights_descriptor, - get_weights_accessor(data_path, "rnn_lstm_cell_kernel_transpose.npy", weights_layout)) - .set_name("h5/transpose"); + TensorDescriptor lstm_weights_descriptor = + TensorDescriptor(TensorShape(4096U, 8192U), common_params.data_type).set_layout(common_params.data_layout); + TensorDescriptor lstm_bias_descriptor = + TensorDescriptor(TensorShape(8192U), common_params.data_type).set_layout(common_params.data_layout); + SubStream lstm_fc_weights(graph); + SubStream lstm_fc_bias(graph); + lstm_fc_weights << ConstantLayer( + lstm_weights_descriptor, + get_weights_accessor(data_path, "rnn_lstm_cell_kernel_transpose.npy", weights_layout)) + .set_name("h5/transpose"); lstm_fc_bias << ConstantLayer(lstm_bias_descriptor, get_weights_accessor(data_path, "rnn_lstm_cell_MatMul_bias.npy")) - .set_name("MatMul_3_bias"); + .set_name("MatMul_3_bias"); // LSTM Block - std::pair<SubStream, SubStream> new_state_1 = add_lstm_cell(unstack_nid, 0, previous_state, previous_state, add_y, lstm_fc_weights, lstm_fc_bias); - std::pair<SubStream, SubStream> new_state_2 = add_lstm_cell(unstack_nid, 1, new_state_1.first, new_state_1.second, add_y, lstm_fc_weights, lstm_fc_bias); - std::pair<SubStream, SubStream> new_state_3 = add_lstm_cell(unstack_nid, 2, new_state_2.first, new_state_2.second, add_y, lstm_fc_weights, lstm_fc_bias); - std::pair<SubStream, SubStream> new_state_4 = add_lstm_cell(unstack_nid, 3, new_state_3.first, new_state_3.second, add_y, lstm_fc_weights, lstm_fc_bias); - std::pair<SubStream, SubStream> new_state_5 = add_lstm_cell(unstack_nid, 4, new_state_4.first, new_state_4.second, add_y, lstm_fc_weights, lstm_fc_bias); - std::pair<SubStream, SubStream> new_state_6 = add_lstm_cell(unstack_nid, 5, new_state_5.first, new_state_5.second, add_y, lstm_fc_weights, lstm_fc_bias); - std::pair<SubStream, SubStream> new_state_7 = add_lstm_cell(unstack_nid, 6, new_state_6.first, new_state_6.second, add_y, lstm_fc_weights, lstm_fc_bias); - std::pair<SubStream, SubStream> new_state_8 = add_lstm_cell(unstack_nid, 7, new_state_7.first, new_state_7.second, add_y, lstm_fc_weights, lstm_fc_bias); - std::pair<SubStream, SubStream> new_state_9 = add_lstm_cell(unstack_nid, 8, new_state_8.first, new_state_8.second, add_y, lstm_fc_weights, lstm_fc_bias); - std::pair<SubStream, SubStream> new_state_10 = add_lstm_cell(unstack_nid, 9, new_state_9.first, new_state_9.second, add_y, lstm_fc_weights, lstm_fc_bias); - std::pair<SubStream, SubStream> new_state_11 = add_lstm_cell(unstack_nid, 10, new_state_10.first, new_state_10.second, add_y, lstm_fc_weights, lstm_fc_bias); - std::pair<SubStream, SubStream> new_state_12 = add_lstm_cell(unstack_nid, 11, new_state_11.first, new_state_11.second, add_y, lstm_fc_weights, lstm_fc_bias); - std::pair<SubStream, SubStream> new_state_13 = add_lstm_cell(unstack_nid, 12, new_state_12.first, new_state_12.second, add_y, lstm_fc_weights, lstm_fc_bias); - std::pair<SubStream, SubStream> new_state_14 = add_lstm_cell(unstack_nid, 13, new_state_13.first, new_state_13.second, add_y, lstm_fc_weights, lstm_fc_bias); - std::pair<SubStream, SubStream> new_state_15 = add_lstm_cell(unstack_nid, 14, new_state_14.first, new_state_14.second, add_y, lstm_fc_weights, lstm_fc_bias); - std::pair<SubStream, SubStream> new_state_16 = add_lstm_cell(unstack_nid, 15, new_state_15.first, new_state_15.second, add_y, lstm_fc_weights, lstm_fc_bias); + std::pair<SubStream, SubStream> new_state_1 = + add_lstm_cell(unstack_nid, 0, previous_state, previous_state, add_y, lstm_fc_weights, lstm_fc_bias); + std::pair<SubStream, SubStream> new_state_2 = + add_lstm_cell(unstack_nid, 1, new_state_1.first, new_state_1.second, add_y, lstm_fc_weights, lstm_fc_bias); + std::pair<SubStream, SubStream> new_state_3 = + add_lstm_cell(unstack_nid, 2, new_state_2.first, new_state_2.second, add_y, lstm_fc_weights, lstm_fc_bias); + std::pair<SubStream, SubStream> new_state_4 = + add_lstm_cell(unstack_nid, 3, new_state_3.first, new_state_3.second, add_y, lstm_fc_weights, lstm_fc_bias); + std::pair<SubStream, SubStream> new_state_5 = + add_lstm_cell(unstack_nid, 4, new_state_4.first, new_state_4.second, add_y, lstm_fc_weights, lstm_fc_bias); + std::pair<SubStream, SubStream> new_state_6 = + add_lstm_cell(unstack_nid, 5, new_state_5.first, new_state_5.second, add_y, lstm_fc_weights, lstm_fc_bias); + std::pair<SubStream, SubStream> new_state_7 = + add_lstm_cell(unstack_nid, 6, new_state_6.first, new_state_6.second, add_y, lstm_fc_weights, lstm_fc_bias); + std::pair<SubStream, SubStream> new_state_8 = + add_lstm_cell(unstack_nid, 7, new_state_7.first, new_state_7.second, add_y, lstm_fc_weights, lstm_fc_bias); + std::pair<SubStream, SubStream> new_state_9 = + add_lstm_cell(unstack_nid, 8, new_state_8.first, new_state_8.second, add_y, lstm_fc_weights, lstm_fc_bias); + std::pair<SubStream, SubStream> new_state_10 = + add_lstm_cell(unstack_nid, 9, new_state_9.first, new_state_9.second, add_y, lstm_fc_weights, lstm_fc_bias); + std::pair<SubStream, SubStream> new_state_11 = add_lstm_cell( + unstack_nid, 10, new_state_10.first, new_state_10.second, add_y, lstm_fc_weights, lstm_fc_bias); + std::pair<SubStream, SubStream> new_state_12 = add_lstm_cell( + unstack_nid, 11, new_state_11.first, new_state_11.second, add_y, lstm_fc_weights, lstm_fc_bias); + std::pair<SubStream, SubStream> new_state_13 = add_lstm_cell( + unstack_nid, 12, new_state_12.first, new_state_12.second, add_y, lstm_fc_weights, lstm_fc_bias); + std::pair<SubStream, SubStream> new_state_14 = add_lstm_cell( + unstack_nid, 13, new_state_13.first, new_state_13.second, add_y, lstm_fc_weights, lstm_fc_bias); + std::pair<SubStream, SubStream> new_state_15 = add_lstm_cell( + unstack_nid, 14, new_state_14.first, new_state_14.second, add_y, lstm_fc_weights, lstm_fc_bias); + std::pair<SubStream, SubStream> new_state_16 = add_lstm_cell( + unstack_nid, 15, new_state_15.first, new_state_15.second, add_y, lstm_fc_weights, lstm_fc_bias); // Concatenate new states on height const int axis = 1; - graph << StackLayer(axis, - std::move(new_state_1.second), - std::move(new_state_2.second), - std::move(new_state_3.second), - std::move(new_state_4.second), - std::move(new_state_5.second), - std::move(new_state_6.second), - std::move(new_state_7.second), - std::move(new_state_8.second), - std::move(new_state_9.second), - std::move(new_state_10.second), - std::move(new_state_11.second), - std::move(new_state_12.second), - std::move(new_state_13.second), - std::move(new_state_14.second), - std::move(new_state_15.second), - std::move(new_state_16.second)) - .set_name("concat"); - - graph << FullyConnectedLayer( - 2048U, - get_weights_accessor(data_path, "h5_transpose.npy", weights_layout), - get_weights_accessor(data_path, "MatMul_3_bias.npy")) - .set_name("fc3") + graph << StackLayer(axis, std::move(new_state_1.second), std::move(new_state_2.second), + std::move(new_state_3.second), std::move(new_state_4.second), std::move(new_state_5.second), + std::move(new_state_6.second), std::move(new_state_7.second), std::move(new_state_8.second), + std::move(new_state_9.second), std::move(new_state_10.second), + std::move(new_state_11.second), std::move(new_state_12.second), + std::move(new_state_13.second), std::move(new_state_14.second), + std::move(new_state_15.second), std::move(new_state_16.second)) + .set_name("concat"); + + graph << FullyConnectedLayer(2048U, get_weights_accessor(data_path, "h5_transpose.npy", weights_layout), + get_weights_accessor(data_path, "MatMul_3_bias.npy")) + .set_name("fc3") << ActivationLayer(ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::BOUNDED_RELU, cell_clip)) - .set_name("Relu3") - << FullyConnectedLayer( - 29U, - get_weights_accessor(data_path, "h6_transpose.npy", weights_layout), - get_weights_accessor(data_path, "MatMul_4_bias.npy")) - .set_name("fc3") + .set_name("Relu3") + << FullyConnectedLayer(29U, get_weights_accessor(data_path, "h6_transpose.npy", weights_layout), + get_weights_accessor(data_path, "MatMul_4_bias.npy")) + .set_name("fc3") << SoftmaxLayer().set_name("logits"); graph << OutputLayer(get_output_accessor(common_params, 5)); @@ -241,7 +241,7 @@ private: return Status{}; } - std::pair<SubStream, SubStream> add_lstm_cell(NodeID unstack_nid, + std::pair<SubStream, SubStream> add_lstm_cell(NodeID unstack_nid, unsigned int unstack_idx, SubStream previous_state_c, SubStream previous_state_h, @@ -250,41 +250,41 @@ private: SubStream lstm_fc_bias) { const std::string cell_name("rnn/lstm_cell_" + std::to_string(unstack_idx)); - const DataLayoutDimension concat_dim = (common_params.data_layout == DataLayout::NHWC) ? DataLayoutDimension::CHANNEL : DataLayoutDimension::WIDTH; + const DataLayoutDimension concat_dim = + (common_params.data_layout == DataLayout::NHWC) ? DataLayoutDimension::CHANNEL : DataLayoutDimension::WIDTH; // Concatenate result of Unstack with previous_state_h - NodeParams concat_params = { cell_name + "/concat", graph.hints().target_hint }; + NodeParams concat_params = {cell_name + "/concat", graph.hints().target_hint}; NodeID concat_nid = graph.graph().add_node<ConcatenateLayerNode>(2, concat_dim); graph.graph().add_connection(unstack_nid, unstack_idx, concat_nid, 0); graph.graph().add_connection(previous_state_h.tail_node(), 0, concat_nid, 1); set_node_params(graph.graph(), concat_nid, concat_params); graph.forward_tail(concat_nid); - graph << FullyConnectedLayer( - 8192U, - lstm_fc_weights, - lstm_fc_bias) - .set_name(cell_name + "/BiasAdd"); + graph << FullyConnectedLayer(8192U, lstm_fc_weights, lstm_fc_bias).set_name(cell_name + "/BiasAdd"); // Split Layer const unsigned int num_splits = 4; const unsigned int split_axis = 0; - NodeParams split_params = { cell_name + "/split", graph.hints().target_hint }; - NodeID split_nid = GraphBuilder::add_split_node(graph.graph(), split_params, { graph.tail_node(), 0 }, num_splits, split_axis); + NodeParams split_params = {cell_name + "/split", graph.hints().target_hint}; + NodeID split_nid = + GraphBuilder::add_split_node(graph.graph(), split_params, {graph.tail_node(), 0}, num_splits, split_axis); - NodeParams sigmoid_1_params = { cell_name + "/Sigmoid_1", graph.hints().target_hint }; - NodeParams add_params = { cell_name + "/add", graph.hints().target_hint }; - NodeParams sigmoid_2_params = { cell_name + "/Sigmoid_2", graph.hints().target_hint }; - NodeParams tanh_params = { cell_name + "/Tanh", graph.hints().target_hint }; + NodeParams sigmoid_1_params = {cell_name + "/Sigmoid_1", graph.hints().target_hint}; + NodeParams add_params = {cell_name + "/add", graph.hints().target_hint}; + NodeParams sigmoid_2_params = {cell_name + "/Sigmoid_2", graph.hints().target_hint}; + NodeParams tanh_params = {cell_name + "/Tanh", graph.hints().target_hint}; // Sigmoid 1 (first split) - NodeID sigmoid_1_nid = graph.graph().add_node<ActivationLayerNode>(ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::LOGISTIC)); + NodeID sigmoid_1_nid = graph.graph().add_node<ActivationLayerNode>( + ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::LOGISTIC)); graph.graph().add_connection(split_nid, 0, sigmoid_1_nid, 0); set_node_params(graph.graph(), sigmoid_1_nid, sigmoid_1_params); // Tanh (second split) - NodeID tanh_nid = graph.graph().add_node<ActivationLayerNode>(ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::TANH, 1.f, 1.f)); + NodeID tanh_nid = graph.graph().add_node<ActivationLayerNode>( + ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::TANH, 1.f, 1.f)); graph.graph().add_connection(split_nid, 1, tanh_nid, 0); set_node_params(graph.graph(), tanh_nid, tanh_params); @@ -292,13 +292,15 @@ private: tanh_ss.forward_tail(tanh_nid); // Add (third split) - NodeID add_nid = graph.graph().add_node<EltwiseLayerNode>(descriptors::EltwiseLayerDescriptor{ EltwiseOperation::Add }); + NodeID add_nid = + graph.graph().add_node<EltwiseLayerNode>(descriptors::EltwiseLayerDescriptor{EltwiseOperation::Add}); graph.graph().add_connection(split_nid, 2, add_nid, 0); graph.graph().add_connection(add_y.tail_node(), 0, add_nid, 1); set_node_params(graph.graph(), add_nid, add_params); // Sigmoid 2 (fourth split) - NodeID sigmoid_2_nid = graph.graph().add_node<ActivationLayerNode>(ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::LOGISTIC)); + NodeID sigmoid_2_nid = graph.graph().add_node<ActivationLayerNode>( + ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::LOGISTIC)); graph.graph().add_connection(split_nid, 3, sigmoid_2_nid, 0); set_node_params(graph.graph(), sigmoid_2_nid, sigmoid_2_params); @@ -306,28 +308,28 @@ private: sigmoid_1_ss.forward_tail(sigmoid_1_nid); SubStream mul_1_ss(sigmoid_1_ss); mul_1_ss << EltwiseLayer(std::move(sigmoid_1_ss), std::move(tanh_ss), EltwiseOperation::Mul) - .set_name(cell_name + "/mul_1"); + .set_name(cell_name + "/mul_1"); SubStream tanh_1_ss_tmp(graph); tanh_1_ss_tmp.forward_tail(add_nid); tanh_1_ss_tmp << ActivationLayer(ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::LOGISTIC)) - .set_name(cell_name + "/Sigmoid"); + .set_name(cell_name + "/Sigmoid"); SubStream tanh_1_ss_tmp2(tanh_1_ss_tmp); tanh_1_ss_tmp2 << EltwiseLayer(std::move(tanh_1_ss_tmp), std::move(previous_state_c), EltwiseOperation::Mul) - .set_name(cell_name + "/mul"); + .set_name(cell_name + "/mul"); SubStream tanh_1_ss(tanh_1_ss_tmp2); tanh_1_ss << EltwiseLayer(std::move(tanh_1_ss_tmp2), std::move(mul_1_ss), EltwiseOperation::Add) - .set_name(cell_name + "/new_state_c"); + .set_name(cell_name + "/new_state_c"); SubStream new_state_c(tanh_1_ss); tanh_1_ss << ActivationLayer(ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::TANH, 1.f, 1.f)) - .set_name(cell_name + "/Tanh_1"); + .set_name(cell_name + "/Tanh_1"); SubStream sigmoid_2_ss(graph); sigmoid_2_ss.forward_tail(sigmoid_2_nid); graph << EltwiseLayer(std::move(sigmoid_2_ss), std::move(tanh_1_ss), EltwiseOperation::Mul) - .set_name(cell_name + "/new_state_h"); + .set_name(cell_name + "/new_state_h"); SubStream new_state_h(graph); return std::pair<SubStream, SubStream>(new_state_c, new_state_h); |