diff options
Diffstat (limited to 'src/runtime/NEON/functions/NELSTMLayerQuantized.cpp')
| -rw-r--r-- | src/runtime/NEON/functions/NELSTMLayerQuantized.cpp | 404 |
1 files changed, 286 insertions, 118 deletions
diff --git a/src/runtime/NEON/functions/NELSTMLayerQuantized.cpp b/src/runtime/NEON/functions/NELSTMLayerQuantized.cpp index 5c0f19a15c..41f9c3d700 100644 --- a/src/runtime/NEON/functions/NELSTMLayerQuantized.cpp +++ b/src/runtime/NEON/functions/NELSTMLayerQuantized.cpp @@ -24,17 +24,10 @@ #include "arm_compute/runtime/NEON/functions/NELSTMLayerQuantized.h" #include "arm_compute/core/Utils.h" -#include "arm_compute/core/Validate.h" #include "arm_compute/core/utils/quantization/AsymmHelpers.h" -#include "src/core/NEON/kernels/NEConvertQuantizedSignednessKernel.h" -#include "src/core/NEON/kernels/NEGEMMInterleave4x4Kernel.h" -#include "src/core/NEON/kernels/NEGEMMLowpMatrixMultiplyKernel.h" -#include "src/core/NEON/kernels/NEGEMMLowpOffsetContributionKernel.h" -#include "src/core/NEON/kernels/NEGEMMLowpOffsetContributionOutputStageKernel.h" -#include "src/core/NEON/kernels/NEGEMMLowpReductionKernel.h" -#include "src/core/NEON/kernels/NEGEMMMatrixAdditionKernel.h" -#include "src/core/NEON/kernels/NEGEMMMatrixMultiplyKernel.h" -#include "src/core/NEON/kernels/NEGEMMTranspose1xWKernel.h" +#include "arm_compute/core/Validate.h" + +#include "src/common/utils/Log.h" #include "src/core/helpers/AutoConfiguration.h" #include <cmath> @@ -54,32 +47,104 @@ const QuantizationInfo qsymm_0(1.f / 32768.f, 0); // qsymm16 with 0 integer bit NELSTMLayerQuantized::~NELSTMLayerQuantized() = default; NELSTMLayerQuantized::NELSTMLayerQuantized(std::shared_ptr<IMemoryManager> memory_manager) - : _memory_group(std::move(memory_manager)), _gemmlowp(), _output_stage(), _transpose_weights(), _concat_input_weights(), _concat_recurrent_weights(), _concat_weights(), _concat_inputs(), - _concat_bias(), _sigmoid_forget_gate(), _sigmoid_input_gate(), _sigmoid_output_gate(), _tanh_modulation_gate(), _tanh_output_state(), _add1(), _add2(), _mul1(), _mul2(), _mul3(), - _slice_input_tensor(), _slice_forget_tensor(), _slice_cell_tensor(), _slice_output_tensor(), _dequantize(), _quantize(), _input_to_input_weights(nullptr), _input_to_forget_weights(nullptr), - _input_to_cell_weights(nullptr), _input_to_output_weights(nullptr), _recurrent_to_input_weights(nullptr), _recurrent_to_forget_weights(nullptr), _recurrent_to_cell_weights(nullptr), - _recurrent_to_output_weights(nullptr), _input_gate_bias(nullptr), _forget_gate_bias(nullptr), _cell_bias(nullptr), _output_gate_bias(nullptr), _recurrent_weights(), _input_weights(), _weights(), - _input(), _weights_transposed(), _output_highp(), _output_lowp(), _bias(), _forget_gate_input(), _input_gate_input(), _output_gate_input(), _input_modulation_gate_input(), _forget_gate_output(), - _input_gate_output(), _output_gate_output(), _input_modulation_gate_output(), _cell_state1(), _cell_state2(), _output_state_tmp(), _output_state_out_symm(), _output_state_out_f32(), + : _memory_group(std::move(memory_manager)), + _gemmlowp(), + _output_stage(), + _transpose_weights(), + _concat_input_weights(), + _concat_recurrent_weights(), + _concat_weights(), + _concat_inputs(), + _concat_bias(), + _sigmoid_forget_gate(), + _sigmoid_input_gate(), + _sigmoid_output_gate(), + _tanh_modulation_gate(), + _tanh_output_state(), + _add1(), + _add2(), + _mul1(), + _mul2(), + _mul3(), + _slice_input_tensor(), + _slice_forget_tensor(), + _slice_cell_tensor(), + _slice_output_tensor(), + _dequantize(), + _quantize(), + _input_to_input_weights(nullptr), + _input_to_forget_weights(nullptr), + _input_to_cell_weights(nullptr), + _input_to_output_weights(nullptr), + _recurrent_to_input_weights(nullptr), + _recurrent_to_forget_weights(nullptr), + _recurrent_to_cell_weights(nullptr), + _recurrent_to_output_weights(nullptr), + _input_gate_bias(nullptr), + _forget_gate_bias(nullptr), + _cell_bias(nullptr), + _output_gate_bias(nullptr), + _recurrent_weights(), + _input_weights(), + _weights(), + _input(), + _weights_transposed(), + _output_highp(), + _output_lowp(), + _bias(), + _forget_gate_input(), + _input_gate_input(), + _output_gate_input(), + _input_modulation_gate_input(), + _forget_gate_output(), + _input_gate_output(), + _output_gate_output(), + _input_modulation_gate_output(), + _cell_state1(), + _cell_state2(), + _output_state_tmp(), + _output_state_out_symm(), + _output_state_out_f32(), _is_prepared(false) { } void NELSTMLayerQuantized::configure(const ITensor *input, - const ITensor *input_to_input_weights, const ITensor *input_to_forget_weights, const ITensor *input_to_cell_weights, const ITensor *input_to_output_weights, - const ITensor *recurrent_to_input_weights, const ITensor *recurrent_to_forget_weights, const ITensor *recurrent_to_cell_weights, const ITensor *recurrent_to_output_weights, - const ITensor *input_gate_bias, const ITensor *forget_gate_bias, const ITensor *cell_bias, const ITensor *output_gate_bias, - ITensor *cell_state_in, const ITensor *output_state_in, - ITensor *cell_state_out, ITensor *output_state_out) + const ITensor *input_to_input_weights, + const ITensor *input_to_forget_weights, + const ITensor *input_to_cell_weights, + const ITensor *input_to_output_weights, + const ITensor *recurrent_to_input_weights, + const ITensor *recurrent_to_forget_weights, + const ITensor *recurrent_to_cell_weights, + const ITensor *recurrent_to_output_weights, + const ITensor *input_gate_bias, + const ITensor *forget_gate_bias, + const ITensor *cell_bias, + const ITensor *output_gate_bias, + ITensor *cell_state_in, + const ITensor *output_state_in, + ITensor *cell_state_out, + ITensor *output_state_out) { - ARM_COMPUTE_ERROR_ON_NULLPTR(input, input_to_input_weights, input_to_forget_weights, input_to_cell_weights, input_to_output_weights, - recurrent_to_input_weights, recurrent_to_forget_weights, recurrent_to_cell_weights, recurrent_to_output_weights, - input_gate_bias, forget_gate_bias, cell_bias, output_gate_bias, cell_state_in, output_state_in, cell_state_out, output_state_out); - - ARM_COMPUTE_ERROR_THROW_ON(NELSTMLayerQuantized::validate(input->info(), input_to_input_weights->info(), input_to_forget_weights->info(), input_to_cell_weights->info(), - input_to_output_weights->info(), - recurrent_to_input_weights->info(), recurrent_to_forget_weights->info(), recurrent_to_cell_weights->info(), recurrent_to_output_weights->info(), - input_gate_bias->info(), forget_gate_bias->info(), cell_bias->info(), output_gate_bias->info(), cell_state_in->info(), output_state_in->info(), cell_state_out->info(), output_state_out->info())); + ARM_COMPUTE_ERROR_ON_NULLPTR(input, input_to_input_weights, input_to_forget_weights, input_to_cell_weights, + input_to_output_weights, recurrent_to_input_weights, recurrent_to_forget_weights, + recurrent_to_cell_weights, recurrent_to_output_weights, input_gate_bias, + forget_gate_bias, cell_bias, output_gate_bias, cell_state_in, output_state_in, + cell_state_out, output_state_out); + + ARM_COMPUTE_ERROR_THROW_ON(NELSTMLayerQuantized::validate( + input->info(), input_to_input_weights->info(), input_to_forget_weights->info(), input_to_cell_weights->info(), + input_to_output_weights->info(), recurrent_to_input_weights->info(), recurrent_to_forget_weights->info(), + recurrent_to_cell_weights->info(), recurrent_to_output_weights->info(), input_gate_bias->info(), + forget_gate_bias->info(), cell_bias->info(), output_gate_bias->info(), cell_state_in->info(), + output_state_in->info(), cell_state_out->info(), output_state_out->info())); + + ARM_COMPUTE_LOG_PARAMS(input, input_to_input_weights, input_to_forget_weights, input_to_cell_weights, + input_to_output_weights, recurrent_to_input_weights, recurrent_to_forget_weights, + recurrent_to_cell_weights, recurrent_to_output_weights, input_gate_bias, forget_gate_bias, + cell_bias, output_gate_bias, cell_state_in, output_state_in, cell_state_out, + output_state_out); const int input_size = input->info()->dimension(0); const int batch_size = input->info()->dimension(1); @@ -87,8 +152,10 @@ void NELSTMLayerQuantized::configure(const ITensor *input, const QuantizationInfo qweights = input_to_input_weights->info()->quantization_info(); // Weights quantization - auto_init_if_empty(*cell_state_out->info(), TensorInfo(TensorShape(batch_size, output_size), 1, DataType::QSYMM16, qsymm_4)); - auto_init_if_empty(*output_state_out->info(), TensorInfo(TensorShape(batch_size, output_size), 1, DataType::QASYMM8, qasymm)); + auto_init_if_empty(*cell_state_out->info(), + TensorInfo(TensorShape(batch_size, output_size), 1, DataType::QSYMM16, qsymm_4)); + auto_init_if_empty(*output_state_out->info(), + TensorInfo(TensorShape(batch_size, output_size), 1, DataType::QASYMM8, qasymm)); _input_to_input_weights = input_to_input_weights; _input_to_forget_weights = input_to_forget_weights; @@ -104,34 +171,41 @@ void NELSTMLayerQuantized::configure(const ITensor *input, _output_gate_bias = output_gate_bias; // Weights concatenation - std::vector<const ITensor *> inputs_weights_vector{ input_to_input_weights, input_to_forget_weights, input_to_cell_weights, input_to_output_weights }; - std::vector<const ITensor *> recurrent_weights_vector{ recurrent_to_input_weights, recurrent_to_forget_weights, recurrent_to_cell_weights, recurrent_to_output_weights }; + std::vector<const ITensor *> inputs_weights_vector{input_to_input_weights, input_to_forget_weights, + input_to_cell_weights, input_to_output_weights}; + std::vector<const ITensor *> recurrent_weights_vector{recurrent_to_input_weights, recurrent_to_forget_weights, + recurrent_to_cell_weights, recurrent_to_output_weights}; - _input_weights.allocator()->init(TensorInfo(TensorShape(input_size, 4 * output_size), 1, DataType::QASYMM8, qweights)); + _input_weights.allocator()->init( + TensorInfo(TensorShape(input_size, 4 * output_size), 1, DataType::QASYMM8, qweights)); _concat_input_weights.configure(inputs_weights_vector, &_input_weights, Window::DimY); - _recurrent_weights.allocator()->init(TensorInfo(TensorShape(output_size, 4 * output_size), 1, DataType::QASYMM8, qweights)); + _recurrent_weights.allocator()->init( + TensorInfo(TensorShape(output_size, 4 * output_size), 1, DataType::QASYMM8, qweights)); _concat_recurrent_weights.configure(recurrent_weights_vector, &_recurrent_weights, Window::DimY); - std::vector<const ITensor *> weights_vector{ &_recurrent_weights, &_input_weights }; - _weights.allocator()->init(TensorInfo(TensorShape(output_size + input_size, 4 * output_size), 1, DataType::QASYMM8, qweights)); + std::vector<const ITensor *> weights_vector{&_recurrent_weights, &_input_weights}; + _weights.allocator()->init( + TensorInfo(TensorShape(output_size + input_size, 4 * output_size), 1, DataType::QASYMM8, qweights)); _concat_weights.configure(weights_vector, &_weights, Window::DimX); _transpose_weights.configure(&_weights, &_weights_transposed); // Input concatenation - std::vector<const ITensor *> input_vector{ input, output_state_in }; + std::vector<const ITensor *> input_vector{input, output_state_in}; _memory_group.manage(&_input); - _input.allocator()->init(TensorInfo(TensorShape(output_size + input_size, batch_size), 1, DataType::QASYMM8, qasymm)); + _input.allocator()->init( + TensorInfo(TensorShape(output_size + input_size, batch_size), 1, DataType::QASYMM8, qasymm)); _concat_inputs.configure(input_vector, &_input, Window::DimX); // Bias concatenation - std::vector<const ITensor *> bias_vector{ input_gate_bias, forget_gate_bias, cell_bias, output_gate_bias }; + std::vector<const ITensor *> bias_vector{input_gate_bias, forget_gate_bias, cell_bias, output_gate_bias}; _bias.allocator()->init(TensorInfo(TensorShape(4 * output_size), 1, DataType::S32)); _concat_bias.configure(bias_vector, &_bias, Window::DimX); // Invert the offset for gemmlowp _input.info()->set_quantization_info(QuantizationInfo(qasymm.uniform().scale, -qasymm.uniform().offset)); - _weights_transposed.info()->set_quantization_info(QuantizationInfo(qweights.uniform().scale, -qweights.uniform().offset)); + _weights_transposed.info()->set_quantization_info( + QuantizationInfo(qweights.uniform().scale, -qweights.uniform().offset)); // Run gemmlowp _memory_group.manage(&_output_highp); @@ -141,7 +215,8 @@ void NELSTMLayerQuantized::configure(const ITensor *input, // Set the offset back _input.info()->set_quantization_info(QuantizationInfo(qasymm.uniform().scale, qasymm.uniform().offset)); - _weights_transposed.info()->set_quantization_info(QuantizationInfo(qweights.uniform().scale, qweights.uniform().offset)); + _weights_transposed.info()->set_quantization_info( + QuantizationInfo(qweights.uniform().scale, qweights.uniform().offset)); // multiplier = (input_scale * weights_scale) / output_scale (2 ^ (-12)) _output_lowp.allocator()->init(TensorInfo(_output_highp.info()->tensor_shape(), 1, DataType::QSYMM16, qsymm_3)); @@ -152,69 +227,91 @@ void NELSTMLayerQuantized::configure(const ITensor *input, quantization::calculate_quantized_multiplier(multiplier, &output_multiplier, &output_shift); _memory_group.manage(&_output_lowp); - _output_stage.configure(&_output_highp, &_bias, &_output_lowp, output_multiplier, output_shift); + + GEMMLowpOutputStageInfo info; + info.type = GEMMLowpOutputStageType::QUANTIZE_DOWN_FIXEDPOINT; + info.gemmlowp_multiplier = output_multiplier; + info.gemmlowp_shift = output_shift; + info.output_data_type = DataType::QSYMM16; + _output_stage.configure(&_output_highp, &_bias, &_output_lowp, info); _output_highp.allocator()->allocate(); _bias.allocator()->allocate(); // Get the gate tensors - if(batch_size > 1) + if (batch_size > 1) { _memory_group.manage(&_input_gate_input); - _slice_input_tensor.configure(&_output_lowp, &_input_gate_input, { 0, 0 }, { output_size, batch_size }); + _slice_input_tensor.configure(&_output_lowp, &_input_gate_input, {0, 0}, {output_size, batch_size}); _memory_group.manage(&_forget_gate_input); - _slice_forget_tensor.configure(&_output_lowp, &_forget_gate_input, { output_size, 0 }, { 2 * output_size, batch_size }); + _slice_forget_tensor.configure(&_output_lowp, &_forget_gate_input, {output_size, 0}, + {2 * output_size, batch_size}); _memory_group.manage(&_input_modulation_gate_input); - _slice_cell_tensor.configure(&_output_lowp, &_input_modulation_gate_input, { 2 * output_size, 0 }, { 3 * output_size, batch_size }); + _slice_cell_tensor.configure(&_output_lowp, &_input_modulation_gate_input, {2 * output_size, 0}, + {3 * output_size, batch_size}); _memory_group.manage(&_output_gate_input); - _slice_output_tensor.configure(&_output_lowp, &_output_gate_input, { 3 * output_size, 0 }, { 4 * output_size, batch_size }); + _slice_output_tensor.configure(&_output_lowp, &_output_gate_input, {3 * output_size, 0}, + {4 * output_size, batch_size}); _output_lowp.allocator()->allocate(); } else { _memory_group.manage(&_input_gate_input); - _slice_input_tensor.configure(&_output_lowp, &_input_gate_input, { 0 }, { output_size }); + _slice_input_tensor.configure(&_output_lowp, &_input_gate_input, {0}, {output_size}); _memory_group.manage(&_forget_gate_input); - _slice_forget_tensor.configure(&_output_lowp, &_forget_gate_input, { output_size }, { 2 * output_size }); + _slice_forget_tensor.configure(&_output_lowp, &_forget_gate_input, {output_size}, {2 * output_size}); _memory_group.manage(&_input_modulation_gate_input); - _slice_cell_tensor.configure(&_output_lowp, &_input_modulation_gate_input, { 2 * output_size }, { 3 * output_size }); + _slice_cell_tensor.configure(&_output_lowp, &_input_modulation_gate_input, {2 * output_size}, + {3 * output_size}); _memory_group.manage(&_output_gate_input); - _slice_output_tensor.configure(&_output_lowp, &_output_gate_input, { 3 * output_size }, { 4 * output_size }); + _slice_output_tensor.configure(&_output_lowp, &_output_gate_input, {3 * output_size}, {4 * output_size}); _output_lowp.allocator()->allocate(); } // Forget gate _memory_group.manage(&_forget_gate_output); - _forget_gate_output.allocator()->init(TensorInfo(_forget_gate_input.info()->tensor_shape(), 1, DataType::QSYMM16, qsymm_0)); - _sigmoid_forget_gate.configure(&_forget_gate_input, &_forget_gate_output, ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::LOGISTIC)); + _forget_gate_output.allocator()->init( + TensorInfo(_forget_gate_input.info()->tensor_shape(), 1, DataType::QSYMM16, qsymm_0)); + _sigmoid_forget_gate.configure(&_forget_gate_input, &_forget_gate_output, + ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::LOGISTIC)); _forget_gate_input.allocator()->allocate(); // Input gate _memory_group.manage(&_input_gate_output); - _input_gate_output.allocator()->init(TensorInfo(_input_gate_input.info()->tensor_shape(), 1, DataType::QSYMM16, qsymm_0)); - _sigmoid_input_gate.configure(&_input_gate_input, &_input_gate_output, ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::LOGISTIC)); + _input_gate_output.allocator()->init( + TensorInfo(_input_gate_input.info()->tensor_shape(), 1, DataType::QSYMM16, qsymm_0)); + _sigmoid_input_gate.configure(&_input_gate_input, &_input_gate_output, + ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::LOGISTIC)); _input_gate_input.allocator()->allocate(); // Input modulation gate equation _memory_group.manage(&_input_modulation_gate_output); - _input_modulation_gate_output.allocator()->init(TensorInfo(_input_modulation_gate_input.info()->tensor_shape(), 1, DataType::QSYMM16, qsymm_0)); - _tanh_modulation_gate.configure(&_input_modulation_gate_input, &_input_modulation_gate_output, ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::TANH, 1.0f, 1.0f)); + _input_modulation_gate_output.allocator()->init( + TensorInfo(_input_modulation_gate_input.info()->tensor_shape(), 1, DataType::QSYMM16, qsymm_0)); + _tanh_modulation_gate.configure(&_input_modulation_gate_input, &_input_modulation_gate_output, + ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::TANH, 1.0f, 1.0f)); _input_modulation_gate_input.allocator()->allocate(); // Output gate _memory_group.manage(&_output_gate_output); - _output_gate_output.allocator()->init(TensorInfo(_output_gate_input.info()->tensor_shape(), 1, DataType::QSYMM16, qsymm_0)); - _sigmoid_output_gate.configure(&_output_gate_input, &_output_gate_output, ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::LOGISTIC)); + _output_gate_output.allocator()->init( + TensorInfo(_output_gate_input.info()->tensor_shape(), 1, DataType::QSYMM16, qsymm_0)); + _sigmoid_output_gate.configure(&_output_gate_input, &_output_gate_output, + ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::LOGISTIC)); _output_gate_input.allocator()->allocate(); // Long term memory _memory_group.manage(&_cell_state1); - _cell_state1.allocator()->init(TensorInfo(_forget_gate_output.info()->tensor_shape(), 1, DataType::QSYMM16, qsymm_4)); - _mul1.configure(&_forget_gate_output, cell_state_in, &_cell_state1, 1, ConvertPolicy::SATURATE, RoundingPolicy::TO_ZERO); + _cell_state1.allocator()->init( + TensorInfo(_forget_gate_output.info()->tensor_shape(), 1, DataType::QSYMM16, qsymm_4)); + _mul1.configure(&_forget_gate_output, cell_state_in, &_cell_state1, 1, ConvertPolicy::SATURATE, + RoundingPolicy::TO_ZERO); _forget_gate_output.allocator()->allocate(); _memory_group.manage(&_cell_state2); - _cell_state2.allocator()->init(TensorInfo(_input_gate_output.info()->tensor_shape(), 1, DataType::QSYMM16, qsymm_4)); - _mul2.configure(&_input_gate_output, &_input_modulation_gate_output, &_cell_state2, 1, ConvertPolicy::SATURATE, RoundingPolicy::TO_ZERO); + _cell_state2.allocator()->init( + TensorInfo(_input_gate_output.info()->tensor_shape(), 1, DataType::QSYMM16, qsymm_4)); + _mul2.configure(&_input_gate_output, &_input_modulation_gate_output, &_cell_state2, 1, ConvertPolicy::SATURATE, + RoundingPolicy::TO_ZERO); _input_modulation_gate_output.allocator()->allocate(); _input_gate_output.allocator()->allocate(); @@ -224,18 +321,23 @@ void NELSTMLayerQuantized::configure(const ITensor *input, // Short term memory _memory_group.manage(&_output_state_tmp); - _output_state_tmp.allocator()->init(TensorInfo(cell_state_out->info()->tensor_shape(), 1, DataType::QSYMM16, qsymm_0)); - _tanh_output_state.configure(cell_state_out, &_output_state_tmp, ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::TANH, 1.0f, 1.0f)); + _output_state_tmp.allocator()->init( + TensorInfo(cell_state_out->info()->tensor_shape(), 1, DataType::QSYMM16, qsymm_0)); + _tanh_output_state.configure(cell_state_out, &_output_state_tmp, + ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::TANH, 1.0f, 1.0f)); _memory_group.manage(&_output_state_out_symm); - _output_state_out_symm.allocator()->init(TensorInfo(_output_gate_output.info()->tensor_shape(), 1, DataType::QSYMM16, qsymm_0)); - _mul3.configure(&_output_state_tmp, &_output_gate_output, &_output_state_out_symm, 1, ConvertPolicy::SATURATE, RoundingPolicy::TO_ZERO); + _output_state_out_symm.allocator()->init( + TensorInfo(_output_gate_output.info()->tensor_shape(), 1, DataType::QSYMM16, qsymm_0)); + _mul3.configure(&_output_state_tmp, &_output_gate_output, &_output_state_out_symm, 1, ConvertPolicy::SATURATE, + RoundingPolicy::TO_ZERO); _output_gate_output.allocator()->allocate(); _output_state_tmp.allocator()->allocate(); // Requantize the output state from QSYMM16 to QASYMM8 _memory_group.manage(&_output_state_out_f32); - _output_state_out_f32.allocator()->init(TensorInfo(_output_state_out_symm.info()->tensor_shape(), 1, DataType::F32)); + _output_state_out_f32.allocator()->init( + TensorInfo(_output_state_out_symm.info()->tensor_shape(), 1, DataType::F32)); _dequantize.configure(&_output_state_out_symm, &_output_state_out_f32); _output_state_out_symm.allocator()->allocate(); @@ -244,15 +346,28 @@ void NELSTMLayerQuantized::configure(const ITensor *input, } Status NELSTMLayerQuantized::validate(const ITensorInfo *input, - const ITensorInfo *input_to_input_weights, const ITensorInfo *input_to_forget_weights, const ITensorInfo *input_to_cell_weights, const ITensorInfo *input_to_output_weights, - const ITensorInfo *recurrent_to_input_weights, const ITensorInfo *recurrent_to_forget_weights, const ITensorInfo *recurrent_to_cell_weights, const ITensorInfo *recurrent_to_output_weights, - const ITensorInfo *input_gate_bias, const ITensorInfo *forget_gate_bias, const ITensorInfo *cell_bias, const ITensorInfo *output_gate_bias, - const ITensorInfo *cell_state_in, const ITensorInfo *output_state_in, - const ITensorInfo *cell_state_out, const ITensorInfo *output_state_out) + const ITensorInfo *input_to_input_weights, + const ITensorInfo *input_to_forget_weights, + const ITensorInfo *input_to_cell_weights, + const ITensorInfo *input_to_output_weights, + const ITensorInfo *recurrent_to_input_weights, + const ITensorInfo *recurrent_to_forget_weights, + const ITensorInfo *recurrent_to_cell_weights, + const ITensorInfo *recurrent_to_output_weights, + const ITensorInfo *input_gate_bias, + const ITensorInfo *forget_gate_bias, + const ITensorInfo *cell_bias, + const ITensorInfo *output_gate_bias, + const ITensorInfo *cell_state_in, + const ITensorInfo *output_state_in, + const ITensorInfo *cell_state_out, + const ITensorInfo *output_state_out) { - ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(input, input_to_input_weights, input_to_forget_weights, input_to_cell_weights, input_to_output_weights, recurrent_to_input_weights, - recurrent_to_forget_weights, recurrent_to_cell_weights, recurrent_to_output_weights, input_gate_bias, forget_gate_bias, cell_bias, output_gate_bias, cell_state_in, - output_state_in, cell_state_out, output_state_out); + ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR( + input, input_to_input_weights, input_to_forget_weights, input_to_cell_weights, input_to_output_weights, + recurrent_to_input_weights, recurrent_to_forget_weights, recurrent_to_cell_weights, recurrent_to_output_weights, + input_gate_bias, forget_gate_bias, cell_bias, output_gate_bias, cell_state_in, output_state_in, cell_state_out, + output_state_out); const int input_size = input->dimension(0); const int batch_size = input->dimension(1); @@ -264,29 +379,51 @@ Status NELSTMLayerQuantized::validate(const ITensorInfo *input, ARM_COMPUTE_RETURN_ERROR_ON(input_gate_bias->num_dimensions() > 1); ARM_COMPUTE_RETURN_ERROR_ON(output_state_in->num_dimensions() > 2); - TensorInfo input_weights_info(input_to_input_weights->clone()->set_tensor_shape(TensorShape(input_size, output_size)).set_data_type(DataType::QASYMM8)); - TensorInfo recurrent_weights_info(input_to_input_weights->clone()->set_tensor_shape(TensorShape(output_size, output_size)).set_data_type(DataType::QASYMM8)); - TensorInfo bias_info(input_gate_bias->clone()->set_tensor_shape(TensorShape(output_size)).set_data_type(DataType::S32)); - TensorInfo output_state_info(cell_state_in->clone()->set_tensor_shape(TensorShape(output_size, batch_size)).set_data_type(DataType::QASYMM8).set_quantization_info(qasymm)); - TensorInfo cell_state_info(cell_state_in->clone()->set_tensor_shape(TensorShape(output_size, batch_size)).set_data_type(DataType::QSYMM16).set_quantization_info(qsymm_4)); + TensorInfo input_weights_info(input_to_input_weights->clone() + ->set_tensor_shape(TensorShape(input_size, output_size)) + .set_data_type(DataType::QASYMM8)); + TensorInfo recurrent_weights_info(input_to_input_weights->clone() + ->set_tensor_shape(TensorShape(output_size, output_size)) + .set_data_type(DataType::QASYMM8)); + TensorInfo bias_info( + input_gate_bias->clone()->set_tensor_shape(TensorShape(output_size)).set_data_type(DataType::S32)); + TensorInfo output_state_info(cell_state_in->clone() + ->set_tensor_shape(TensorShape(output_size, batch_size)) + .set_data_type(DataType::QASYMM8) + .set_quantization_info(qasymm)); + TensorInfo cell_state_info(cell_state_in->clone() + ->set_tensor_shape(TensorShape(output_size, batch_size)) + .set_data_type(DataType::QSYMM16) + .set_quantization_info(qsymm_4)); // Shape checks - ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_SHAPES(&input_weights_info, input_to_input_weights, input_to_forget_weights, input_to_cell_weights, input_to_output_weights); - ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_SHAPES(&recurrent_weights_info, recurrent_to_input_weights, recurrent_to_forget_weights, recurrent_to_cell_weights, recurrent_to_output_weights); - ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_SHAPES(&bias_info, input_gate_bias, forget_gate_bias, cell_bias, output_gate_bias); + ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_SHAPES(&input_weights_info, input_to_input_weights, input_to_forget_weights, + input_to_cell_weights, input_to_output_weights); + ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_SHAPES(&recurrent_weights_info, recurrent_to_input_weights, + recurrent_to_forget_weights, recurrent_to_cell_weights, + recurrent_to_output_weights); + ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_SHAPES(&bias_info, input_gate_bias, forget_gate_bias, cell_bias, + output_gate_bias); ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_SHAPES(&cell_state_info, cell_state_in); ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_SHAPES(&output_state_info, output_state_in); // Data type checks - ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(&input_weights_info, input, input_to_input_weights, input_to_forget_weights, input_to_cell_weights, input_to_output_weights); - ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(recurrent_to_input_weights, recurrent_to_forget_weights, recurrent_to_cell_weights, recurrent_to_output_weights); - ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(&bias_info, input_gate_bias, forget_gate_bias, cell_bias, output_gate_bias); + ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(&input_weights_info, input, input_to_input_weights, + input_to_forget_weights, input_to_cell_weights, + input_to_output_weights); + ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(recurrent_to_input_weights, recurrent_to_forget_weights, + recurrent_to_cell_weights, recurrent_to_output_weights); + ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(&bias_info, input_gate_bias, forget_gate_bias, cell_bias, + output_gate_bias); ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(&cell_state_info, cell_state_in); ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(&output_state_info, output_state_in); // Quantization checks - ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_QUANTIZATION_INFO(&input_weights_info, input_to_input_weights, input_to_forget_weights, input_to_cell_weights, input_to_output_weights); - ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_QUANTIZATION_INFO(recurrent_to_input_weights, recurrent_to_forget_weights, recurrent_to_cell_weights, recurrent_to_output_weights); + ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_QUANTIZATION_INFO(&input_weights_info, input_to_input_weights, + input_to_forget_weights, input_to_cell_weights, + input_to_output_weights); + ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_QUANTIZATION_INFO(recurrent_to_input_weights, recurrent_to_forget_weights, + recurrent_to_cell_weights, recurrent_to_output_weights); ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_QUANTIZATION_INFO(&cell_state_info, cell_state_in); ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_QUANTIZATION_INFO(&output_state_info, output_state_in); @@ -308,7 +445,8 @@ Status NELSTMLayerQuantized::validate(const ITensorInfo *input, recurrent_weights_vector.emplace_back(recurrent_to_cell_weights); recurrent_weights_vector.emplace_back(recurrent_to_output_weights); const TensorInfo recurrent_weights(TensorShape(output_size, 4 * output_size), 1, DataType::QASYMM8, qweights); - ARM_COMPUTE_RETURN_ON_ERROR(NEConcatenateLayer::validate(recurrent_weights_vector, &recurrent_weights, Window::DimY)); + ARM_COMPUTE_RETURN_ON_ERROR( + NEConcatenateLayer::validate(recurrent_weights_vector, &recurrent_weights, Window::DimY)); // _concat_weights std::vector<const ITensorInfo *> weights_vector; @@ -318,7 +456,7 @@ Status NELSTMLayerQuantized::validate(const ITensorInfo *input, ARM_COMPUTE_RETURN_ON_ERROR(NEConcatenateLayer::validate(weights_vector, &weights, Window::DimX)); // _transpose_weights const TensorShape weights_transposed_shape(weights.tensor_shape()[1], weights.tensor_shape()[0]); - TensorInfo weights_transposed = weights.clone()->set_is_resizable(true).set_tensor_shape(weights_transposed_shape); + TensorInfo weights_transposed = weights.clone()->set_is_resizable(true).set_tensor_shape(weights_transposed_shape); ARM_COMPUTE_RETURN_ON_ERROR(NETranspose::validate(&weights, &weights_transposed)); // _concat_inputs @@ -344,7 +482,8 @@ Status NELSTMLayerQuantized::validate(const ITensorInfo *input, // _gemmlowp const TensorInfo output_highp(TensorShape(4 * output_size, batch_size), 1, DataType::S32); - ARM_COMPUTE_RETURN_ON_ERROR(NEGEMMLowpMatrixMultiplyCore::validate(&input_concatenated, &weights_transposed, nullptr, &output_highp)); + ARM_COMPUTE_RETURN_ON_ERROR( + NEGEMMLowpMatrixMultiplyCore::validate(&input_concatenated, &weights_transposed, nullptr, &output_highp)); // Set the offset back input_concatenated.set_quantization_info(QuantizationInfo(qasymm.uniform().scale, qasymm.uniform().offset)); @@ -355,78 +494,107 @@ Status NELSTMLayerQuantized::validate(const ITensorInfo *input, const float multiplier = 4096.f * qasymm.uniform().scale * qweights.uniform().scale; int32_t output_multiplier = 0; int32_t output_shift = 0; - ARM_COMPUTE_RETURN_ON_ERROR(quantization::calculate_quantized_multiplier(multiplier, &output_multiplier, &output_shift)); + ARM_COMPUTE_RETURN_ON_ERROR( + quantization::calculate_quantized_multiplier(multiplier, &output_multiplier, &output_shift)); // _output_stage - ARM_COMPUTE_RETURN_ON_ERROR(NEGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPoint::validate(&output_highp, &bias_concatenated, &output_lowp)); + GEMMLowpOutputStageInfo info; + info.type = GEMMLowpOutputStageType::QUANTIZE_DOWN_FIXEDPOINT; + info.gemmlowp_multiplier = output_multiplier; + info.gemmlowp_shift = output_shift; + info.output_data_type = DataType::QSYMM16; + ARM_COMPUTE_RETURN_ON_ERROR(NEGEMMLowpOutputStage::validate(&output_highp, &bias_concatenated, &output_lowp, info)); TensorInfo input_gate_input; TensorInfo forget_gate_input; TensorInfo input_modulation_gate_input; TensorInfo output_gate_input; - if(batch_size > 1) + if (batch_size > 1) { // _slice_input_tensor input_gate_input = TensorInfo(TensorShape(output_size, batch_size), 1, DataType::QSYMM16, qsymm_3); - ARM_COMPUTE_RETURN_ON_ERROR(NESlice::validate(&output_lowp, &input_gate_input, { 0, 0 }, { output_size, batch_size })); + ARM_COMPUTE_RETURN_ON_ERROR( + NESlice::validate(&output_lowp, &input_gate_input, {0, 0}, {output_size, batch_size})); // _slice_forget_tensor forget_gate_input = TensorInfo(TensorShape(output_size, batch_size), 1, DataType::QSYMM16, qsymm_3); - ARM_COMPUTE_RETURN_ON_ERROR(NESlice::validate(&output_lowp, &forget_gate_input, { output_size, 0 }, { 2 * output_size, batch_size })); + ARM_COMPUTE_RETURN_ON_ERROR( + NESlice::validate(&output_lowp, &forget_gate_input, {output_size, 0}, {2 * output_size, batch_size})); // _slice_cell_tensor input_modulation_gate_input = TensorInfo(TensorShape(output_size, batch_size), 1, DataType::QSYMM16, qsymm_3); - ARM_COMPUTE_RETURN_ON_ERROR(NESlice::validate(&output_lowp, &input_modulation_gate_input, { 2 * output_size, 0 }, { 3 * output_size, batch_size })); + ARM_COMPUTE_RETURN_ON_ERROR(NESlice::validate(&output_lowp, &input_modulation_gate_input, {2 * output_size, 0}, + {3 * output_size, batch_size})); // _slice_output_tensor output_gate_input = TensorInfo(TensorShape(output_size, batch_size), 1, DataType::QSYMM16, qsymm_3); - ARM_COMPUTE_RETURN_ON_ERROR(NESlice::validate(&output_lowp, &output_gate_input, { 3 * output_size, 0 }, { 4 * output_size, batch_size })); + ARM_COMPUTE_RETURN_ON_ERROR( + NESlice::validate(&output_lowp, &output_gate_input, {3 * output_size, 0}, {4 * output_size, batch_size})); } else { // _slice_input_tensor input_gate_input = TensorInfo(TensorShape(output_size), 1, DataType::QSYMM16, qsymm_3); - ARM_COMPUTE_RETURN_ON_ERROR(NESlice::validate(&output_lowp, &input_gate_input, { 0 }, { output_size })); + ARM_COMPUTE_RETURN_ON_ERROR(NESlice::validate(&output_lowp, &input_gate_input, {0}, {output_size})); // _slice_forget_tensor forget_gate_input = TensorInfo(TensorShape(output_size), 1, DataType::QSYMM16, qsymm_3); - ARM_COMPUTE_RETURN_ON_ERROR(NESlice::validate(&output_lowp, &forget_gate_input, { output_size }, { 2 * output_size })); + ARM_COMPUTE_RETURN_ON_ERROR( + NESlice::validate(&output_lowp, &forget_gate_input, {output_size}, {2 * output_size})); // _slice_cell_tensor input_modulation_gate_input = TensorInfo(TensorShape(output_size), 1, DataType::QSYMM16, qsymm_3); - ARM_COMPUTE_RETURN_ON_ERROR(NESlice::validate(&output_lowp, &input_modulation_gate_input, { 2 * output_size }, { 3 * output_size })); + ARM_COMPUTE_RETURN_ON_ERROR( + NESlice::validate(&output_lowp, &input_modulation_gate_input, {2 * output_size}, {3 * output_size})); // _slice_output_tensor output_gate_input = TensorInfo(TensorShape(output_size), 1, DataType::QSYMM16, qsymm_3); - ARM_COMPUTE_RETURN_ON_ERROR(NESlice::validate(&output_lowp, &output_gate_input, { 3 * output_size }, { 4 * output_size })); + ARM_COMPUTE_RETURN_ON_ERROR( + NESlice::validate(&output_lowp, &output_gate_input, {3 * output_size}, {4 * output_size})); } // _sigmoid_forget_gate const TensorInfo forget_gate_output(forget_gate_input.tensor_shape(), 1, DataType::QSYMM16, qsymm_0); - ARM_COMPUTE_RETURN_ON_ERROR(NEActivationLayer::validate(&forget_gate_input, &forget_gate_output, ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::LOGISTIC))); + ARM_COMPUTE_RETURN_ON_ERROR( + NEActivationLayer::validate(&forget_gate_input, &forget_gate_output, + ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::LOGISTIC))); // _sigmoid_input_gate const TensorInfo input_gate_output(input_gate_input.tensor_shape(), 1, DataType::QSYMM16, qsymm_0); - ARM_COMPUTE_RETURN_ON_ERROR(NEActivationLayer::validate(&input_gate_input, &input_gate_output, ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::LOGISTIC))); + ARM_COMPUTE_RETURN_ON_ERROR(NEActivationLayer::validate( + &input_gate_input, &input_gate_output, ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::LOGISTIC))); // _tanh_modulation_gate - const TensorInfo input_modulation_gate_output(input_modulation_gate_input.tensor_shape(), 1, DataType::QSYMM16, qsymm_0); - ARM_COMPUTE_RETURN_ON_ERROR(NEActivationLayer::validate(&input_modulation_gate_input, &input_modulation_gate_output, ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::TANH, 1.0f, 1.0f))); + const TensorInfo input_modulation_gate_output(input_modulation_gate_input.tensor_shape(), 1, DataType::QSYMM16, + qsymm_0); + ARM_COMPUTE_RETURN_ON_ERROR( + NEActivationLayer::validate(&input_modulation_gate_input, &input_modulation_gate_output, + ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::TANH, 1.0f, 1.0f))); // _sigmoid_output_gate const TensorInfo output_gate_output(output_gate_input.tensor_shape(), 1, DataType::QSYMM16, qsymm_0); - ARM_COMPUTE_RETURN_ON_ERROR(NEActivationLayer::validate(&output_gate_input, &output_gate_output, ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::LOGISTIC))); + ARM_COMPUTE_RETURN_ON_ERROR( + NEActivationLayer::validate(&output_gate_input, &output_gate_output, + ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::LOGISTIC))); // _mul_forget_gate_cell_state const TensorInfo cell_state_tmp1(forget_gate_output.tensor_shape(), 1, DataType::QSYMM16, qsymm_4); - ARM_COMPUTE_RETURN_ON_ERROR(NEPixelWiseMultiplication::validate(&forget_gate_output, cell_state_in, &cell_state_tmp1, 1, ConvertPolicy::SATURATE, RoundingPolicy::TO_ZERO)); + ARM_COMPUTE_RETURN_ON_ERROR(NEPixelWiseMultiplication::validate( + &forget_gate_output, cell_state_in, &cell_state_tmp1, 1, ConvertPolicy::SATURATE, RoundingPolicy::TO_ZERO)); // _mul_input_gate_input_mod_gate const TensorInfo cell_state_tmp2(input_gate_output.tensor_shape(), 1, DataType::QSYMM16, qsymm_4); - ARM_COMPUTE_RETURN_ON_ERROR(NEPixelWiseMultiplication::validate(&input_gate_output, &input_modulation_gate_output, &cell_state_tmp2, 1, ConvertPolicy::SATURATE, RoundingPolicy::TO_ZERO)); + ARM_COMPUTE_RETURN_ON_ERROR(NEPixelWiseMultiplication::validate(&input_gate_output, &input_modulation_gate_output, + &cell_state_tmp2, 1, ConvertPolicy::SATURATE, + RoundingPolicy::TO_ZERO)); // _add_cell_state_tmps - ARM_COMPUTE_RETURN_ON_ERROR(NEArithmeticAddition::validate(&cell_state_tmp1, &cell_state_tmp2, cell_state_out, ConvertPolicy::SATURATE)); + ARM_COMPUTE_RETURN_ON_ERROR( + NEArithmeticAddition::validate(&cell_state_tmp1, &cell_state_tmp2, cell_state_out, ConvertPolicy::SATURATE)); // _tanh_modulation_gate const TensorInfo output_state_tmp(cell_state_out->tensor_shape(), 1, DataType::QSYMM16, qsymm_0); - ARM_COMPUTE_RETURN_ON_ERROR(NEActivationLayer::validate(cell_state_out, &output_state_tmp, ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::TANH, 1.0f, 1.0f))); + ARM_COMPUTE_RETURN_ON_ERROR( + NEActivationLayer::validate(cell_state_out, &output_state_tmp, + ActivationLayerInfo(ActivationLayerInfo::ActivationFunction::TANH, 1.0f, 1.0f))); // _mul_output_state_tmp_output_gate const TensorInfo output_state_out_symm(output_gate_output.tensor_shape(), 1, DataType::QSYMM16, qsymm_0); - ARM_COMPUTE_RETURN_ON_ERROR(NEPixelWiseMultiplication::validate(&output_state_tmp, &output_gate_output, &output_state_out_symm, 1, ConvertPolicy::SATURATE, RoundingPolicy::TO_ZERO)); + ARM_COMPUTE_RETURN_ON_ERROR(NEPixelWiseMultiplication::validate(&output_state_tmp, &output_gate_output, + &output_state_out_symm, 1, ConvertPolicy::SATURATE, + RoundingPolicy::TO_ZERO)); // _dequantize const TensorInfo output_state_out_f32(output_state_out_symm.tensor_shape(), 1, DataType::F32); @@ -435,14 +603,14 @@ Status NELSTMLayerQuantized::validate(const ITensorInfo *input, // _quantize ARM_COMPUTE_RETURN_ON_ERROR(NEQuantizationLayer::validate(&output_state_out_f32, output_state_out)); - if(cell_state_out->total_size() != 0) + if (cell_state_out->total_size() != 0) { ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(&cell_state_info, cell_state_out); ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_SHAPES(&cell_state_info, cell_state_out); ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_QUANTIZATION_INFO(&cell_state_info, cell_state_out); } - if(output_state_out->total_size() != 0) + if (output_state_out->total_size() != 0) { ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(&output_state_info, output_state_out); ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_SHAPES(&output_state_info, output_state_out); @@ -501,7 +669,7 @@ void NELSTMLayerQuantized::run() void NELSTMLayerQuantized::prepare() { - if(!_is_prepared) + if (!_is_prepared) { _input_weights.allocator()->allocate(); _concat_input_weights.run(); |