|
|
@@ -0,0 +1,2955 @@
|
|
|
+/* Copyright 2020 The TensorFlow Authors. All Rights Reserved.
|
|
|
+
|
|
|
+Licensed under the Apache License, Version 2.0 (the "License");
|
|
|
+you may not use this file except in compliance with the License.
|
|
|
+You may obtain a copy of the License at
|
|
|
+
|
|
|
+ http://www.apache.org/licenses/LICENSE-2.0
|
|
|
+
|
|
|
+Unless required by applicable law or agreed to in writing, software
|
|
|
+distributed under the License is distributed on an "AS IS" BASIS,
|
|
|
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
|
+See the License for the specific language governing permissions and
|
|
|
+limitations under the License.
|
|
|
+==============================================================================*/
|
|
|
+#include "tensorflow/lite/micro/kernels/lstm_eval.h"
|
|
|
+
|
|
|
+#include <cmath>
|
|
|
+#include <cstdint>
|
|
|
+#include <cstring>
|
|
|
+#include <memory>
|
|
|
+
|
|
|
+#include "tensorflow/lite/c/builtin_op_data.h"
|
|
|
+#include "tensorflow/lite/c/common.h"
|
|
|
+#include "tensorflow/lite/kernels/internal/compatibility.h"
|
|
|
+#include "tensorflow/lite/kernels/internal/tensor_ctypes.h"
|
|
|
+#include "tensorflow/lite/kernels/op_macros.h"
|
|
|
+#include "tensorflow/lite/micro/kernels/kernel_util.h"
|
|
|
+#include "tensorflow/lite/micro/kernels/micro_tensor_utils.h"
|
|
|
+namespace tflite {
|
|
|
+namespace {
|
|
|
+
|
|
|
+void ComputeRowSums(
|
|
|
+ int32_t* input_to_input_row_sums, int32_t* input_to_forget_row_sums,
|
|
|
+ int32_t* input_to_cell_row_sums, int32_t* input_to_output_row_sums,
|
|
|
+ int32_t* aux_input_to_input_row_sums, int32_t* aux_input_to_forget_row_sums,
|
|
|
+ int32_t* aux_input_to_cell_row_sums, int32_t* aux_input_to_output_row_sums,
|
|
|
+ int32_t* recurrent_to_input_row_sums, int32_t* recurrent_to_forget_row_sums,
|
|
|
+ int32_t* recurrent_to_cell_row_sums, int32_t* recurrent_to_output_row_sums,
|
|
|
+ int32_t* projection_weights_row_sums, int32_t* row_sums, int n_cell,
|
|
|
+ int n_input, int n_aux_input, int n_output,
|
|
|
+ const int8_t* input_to_input_weights_ptr,
|
|
|
+ const int8_t* input_to_forget_weights_ptr,
|
|
|
+ const int8_t* input_to_cell_weights_ptr,
|
|
|
+ const int8_t* input_to_output_weights_ptr,
|
|
|
+ const int8_t* aux_input_to_input_weights_ptr,
|
|
|
+ const int8_t* aux_input_to_forget_weights_ptr,
|
|
|
+ const int8_t* aux_input_to_cell_weights_ptr,
|
|
|
+ const int8_t* aux_input_to_output_weights_ptr,
|
|
|
+ const int8_t* recurrent_to_input_weights_ptr,
|
|
|
+ const int8_t* recurrent_to_forget_weights_ptr,
|
|
|
+ const int8_t* recurrent_to_cell_weights_ptr,
|
|
|
+ const int8_t* recurrent_to_output_weights_ptr,
|
|
|
+ const int8_t* projection_weights_ptr, bool use_cifg,
|
|
|
+ const float* aux_input_ptr) {
|
|
|
+ // Compute the row sums for dequantization
|
|
|
+ if (!use_cifg) {
|
|
|
+ micro_tensor_utils::ReductionSumVector(
|
|
|
+ input_to_input_weights_ptr, input_to_input_row_sums, n_cell, n_input);
|
|
|
+ }
|
|
|
+ micro_tensor_utils::ReductionSumVector(
|
|
|
+ input_to_forget_weights_ptr, input_to_forget_row_sums, n_cell, n_input);
|
|
|
+ micro_tensor_utils::ReductionSumVector(
|
|
|
+ input_to_cell_weights_ptr, input_to_cell_row_sums, n_cell, n_input);
|
|
|
+ micro_tensor_utils::ReductionSumVector(
|
|
|
+ input_to_output_weights_ptr, input_to_output_row_sums, n_cell, n_input);
|
|
|
+
|
|
|
+ if (aux_input_ptr) {
|
|
|
+ if (!use_cifg) {
|
|
|
+ micro_tensor_utils::ReductionSumVector(aux_input_to_input_weights_ptr,
|
|
|
+ aux_input_to_input_row_sums,
|
|
|
+ n_cell, n_aux_input);
|
|
|
+ }
|
|
|
+ micro_tensor_utils::ReductionSumVector(aux_input_to_forget_weights_ptr,
|
|
|
+ aux_input_to_forget_row_sums, n_cell,
|
|
|
+ n_aux_input);
|
|
|
+ micro_tensor_utils::ReductionSumVector(aux_input_to_cell_weights_ptr,
|
|
|
+ aux_input_to_cell_row_sums, n_cell,
|
|
|
+ n_aux_input);
|
|
|
+ micro_tensor_utils::ReductionSumVector(aux_input_to_output_weights_ptr,
|
|
|
+ aux_input_to_output_row_sums, n_cell,
|
|
|
+ n_aux_input);
|
|
|
+ }
|
|
|
+ if (!use_cifg) {
|
|
|
+ micro_tensor_utils::ReductionSumVector(recurrent_to_input_weights_ptr,
|
|
|
+ recurrent_to_input_row_sums, n_cell,
|
|
|
+ n_output);
|
|
|
+ }
|
|
|
+ micro_tensor_utils::ReductionSumVector(recurrent_to_forget_weights_ptr,
|
|
|
+ recurrent_to_forget_row_sums, n_cell,
|
|
|
+ n_output);
|
|
|
+ micro_tensor_utils::ReductionSumVector(recurrent_to_cell_weights_ptr,
|
|
|
+ recurrent_to_cell_row_sums, n_cell,
|
|
|
+ n_output);
|
|
|
+ micro_tensor_utils::ReductionSumVector(recurrent_to_output_weights_ptr,
|
|
|
+ recurrent_to_output_row_sums, n_cell,
|
|
|
+ n_output);
|
|
|
+
|
|
|
+ if (projection_weights_ptr != nullptr) {
|
|
|
+ micro_tensor_utils::ReductionSumVector(
|
|
|
+ projection_weights_ptr, projection_weights_row_sums, n_output, n_cell);
|
|
|
+ }
|
|
|
+}
|
|
|
+
|
|
|
+// Calculates a single LSTM gate.
|
|
|
+//
|
|
|
+// Implements the following formula: (* is matrix multiply)
|
|
|
+// gate = activate(W_input * input + W_aux * aux_input +
|
|
|
+// W_peephole * cell + W_recurrent * prev_output + bias)
|
|
|
+// with layer norm:
|
|
|
+// gate = activate(W_norm * normalize(...) + bias) // not adding bias inside
|
|
|
+//
|
|
|
+// Activation is sigmoid except for the "cell" gate (configurable, usually tanh)
|
|
|
+//
|
|
|
+// Parameters:
|
|
|
+// Input vectors (to LSTM): | Size: | Optional?
|
|
|
+// input | n_input |
|
|
|
+// aux_input | n_aux_input | y (bidir LSTM)
|
|
|
+// Input vectors (persistent states):
|
|
|
+// output_state | n_output |
|
|
|
+// cell_state | n_cell |
|
|
|
+// 'Constant' inputs:
|
|
|
+// input_to_gate_weights | n_cell * n_input |
|
|
|
+// aux_input_to_gate_weights | n_cell * n_aux_input | y (bidir LSTM)
|
|
|
+// recurrent_to_gate_weights | n_cell * n_output |
|
|
|
+// cell_to_gate_weights | n_cell | y (peephole)
|
|
|
+// gate_bias | n_cell |
|
|
|
+// layer_norm_coefficients | n_cell | y (layer norm)
|
|
|
+// Output vector:
|
|
|
+// gate | n_cell |
|
|
|
+// Scalar parameters:
|
|
|
+// n_batch - batch size / number of vectors
|
|
|
+// n_input, n_aux_input, n_output, n_cell - size of vectors.
|
|
|
+// activation - activation to use.
|
|
|
+// is_input_all_zeros, is_aux_input_all_zeros - if input vectors are all zero.
|
|
|
+// use_layer_norm - if doing layer norm LSTM.
|
|
|
+inline void CalculateLstmGateFloat(
|
|
|
+ const float* input, const float* input_to_gate_weights,
|
|
|
+ const float* aux_input, const float* aux_input_to_gate_weights,
|
|
|
+ const float* output_state, const float* recurrent_to_gate_weights,
|
|
|
+ const float* cell_state, const float* cell_to_gate_weights,
|
|
|
+ const float* layer_norm_coefficients, const float* gate_bias,
|
|
|
+ const int n_batch, const int n_input, const int n_aux_input,
|
|
|
+ const int n_output, const int n_cell,
|
|
|
+ const TfLiteFusedActivation activation, float* gate,
|
|
|
+ const bool is_input_all_zeros, const bool is_aux_input_all_zeros) {
|
|
|
+ const bool use_peephole = (cell_to_gate_weights != nullptr);
|
|
|
+ const bool use_layer_norm = (layer_norm_coefficients != nullptr);
|
|
|
+
|
|
|
+ // Initialize scratch buffers with bias for regular lstm or initialize with
|
|
|
+ // zero for layer norm lstm.
|
|
|
+ if (use_layer_norm) {
|
|
|
+ memset(gate, 0, n_cell * n_batch * sizeof(float));
|
|
|
+ } else {
|
|
|
+ micro_tensor_utils::VectorBatchVectorAssign(gate_bias, n_cell, n_batch,
|
|
|
+ gate);
|
|
|
+ }
|
|
|
+ // For each batch and cell: compute input_weight * input.
|
|
|
+ // Skip if input is all zeros.
|
|
|
+ if (!is_input_all_zeros) {
|
|
|
+ micro_tensor_utils::MatrixBatchVectorMultiplyAccumulate(
|
|
|
+ input_to_gate_weights, n_cell, n_input, input, n_batch, gate);
|
|
|
+ }
|
|
|
+ // For each batch and cell: compute aux_input_weight * aux_input.
|
|
|
+ // Skip if auxiliary input is not available or all zeros.
|
|
|
+ if (!is_aux_input_all_zeros) {
|
|
|
+ micro_tensor_utils::MatrixBatchVectorMultiplyAccumulate(
|
|
|
+ aux_input_to_gate_weights, n_cell, n_aux_input, aux_input, n_batch,
|
|
|
+ gate);
|
|
|
+ }
|
|
|
+ // For each batch and cell: compute recurrent_weight * output_state.
|
|
|
+ micro_tensor_utils::MatrixBatchVectorMultiplyAccumulate(
|
|
|
+ recurrent_to_gate_weights, n_cell, n_output, output_state, n_batch, gate);
|
|
|
+ // For each batch and cell: compute cell_weight .* cell_state (peephole LSTM)
|
|
|
+ if (use_peephole) {
|
|
|
+ micro_tensor_utils::VectorBatchVectorCwiseProductAccumulate(
|
|
|
+ cell_to_gate_weights, n_cell, cell_state, n_batch, gate);
|
|
|
+ }
|
|
|
+ // Do layer normalization (if layer norm LSTM)
|
|
|
+ if (use_layer_norm) {
|
|
|
+ micro_tensor_utils::MeanStddevNormalization(gate, gate, n_cell, n_batch);
|
|
|
+ micro_tensor_utils::VectorBatchVectorCwiseProduct(
|
|
|
+ layer_norm_coefficients, n_cell, gate, n_batch, gate);
|
|
|
+ micro_tensor_utils::VectorBatchVectorAdd(gate_bias, n_cell, n_batch, gate);
|
|
|
+ }
|
|
|
+ // Apply activation
|
|
|
+ micro_tensor_utils::ApplyActivationToVector(gate, n_batch * n_cell,
|
|
|
+ activation, gate);
|
|
|
+}
|
|
|
+
|
|
|
+// Updates the LSTM cell state, used by both float and hybrid LSTM versions.
|
|
|
+//
|
|
|
+// Implements the following formula:
|
|
|
+// cell_state_new = clip(forget_gate * cell_state + input_gate * cell_gate)
|
|
|
+//
|
|
|
+// With CIFG LSTM, input gate is replaced by (1-forget_gate).
|
|
|
+//
|
|
|
+// Parameters:
|
|
|
+// - n_batch, n_cell: sizes of vectors
|
|
|
+// - cell_state: input/output vector, size n_batch*n_cell
|
|
|
+// - input_gate: input vector, size n_batch*n_cell.
|
|
|
+// - forget_gate: input/scratch vector, size n_batch*n_cell, modified with CIFG
|
|
|
+// - cell_gate: input vector, size n_batch*n_cell.
|
|
|
+// - use_cifg: use 1-forget_gate instead of input_gate.
|
|
|
+// - clip: if > 0, clip the resulting cell state to [-clip, +clip].
|
|
|
+void UpdateLstmCellFloat(int n_batch, int n_cell, float* cell_state,
|
|
|
+ const float* input_gate, float* forget_gate,
|
|
|
+ const float* cell_gate, bool use_cifg, float clip) {
|
|
|
+ micro_tensor_utils::VectorVectorCwiseProduct(forget_gate, cell_state,
|
|
|
+ n_batch * n_cell, cell_state);
|
|
|
+
|
|
|
+ if (use_cifg) {
|
|
|
+ // With CIFG, input_gate = 1-forget_gate. Use the forget_gate array as
|
|
|
+ // scratch, as input_gate array is not allocated in this case. (Be careful
|
|
|
+ // not to write to the scratch before reading the forget gate data.)
|
|
|
+ float* scratch = forget_gate;
|
|
|
+ micro_tensor_utils::Sub1Vector(forget_gate, n_batch * n_cell, scratch);
|
|
|
+ micro_tensor_utils::VectorVectorCwiseProductAccumulate(
|
|
|
+ cell_gate, scratch, n_batch * n_cell, cell_state);
|
|
|
+ } else {
|
|
|
+ micro_tensor_utils::VectorVectorCwiseProductAccumulate(
|
|
|
+ cell_gate, input_gate, n_batch * n_cell, cell_state);
|
|
|
+ }
|
|
|
+ if (clip > 0.0f) {
|
|
|
+ micro_tensor_utils::CwiseClipping(cell_state, n_batch * n_cell, clip);
|
|
|
+ }
|
|
|
+}
|
|
|
+
|
|
|
+// Calculates the output state tensor of an LSTM step.
|
|
|
+//
|
|
|
+// Implements the following formula:
|
|
|
+// output_no_projection = output_gate .* activate(cell_state)
|
|
|
+// (elementwise vector product)
|
|
|
+// If no projection is used:
|
|
|
+// output = output_state = output_no_projection
|
|
|
+// With projection:
|
|
|
+// output = output_state = clip(W*output_no_projection + bias)
|
|
|
+//
|
|
|
+// Output might not have a different 'stride' than n_batch, so we need to copy.
|
|
|
+//
|
|
|
+// Parameters:
|
|
|
+// - n_batch: batches: the number of distinct vectors in each array.
|
|
|
+// - n_cell, n_output: sizes of vectors.
|
|
|
+// - cell_state, output_gate: input vectors, size n_batch*n_cell.
|
|
|
+// - projection_weights, projection_weights_scale, projection_bias:
|
|
|
+// constant inputs, describing projection matrix and bias.
|
|
|
+// - proj_clip: if > 0, clip the output of the projection.
|
|
|
+// - output_state: output vector, size n_batch*n_output. Must be contigous.
|
|
|
+// - scratch: scratch area, size n_batch*n_cell.
|
|
|
+void CalculateLstmOutputFloat(int n_batch, int n_cell, int n_output,
|
|
|
+ const float* cell_state, const float* output_gate,
|
|
|
+ TfLiteFusedActivation activation,
|
|
|
+ const float* projection_weights,
|
|
|
+ const float* projection_bias,
|
|
|
+ const float proj_clip, float* output_state,
|
|
|
+ float* scratch) {
|
|
|
+ micro_tensor_utils::ApplyActivationToVector(cell_state, n_batch * n_cell,
|
|
|
+ activation, scratch);
|
|
|
+ micro_tensor_utils::VectorVectorCwiseProduct(output_gate, scratch,
|
|
|
+ n_batch * n_cell, scratch);
|
|
|
+
|
|
|
+ const bool use_projection = (projection_weights != nullptr);
|
|
|
+ const bool use_projection_bias = (projection_bias != nullptr);
|
|
|
+
|
|
|
+ if (use_projection) {
|
|
|
+ if (use_projection_bias) {
|
|
|
+ micro_tensor_utils::VectorBatchVectorAssign(projection_bias, n_output,
|
|
|
+ n_batch, output_state);
|
|
|
+ } else {
|
|
|
+ memset(output_state, 0, n_batch * n_output * sizeof(float));
|
|
|
+ }
|
|
|
+ micro_tensor_utils::MatrixBatchVectorMultiplyAccumulate(
|
|
|
+ projection_weights, n_output, n_cell, scratch, n_batch, output_state);
|
|
|
+ if (proj_clip > 0.0f) {
|
|
|
+ micro_tensor_utils::CwiseClipping(output_state, n_batch * n_output,
|
|
|
+ proj_clip);
|
|
|
+ }
|
|
|
+ } else {
|
|
|
+ std::memcpy(output_state, scratch, n_batch * n_output * sizeof(float));
|
|
|
+ }
|
|
|
+}
|
|
|
+
|
|
|
+// Calculates a single LSTM gate, hybrid version.
|
|
|
+// Implements the same functionality as CalculateLstmGateFloat.
|
|
|
+void CalculateLstmGateHybrid(
|
|
|
+ // Input and weights
|
|
|
+ const int8_t* input, const float* input_sf, const int32_t* input_zp,
|
|
|
+ const int8_t* input_to_gate_weights,
|
|
|
+ const uint8_t* input_to_gate_weights_ledger,
|
|
|
+ const float input_to_gate_weights_scale, int32_t* input_to_gate_row_sums,
|
|
|
+ // Aux input and weights
|
|
|
+ const int8_t* aux_input, const float* aux_input_sf,
|
|
|
+ const int32_t* aux_input_zp, const int8_t* aux_input_to_gate_weights,
|
|
|
+ const float aux_input_to_gate_weights_scale,
|
|
|
+ int32_t* aux_input_to_gate_row_sums,
|
|
|
+ // Output state and weights
|
|
|
+ const int8_t* output_state, const float* output_state_sf,
|
|
|
+ const int32_t* output_state_zp, const int8_t* recurrent_to_gate_weights,
|
|
|
+ const uint8_t* recurrent_to_gate_weights_ledger,
|
|
|
+ const float recurrent_to_gate_weights_scale,
|
|
|
+ int32_t* recurrent_to_gate_row_sums,
|
|
|
+ // Cell state and weights (peephole LSTM)
|
|
|
+ const float* cell_state, const int8_t* cell_to_gate_weights,
|
|
|
+ const float cell_to_gate_weights_scale,
|
|
|
+ // Layer normalization coefficients (layer norm LSTM) + gate bias
|
|
|
+ const float* layer_norm_coefficients, const float* gate_bias,
|
|
|
+ // Array sizes
|
|
|
+ const int n_batch, const int n_input, const int n_aux_input,
|
|
|
+ const int n_output, const int n_cell,
|
|
|
+ const TfLiteFusedActivation activation,
|
|
|
+ // Output
|
|
|
+ float* gate,
|
|
|
+ // Parameters for performance optimizations
|
|
|
+ const bool is_input_all_zeros, const bool is_aux_input_all_zeros,
|
|
|
+ const bool is_output_state_all_zeros, bool* compute_row_sums,
|
|
|
+ // Scratch arrays
|
|
|
+ float* scratch0, // size: n_batch
|
|
|
+ float* scratch1, // size: n_cell, only used if peephole LSTM
|
|
|
+ float* scales, // size: n_batch
|
|
|
+ int32_t* accum_scratch // For MatrixBatchVectorMultiplyAccumulate
|
|
|
+) {
|
|
|
+ const bool use_peephole = (cell_to_gate_weights != nullptr);
|
|
|
+ const bool use_layer_norm = (layer_norm_coefficients != nullptr);
|
|
|
+
|
|
|
+ // Initialize scratch buffers with bias for regular lstm or initialize with
|
|
|
+ // zero for layer norm lstm.
|
|
|
+ if (use_layer_norm) {
|
|
|
+ memset(gate, 0, n_cell * n_batch * sizeof(float));
|
|
|
+ } else {
|
|
|
+ micro_tensor_utils::VectorBatchVectorAssign(gate_bias, n_cell, n_batch,
|
|
|
+ gate);
|
|
|
+ }
|
|
|
+ // For each batch and cell: compute input_weight * input.
|
|
|
+ // Skip if input is all zeros.
|
|
|
+ if (!is_input_all_zeros) {
|
|
|
+ if (input_to_gate_weights_ledger != nullptr) {
|
|
|
+ for (int i = 0; i < n_batch; i++) {
|
|
|
+ scales[i] = input_to_gate_weights_scale * input_sf[i];
|
|
|
+ }
|
|
|
+ micro_tensor_utils::SparseMatrixBatchVectorMultiplyAccumulate(
|
|
|
+ input_to_gate_weights, input_to_gate_weights_ledger, n_cell, n_input,
|
|
|
+ input, scales, n_batch, gate);
|
|
|
+
|
|
|
+ } else {
|
|
|
+ micro_tensor_utils::MatrixBatchVectorMultiplyAccumulate(
|
|
|
+ input_to_gate_weights, n_cell, n_input, input,
|
|
|
+ input_to_gate_weights_scale, input_sf, n_batch, gate,
|
|
|
+ /*per_channel_scale=*/nullptr, input_zp, accum_scratch,
|
|
|
+ input_to_gate_row_sums, compute_row_sums, scratch0, nullptr);
|
|
|
+ }
|
|
|
+ }
|
|
|
+ // For each batch and cell: compute aux_input_weight * aux_input.
|
|
|
+ // Skip if auxiliary input is not available or all zeros.
|
|
|
+ if (!is_aux_input_all_zeros) {
|
|
|
+ micro_tensor_utils::MatrixBatchVectorMultiplyAccumulate(
|
|
|
+ aux_input_to_gate_weights, n_cell, n_aux_input, aux_input,
|
|
|
+ aux_input_to_gate_weights_scale, aux_input_sf, n_batch, gate,
|
|
|
+ /*per_channel_scale=*/nullptr, aux_input_zp, accum_scratch,
|
|
|
+ aux_input_to_gate_row_sums, compute_row_sums, scratch0, nullptr);
|
|
|
+ }
|
|
|
+ // For each batch and cell: compute recurrent_weight * output_state.
|
|
|
+ // Skip if output state is all zeros.
|
|
|
+ if (!is_output_state_all_zeros) {
|
|
|
+ if (recurrent_to_gate_weights_ledger != nullptr) {
|
|
|
+ for (int i = 0; i < n_batch; i++) {
|
|
|
+ scales[i] = recurrent_to_gate_weights_scale * input_sf[i];
|
|
|
+ }
|
|
|
+ micro_tensor_utils::SparseMatrixBatchVectorMultiplyAccumulate(
|
|
|
+ recurrent_to_gate_weights, recurrent_to_gate_weights_ledger, n_cell,
|
|
|
+ n_output, output_state, scales, n_batch, gate);
|
|
|
+ } else {
|
|
|
+ micro_tensor_utils::MatrixBatchVectorMultiplyAccumulate(
|
|
|
+ recurrent_to_gate_weights, n_cell, n_output, output_state,
|
|
|
+ recurrent_to_gate_weights_scale, output_state_sf, n_batch, gate,
|
|
|
+ /*per_channel_scale=*/nullptr, output_state_zp, accum_scratch,
|
|
|
+ recurrent_to_gate_row_sums, compute_row_sums, scratch0, nullptr);
|
|
|
+ }
|
|
|
+ }
|
|
|
+ // For each batch and cell: compute cell_weight .* cell_state (peephole LSTM)
|
|
|
+ if (use_peephole) {
|
|
|
+ float* recovered_cell_weights = scratch1;
|
|
|
+ micro_tensor_utils::VectorScalarMultiply(cell_to_gate_weights, n_cell,
|
|
|
+ cell_to_gate_weights_scale,
|
|
|
+ recovered_cell_weights);
|
|
|
+ micro_tensor_utils::VectorBatchVectorCwiseProductAccumulate(
|
|
|
+ recovered_cell_weights, n_cell, cell_state, n_batch, gate);
|
|
|
+ }
|
|
|
+ // Do layer normalization (if layer norm LSTM)
|
|
|
+ if (use_layer_norm) {
|
|
|
+ micro_tensor_utils::MeanStddevNormalization(gate, gate, n_cell, n_batch);
|
|
|
+ micro_tensor_utils::VectorBatchVectorCwiseProduct(
|
|
|
+ layer_norm_coefficients, n_cell, gate, n_batch, gate);
|
|
|
+ micro_tensor_utils::VectorBatchVectorAdd(gate_bias, n_cell, n_batch, gate);
|
|
|
+ }
|
|
|
+ // Apply activation
|
|
|
+ micro_tensor_utils::ApplyActivationToVector(gate, n_cell * n_batch,
|
|
|
+ activation, gate);
|
|
|
+}
|
|
|
+
|
|
|
+// Calculates the output state tensor of an LSTM step. See Float version too.
|
|
|
+//
|
|
|
+// Parameters:
|
|
|
+// - n_batch: batches: the number of distinct vectors in each array.
|
|
|
+// - n_cell, n_output: sizes of vectors.
|
|
|
+// - cell_state, output_gate: input vectors, size n_batch*n_cell.
|
|
|
+// - projection_weights, projection_weights_scale, projection_bias:
|
|
|
+// constant inputs, describing projection matrix and bias.
|
|
|
+// - proj_clip: if > 0, clip the output of the projection.
|
|
|
+// - output_state: output vector, size n_batch*n_output. Must be contigous.
|
|
|
+// - asymmetric_quantize_inputs: parameter to control quantization.
|
|
|
+// - projection_weights_row_sums, compute_row_sums: Data for optimized
|
|
|
+// MatrixBatchVectorMultiplyAccumulate.
|
|
|
+// - scratch0: scratch area of size n_batch*n_cell
|
|
|
+// - scratch1: scratch area of size n_batch*n_cell
|
|
|
+// - scratch2: scratch area of size n_batch
|
|
|
+// - scratch3: scratch area of size n_batch
|
|
|
+// - scratch4: scratch area used by MatrixBatchVectorMultiplyAccumulate
|
|
|
+// - scales: scratch area of size n_batch
|
|
|
+void CalculateLstmOutputHybrid(
|
|
|
+ int n_batch, int n_cell, int n_output, const float* cell_state,
|
|
|
+ const float* output_gate, TfLiteFusedActivation activation,
|
|
|
+ const int8_t* projection_weights, const uint8_t* projection_weights_ledger,
|
|
|
+ float projection_weights_scale, const float* projection_bias,
|
|
|
+ const float proj_clip, float* output_state, bool asymmetric_quantize_inputs,
|
|
|
+ int32_t* projection_weights_row_sums, bool* compute_row_sums,
|
|
|
+ float* scratch0, int8_t* scratch1, float* scratch2, int32_t* scratch3,
|
|
|
+ int32_t* scratch4, float* scales) {
|
|
|
+ micro_tensor_utils::ApplyActivationToVector(cell_state, n_batch * n_cell,
|
|
|
+ activation, scratch0);
|
|
|
+ micro_tensor_utils::VectorVectorCwiseProduct(output_gate, scratch0,
|
|
|
+ n_batch * n_cell, scratch0);
|
|
|
+
|
|
|
+ const bool use_projection = (projection_weights != nullptr);
|
|
|
+ const bool use_projection_bias = (projection_bias != nullptr);
|
|
|
+
|
|
|
+ if (use_projection) {
|
|
|
+ if (use_projection_bias) {
|
|
|
+ micro_tensor_utils::VectorBatchVectorAssign(projection_bias, n_output,
|
|
|
+ n_batch, output_state);
|
|
|
+ } else {
|
|
|
+ memset(output_state, 0, n_batch * n_output * sizeof(float));
|
|
|
+ }
|
|
|
+ if (!micro_tensor_utils::IsZeroVector(scratch0, n_batch * n_cell)) {
|
|
|
+ // Save quantization and matmul computation for all zero output.
|
|
|
+ micro_tensor_utils::BatchQuantizeFloats(scratch0, n_batch, n_cell,
|
|
|
+ scratch1, scratch2, scratch3,
|
|
|
+ asymmetric_quantize_inputs);
|
|
|
+ if (projection_weights_ledger != nullptr) {
|
|
|
+ for (int i = 0; i < n_batch; i++) {
|
|
|
+ scales[i] = projection_weights_scale * scratch2[i];
|
|
|
+ }
|
|
|
+ micro_tensor_utils::SparseMatrixBatchVectorMultiplyAccumulate(
|
|
|
+ projection_weights, projection_weights_ledger, n_output, n_cell,
|
|
|
+ scratch1, scales, n_batch, output_state);
|
|
|
+ } else {
|
|
|
+ micro_tensor_utils::MatrixBatchVectorMultiplyAccumulate(
|
|
|
+ projection_weights, n_output, n_cell, scratch1,
|
|
|
+ projection_weights_scale, scratch2, n_batch, output_state,
|
|
|
+ /*per_channel_scale=*/nullptr, scratch3, scratch4,
|
|
|
+ projection_weights_row_sums, compute_row_sums, scratch2, nullptr);
|
|
|
+ }
|
|
|
+ }
|
|
|
+ if (proj_clip > 0.0f) {
|
|
|
+ micro_tensor_utils::CwiseClipping(output_state, n_batch * n_output,
|
|
|
+ proj_clip);
|
|
|
+ }
|
|
|
+ } else {
|
|
|
+ std::memcpy(output_state, scratch0, n_batch * n_output * sizeof(float));
|
|
|
+ }
|
|
|
+}
|
|
|
+
|
|
|
+// Calculates a single LSTM gate, int8x8_16 version.
|
|
|
+// Implements the same functionality as CalculateLstmGateFloat.
|
|
|
+void CalculateLstmGateInteger8x8_16(
|
|
|
+ // Input and weights
|
|
|
+ const int8_t* input, const int8_t* input_to_gate_weights,
|
|
|
+ const int32_t* input_to_gate_bias, const int32_t input_to_gate_scale_a,
|
|
|
+ const int32_t input_to_gate_scale_b,
|
|
|
+ // Output state and weights
|
|
|
+ const int8_t* output_state, const int8_t* recurrent_to_gate_weights,
|
|
|
+ const int32_t* recurrent_to_gate_bias,
|
|
|
+ const int32_t recurrent_to_gate_scale_a,
|
|
|
+ const int32_t recurrent_to_gate_scale_b,
|
|
|
+ // Cell state and weights
|
|
|
+ const int16_t* cell_state, const int16_t* cell_to_gate_weights,
|
|
|
+ const int32_t cell_to_gate_scale_a, const int32_t cell_to_gate_scale_b,
|
|
|
+ // Layer normalization parameters (layer norm LSTM)
|
|
|
+ const int16_t* layer_norm_coefficients, const int32_t* layer_norm_bias,
|
|
|
+ const int32_t layer_norm_input_scale_a,
|
|
|
+ const int32_t layer_norm_input_scale_b,
|
|
|
+ const int32_t layer_norm_variance_guard,
|
|
|
+ // Array sizes
|
|
|
+ const int n_batch, const int n_input, const int n_output, const int n_cell,
|
|
|
+ const TfLiteFusedActivation activation,
|
|
|
+ // Output
|
|
|
+ int16_t* gate,
|
|
|
+ // Parameters for performance optimizations
|
|
|
+ // Scratch arrays
|
|
|
+ int32_t* scratch5) {
|
|
|
+ const bool use_peephole = (cell_to_gate_weights != nullptr);
|
|
|
+ const bool use_layer_norm = (layer_norm_coefficients != nullptr);
|
|
|
+
|
|
|
+ // Initialize scratch buffers with zeros. Note that unlike float and hybrid
|
|
|
+ // versions, bias is only used in layer normalization.
|
|
|
+ memset(gate, 0, n_batch * n_cell * sizeof(int16_t));
|
|
|
+ // For each batch and cell: compute input_weight * input.
|
|
|
+ micro_tensor_utils::MatrixBatchVectorMultiplyAccumulate(
|
|
|
+ input, input_to_gate_bias, input_to_gate_weights, input_to_gate_scale_a,
|
|
|
+ input_to_gate_scale_b, n_batch, n_input, n_cell, 0, scratch5, gate,
|
|
|
+ nullptr);
|
|
|
+ // Note: no aux_input.
|
|
|
+
|
|
|
+ // For each batch and cell: compute recurrent_weight * output_state.
|
|
|
+ micro_tensor_utils::MatrixBatchVectorMultiplyAccumulate(
|
|
|
+ output_state, recurrent_to_gate_bias, recurrent_to_gate_weights,
|
|
|
+ recurrent_to_gate_scale_a, recurrent_to_gate_scale_b, n_batch, n_output,
|
|
|
+ n_cell, 0, scratch5, gate, nullptr);
|
|
|
+ // For each batch and cell: compute cell_weight * cell_state (peephole LSTM)
|
|
|
+ if (use_peephole) {
|
|
|
+ micro_tensor_utils::VectorBatchVectorCwiseProductAccumulate(
|
|
|
+ cell_to_gate_weights, n_output, cell_state, n_batch,
|
|
|
+ cell_to_gate_scale_a, cell_to_gate_scale_b, gate);
|
|
|
+ }
|
|
|
+ // Do layer normalization (if layer norm LSTM)
|
|
|
+ if (use_layer_norm) {
|
|
|
+ micro_tensor_utils::ApplyLayerNorm(
|
|
|
+ gate, layer_norm_coefficients, layer_norm_bias,
|
|
|
+ layer_norm_input_scale_a, layer_norm_input_scale_b,
|
|
|
+ layer_norm_variance_guard, n_batch, n_cell, gate);
|
|
|
+ }
|
|
|
+ // Apply activation
|
|
|
+ switch (activation) {
|
|
|
+ case kTfLiteActSigmoid:
|
|
|
+ micro_tensor_utils::ApplySigmoid(gate, n_batch, n_cell, gate);
|
|
|
+ break;
|
|
|
+ case kTfLiteActTanh:
|
|
|
+ micro_tensor_utils::ApplyTanh(3, gate, n_batch, n_cell, gate);
|
|
|
+ break;
|
|
|
+ default:
|
|
|
+ // Only Sigmoid or Tanh is used.
|
|
|
+ TFLITE_ASSERT_FALSE;
|
|
|
+ }
|
|
|
+}
|
|
|
+
|
|
|
+// Updates the LSTM cell state, used by both integer LSTM versions.
|
|
|
+// Also see UpdateLstmCellFloat.
|
|
|
+//
|
|
|
+// Parameters:
|
|
|
+// - n_batch, n_cell: sizes of vectors
|
|
|
+// - cell_state: input/output vector, size n_batch*n_cell
|
|
|
+// - cell_state_scale: scaling factor of cell state.
|
|
|
+// - input_gate: input vector, size n_batch*n_cell.
|
|
|
+// - forget_gate: input/scratch vector, size n_batch*n_cell, always modified.
|
|
|
+// - cell_gate: input vector, size n_batch*n_cell.
|
|
|
+// - use_cifg: use 1-forget_gate instead of input_gate.
|
|
|
+// - clip: if > 0, clip the resulting cell state to [-clip, +clip].
|
|
|
+void UpdateLstmCellInteger(int n_batch, int n_cell, int16_t* cell_state,
|
|
|
+ int32_t cell_state_scale, const int16_t* input_gate,
|
|
|
+ int16_t* forget_gate, const int16_t* cell_gate,
|
|
|
+ bool use_cifg, int16_t clip) {
|
|
|
+ // Use the forget_gate array as scratch, as input_gate array is not allocated
|
|
|
+ // in CIFG case. (Be careful not to write to the scratch before reading the
|
|
|
+ // forget gate data.)
|
|
|
+ int16_t* scratch = forget_gate;
|
|
|
+
|
|
|
+ micro_tensor_utils::CwiseMul(forget_gate, cell_state, n_batch, n_cell, 15,
|
|
|
+ cell_state);
|
|
|
+ if (use_cifg) {
|
|
|
+ micro_tensor_utils::Sub1Vector(forget_gate, n_batch * n_cell, scratch);
|
|
|
+ micro_tensor_utils::CwiseMul(scratch, cell_gate, n_batch, n_cell,
|
|
|
+ 30 + cell_state_scale, scratch);
|
|
|
+ } else {
|
|
|
+ micro_tensor_utils::CwiseMul(input_gate, cell_gate, n_batch, n_cell,
|
|
|
+ 30 + cell_state_scale, scratch);
|
|
|
+ }
|
|
|
+ micro_tensor_utils::CwiseAdd(cell_state, scratch, n_batch, n_cell,
|
|
|
+ cell_state);
|
|
|
+
|
|
|
+ if (clip > 0) {
|
|
|
+ micro_tensor_utils::CwiseClipping(cell_state, n_batch * n_cell, clip);
|
|
|
+ }
|
|
|
+}
|
|
|
+
|
|
|
+// Calculates the output state tensor of an LSTM step. See Float and hybrid
|
|
|
+// versions as well.
|
|
|
+//
|
|
|
+// Parameters:
|
|
|
+// - n_batch: batches: the number of distinct vectors in each array.
|
|
|
+// - n_cell, n_output: sizes of vectors.
|
|
|
+// - cell_state, output_gate: input vectors, size n_batch*n_cell.
|
|
|
+// - cell_state_scale: scaling of cell_state.
|
|
|
+// - hidden_scale_[a|b]: effective scale of cell_state.*output_gate
|
|
|
+// - hidden_zp: zero_point for cell_state.*output_gate
|
|
|
+// - projection_weights, proj_scale_[a|b], projection_bias:
|
|
|
+// constant inputs, describing projection matrix and bias.
|
|
|
+// - output_state_zp: zero point of output_state. (Input, calibrated value.)
|
|
|
+// - quantized_proj_clip: if > 0, clip the output of the projection.
|
|
|
+// - output_state: output vector, size n_batch*n_output. Must be contigous.
|
|
|
+// - scratch0: scratch area of size n_batch*n_cell
|
|
|
+// - scratch1: scratch area of size n_batch*n_cell
|
|
|
+// - scratch2: scratch area used by MatrixBatchVectorMultiplyAccumulate
|
|
|
+void CalculateLstmOutputInteger8x8_16(
|
|
|
+ int n_batch, int n_cell, int n_output, const int16_t* cell_state,
|
|
|
+ int32_t cell_state_scale, const int16_t* output_gate,
|
|
|
+ int32_t hidden_scale_a, int32_t hidden_scale_b, int32_t hidden_zp,
|
|
|
+ const int8_t* projection_weights, int32_t proj_scale_a,
|
|
|
+ int32_t proj_scale_b, const int32_t* projection_bias,
|
|
|
+ int32_t output_state_zp, int8_t quantized_proj_clip, int8_t* output_state,
|
|
|
+ int16_t* scratch0, int8_t* scratch1, int32_t* scratch2) {
|
|
|
+ // Note: unlike float/hybrid, the activation is always Tanh.
|
|
|
+ micro_tensor_utils::ApplyTanh(15 + cell_state_scale, cell_state, n_batch,
|
|
|
+ n_cell, scratch0);
|
|
|
+ micro_tensor_utils::CwiseMul(output_gate, scratch0, hidden_scale_a,
|
|
|
+ hidden_scale_b, n_batch, n_cell, hidden_zp,
|
|
|
+ scratch1);
|
|
|
+
|
|
|
+ const bool use_projection = (projection_weights != nullptr);
|
|
|
+
|
|
|
+ if (use_projection) {
|
|
|
+ // Note: no bias like in float/hybrid
|
|
|
+ memset(output_state, 0, n_batch * n_output * sizeof(int8_t));
|
|
|
+ micro_tensor_utils::MatrixBatchVectorMultiplyAccumulate(
|
|
|
+ scratch1, projection_bias, projection_weights, proj_scale_a,
|
|
|
+ proj_scale_b, n_batch, n_cell, n_output, output_state_zp, scratch2,
|
|
|
+ output_state, nullptr);
|
|
|
+ if (quantized_proj_clip > 0) {
|
|
|
+ micro_tensor_utils::CwiseClipping(output_state, n_batch * n_output,
|
|
|
+ quantized_proj_clip);
|
|
|
+ }
|
|
|
+ } else {
|
|
|
+ std::memcpy(output_state, scratch1, n_batch * n_output * sizeof(int8_t));
|
|
|
+ }
|
|
|
+}
|
|
|
+
|
|
|
+// Calculates a single LSTM gate, int8x8_8 version.
|
|
|
+// Implements the same functionality as CalculateLstmGateFloat.
|
|
|
+void CalculateLstmGateInteger8x8_8(
|
|
|
+ // Inputs and weights
|
|
|
+ const int8_t* input, int32_t input_zp, const int8_t* input_to_gate_weight,
|
|
|
+ const int32_t input_to_gate_scale_a, const int32_t input_to_gate_scale_b,
|
|
|
+ const int32_t input_times_weights_scale_a,
|
|
|
+ const int32_t input_times_weights_scale_b,
|
|
|
+ const int32_t input_times_weights_zp,
|
|
|
+ // Output state and weights
|
|
|
+ const int8_t* output_state, const int32_t output_state_zp,
|
|
|
+ const int8_t* recurrent_to_gate_weight,
|
|
|
+ const int32_t recurrent_to_gate_scale_a,
|
|
|
+ const int32_t recurrent_to_gate_scale_b,
|
|
|
+ const int32_t output_state_times_weights_scale_a,
|
|
|
+ const int32_t output_state_times_weights_scale_b,
|
|
|
+ const int32_t output_state_times_weights_zp,
|
|
|
+ // Layer normalization parameters (layer norm LSTM)
|
|
|
+ const int16_t* layer_norm_gate_weight,
|
|
|
+ const int32_t layer_norm_gate_scale_a,
|
|
|
+ const int32_t layer_norm_gate_scale_b, const int32_t* gate_bias,
|
|
|
+ // Array sizes
|
|
|
+ const int n_batch, const int n_input, const int n_output, const int n_cell,
|
|
|
+ const TfLiteFusedActivation activation,
|
|
|
+ // Output
|
|
|
+ int16_t* gate,
|
|
|
+ // Scratch arrays, both sized n_batch*n_cell
|
|
|
+ int8_t* scratch0, int8_t* scratch1) {
|
|
|
+ // Multiply input * input_weights => scratch0
|
|
|
+ micro_tensor_utils::MatrixBatchVectorMultiply(
|
|
|
+ input, input_zp, input_to_gate_weight, input_to_gate_scale_a,
|
|
|
+ input_to_gate_scale_b, n_batch, n_input, n_cell, scratch0,
|
|
|
+ input_times_weights_zp);
|
|
|
+ // Multiply output_state * recurrent_weights => scratch1
|
|
|
+ micro_tensor_utils::MatrixBatchVectorMultiply(
|
|
|
+ output_state, output_state_zp, recurrent_to_gate_weight,
|
|
|
+ recurrent_to_gate_scale_a, recurrent_to_gate_scale_b, n_batch, n_output,
|
|
|
+ n_cell, scratch1, output_state_times_weights_zp);
|
|
|
+ // Add scratch0 + scratch1 => gate
|
|
|
+ micro_tensor_utils::TwoGateSaturatingAdd(
|
|
|
+ scratch0, input_times_weights_zp, scratch1, output_state_times_weights_zp,
|
|
|
+ input_times_weights_scale_a, input_times_weights_scale_b,
|
|
|
+ output_state_times_weights_scale_a, output_state_times_weights_scale_b,
|
|
|
+ n_batch, n_cell, gate);
|
|
|
+ // Apply layer normalization.
|
|
|
+ micro_tensor_utils::ApplyLayerNormFloat(
|
|
|
+ gate, layer_norm_gate_weight, layer_norm_gate_scale_a,
|
|
|
+ layer_norm_gate_scale_b, gate_bias, n_batch, n_cell, gate);
|
|
|
+ // Apply activation.
|
|
|
+ switch (activation) {
|
|
|
+ case kTfLiteActSigmoid:
|
|
|
+ micro_tensor_utils::ApplySigmoidFloat(gate, n_batch, n_cell, gate);
|
|
|
+ break;
|
|
|
+ case kTfLiteActTanh:
|
|
|
+ micro_tensor_utils::ApplyTanhFloat(gate, n_batch, n_cell, -12, gate);
|
|
|
+ break;
|
|
|
+ default:
|
|
|
+ // Only Sigmoid or Tanh is used.
|
|
|
+ TFLITE_ASSERT_FALSE;
|
|
|
+ }
|
|
|
+}
|
|
|
+
|
|
|
+// Calculates the output state tensor of an LSTM step. See Float and hybrid
|
|
|
+// versions as well.
|
|
|
+//
|
|
|
+// Parameters:
|
|
|
+// - n_batch: batches: the number of distinct vectors in each array.
|
|
|
+// - n_cell, n_output: sizes of vectors.
|
|
|
+// - cell_state, output_gate: input vectors, size n_batch*n_cell.
|
|
|
+// - projection_weights, proj_scale_[a|b], projection_bias:
|
|
|
+// constant inputs, describing projection matrix and bias.
|
|
|
+// - output_state_zp: zero point of the output state.
|
|
|
+// - quantized_proj_clip: if > 0, clip the output of the projection.
|
|
|
+// - output_state: output vector, size n_batch*n_output. Must be contigous.
|
|
|
+// - scratch: scratch area of size n_batch*n_cell
|
|
|
+void CalculateLstmOutputInteger8x8_8(
|
|
|
+ int n_batch, int n_cell, int n_output, const int16_t* cell_state,
|
|
|
+ const int16_t* output_gate, const int8_t* projection_weights,
|
|
|
+ int32_t proj_scale_a, int32_t proj_scale_b, const int32_t* projection_bias,
|
|
|
+ int32_t output_state_zp, int32_t quantized_proj_clip, int8_t* output_state,
|
|
|
+ int16_t* scratch) {
|
|
|
+ // Note: unlike float/hybrid, the activation is always Tanh.
|
|
|
+ micro_tensor_utils::ApplyTanhFloat(cell_state, n_batch, n_cell, -15, scratch);
|
|
|
+ micro_tensor_utils::CwiseMul(output_gate, scratch, n_batch, n_cell,
|
|
|
+ 15 + 15 - 15, scratch);
|
|
|
+ // Note: no bias like in float/hybrid
|
|
|
+ micro_tensor_utils::MatrixBatchVectorMultiply(
|
|
|
+ scratch, projection_weights, proj_scale_a, proj_scale_b, projection_bias,
|
|
|
+ n_batch, n_cell, n_output, output_state_zp, output_state);
|
|
|
+ if (quantized_proj_clip > 0) {
|
|
|
+ micro_tensor_utils::CwiseClipping(output_state, n_batch * n_output,
|
|
|
+ quantized_proj_clip);
|
|
|
+ }
|
|
|
+}
|
|
|
+
|
|
|
+// Performs an LSTM batch inference step for input specified by input_ptr.
|
|
|
+// The LSTM cell is specified by the pointers to its weights (*_weights_ptr) and
|
|
|
+// biases (*_bias_ptr), and buffers (*_scratch), along with additional
|
|
|
+// parameters:
|
|
|
+// - params: various LSTM params including activation, clipping, etc.,
|
|
|
+// - n_batch: size of batch,
|
|
|
+// - n_cell: number of cells (or units),
|
|
|
+// - n_input: the input size,
|
|
|
+// - n_aux_input: the auxiliary input size.
|
|
|
+// - n_output: the output size.
|
|
|
+// - output_batch_leading_dim: the leading dimension of the output buffer.
|
|
|
+//
|
|
|
+// Input of size 'n_batch * n_input':
|
|
|
+// input_ptr
|
|
|
+// Input of size 'n_batch * n_aux_input':
|
|
|
+// aux_input_ptr - optional (can be nullptr)
|
|
|
+//
|
|
|
+// LSTM weights:
|
|
|
+// Input weights of size 'n_cell * n_input':
|
|
|
+// input_to_input_weights - optional
|
|
|
+// input_to_forget_weights
|
|
|
+// input_to_cell_weights
|
|
|
+// input_to_output_weights
|
|
|
+// Auxiliary input weights of size 'n_cell * n_aux_input':
|
|
|
+// aux_input_to_input_weights - optional
|
|
|
+// aux_input_to_forget_weights - optional
|
|
|
+// aux_input_to_cell_weights - optional
|
|
|
+// aux_input_to_output_weights - optional
|
|
|
+// Recurrent weights of size 'n_cell * n_output':
|
|
|
+// recurrent_to_input_weights - optional
|
|
|
+// recurrent_to_forget_weights
|
|
|
+// recurrent_to_cell_weights
|
|
|
+// recurrent_to_input_weights
|
|
|
+// Peephole weights of size 'n_cell', representing diagonal matrices.
|
|
|
+// cell_to_input_weights - optional
|
|
|
+// cell_to_cell_weights - optional
|
|
|
+// cell_to_output_weights - optional
|
|
|
+// Projection weights of size 'n_output * n_cell'
|
|
|
+// projection_weights_ptr - optional
|
|
|
+// Gate biases of size 'n_cell':
|
|
|
+// input_gate_bias_ptr - optional
|
|
|
+// forget_gate_bias_ptr
|
|
|
+// cell_gate_bias_ptr
|
|
|
+// output_gate_bias_ptr
|
|
|
+//
|
|
|
+// Layer norm coefficients of size 'n_cell', representing diagonal matrices.
|
|
|
+// input_layer_norm_coefficients_ptr - optional
|
|
|
+// forget_layer_norm_coefficients_ptr - optional
|
|
|
+// cell_layer_norm_coefficients_ptr - optional
|
|
|
+// output_layer_norm_coefficients_ptr - optional
|
|
|
+//
|
|
|
+// The pointers to the cell and output state and the output are updated.
|
|
|
+//
|
|
|
+// The pointers input_ptr, aux_input_ptr, and output_ptr point to data aligned
|
|
|
+// in batch_major order, and each step processes batch_size many inputs from
|
|
|
+// input_ptr, and updates batch_size many cell and output states.
|
|
|
+//
|
|
|
+// The output_batch_dim is output.shape[-1], i.e. the outermost dimension of the
|
|
|
+// output tensor, and in most cases will be equal to n_output. It is usually not
|
|
|
+// when we want to store the LSTM output into a slice of the output tensor, e.g.
|
|
|
+// for bidirectional LSTMs with merge_outputs. In this case, the batched
|
|
|
+// operations cannot be used since they assume that the batched outputs are
|
|
|
+// contiguous, and we manually loop over the batched outputs.
|
|
|
+inline void LstmStepFloat(
|
|
|
+ const float* input_ptr, const float* input_to_input_weights_ptr,
|
|
|
+ const float* input_to_forget_weights_ptr,
|
|
|
+ const float* input_to_cell_weights_ptr,
|
|
|
+ const float* input_to_output_weights_ptr, const float* aux_input_ptr,
|
|
|
+ const float* aux_input_to_input_weights_ptr,
|
|
|
+ const float* aux_input_to_forget_weights_ptr,
|
|
|
+ const float* aux_input_to_cell_weights_ptr,
|
|
|
+ const float* aux_input_to_output_weights_ptr,
|
|
|
+ const float* recurrent_to_input_weights_ptr,
|
|
|
+ const float* recurrent_to_forget_weights_ptr,
|
|
|
+ const float* recurrent_to_cell_weights_ptr,
|
|
|
+ const float* recurrent_to_output_weights_ptr,
|
|
|
+ const float* cell_to_input_weights_ptr,
|
|
|
+ const float* cell_to_forget_weights_ptr,
|
|
|
+ const float* cell_to_output_weights_ptr,
|
|
|
+ const float* input_layer_norm_coefficients_ptr,
|
|
|
+ const float* forget_layer_norm_coefficients_ptr,
|
|
|
+ const float* cell_layer_norm_coefficients_ptr,
|
|
|
+ const float* output_layer_norm_coefficients_ptr,
|
|
|
+ const float* input_gate_bias_ptr, const float* forget_gate_bias_ptr,
|
|
|
+ const float* cell_gate_bias_ptr, const float* output_gate_bias_ptr,
|
|
|
+ const float* projection_weights_ptr, const float* projection_bias_ptr,
|
|
|
+ const TfLiteLSTMParams* params, int n_batch, int n_cell, int n_input,
|
|
|
+ int n_aux_input, int n_output, int output_batch_leading_dim,
|
|
|
+ float* output_state_ptr, float* cell_state_ptr, float* scratch0,
|
|
|
+ float* scratch1, float* scratch2, float* scratch3, float* output_ptr) {
|
|
|
+ // Since we have already checked that weights are all there or none, we can
|
|
|
+ // check the existence of only one to the get the condition.
|
|
|
+ const bool use_cifg = (input_to_input_weights_ptr == nullptr);
|
|
|
+
|
|
|
+ // Make named scratch buffers.
|
|
|
+ float* input_gate_scratch = scratch0;
|
|
|
+ float* forget_gate_scratch = scratch1;
|
|
|
+ float* cell_gate_scratch = scratch2;
|
|
|
+ float* output_gate_scratch = scratch3;
|
|
|
+
|
|
|
+ // Check if inputs are all zeros so we can skip some computations.
|
|
|
+ const bool is_input_all_zeros =
|
|
|
+ micro_tensor_utils::IsZeroVector(input_ptr, n_batch * n_input);
|
|
|
+ const bool is_aux_input_all_zeros =
|
|
|
+ (aux_input_ptr == nullptr ||
|
|
|
+ micro_tensor_utils::IsZeroVector(aux_input_ptr, n_batch * n_aux_input));
|
|
|
+ if (!use_cifg) {
|
|
|
+ // Calculate the input gate. (If not CIFG.)
|
|
|
+ CalculateLstmGateFloat(
|
|
|
+ input_ptr, input_to_input_weights_ptr, aux_input_ptr,
|
|
|
+ aux_input_to_input_weights_ptr, output_state_ptr,
|
|
|
+ recurrent_to_input_weights_ptr, cell_state_ptr,
|
|
|
+ cell_to_input_weights_ptr, input_layer_norm_coefficients_ptr,
|
|
|
+ input_gate_bias_ptr, n_batch, n_input, n_aux_input, n_output, n_cell,
|
|
|
+ /*activation=*/kTfLiteActSigmoid, input_gate_scratch,
|
|
|
+ is_input_all_zeros, is_aux_input_all_zeros);
|
|
|
+ }
|
|
|
+ // Calculate the forget gate.
|
|
|
+ CalculateLstmGateFloat(
|
|
|
+ input_ptr, input_to_forget_weights_ptr, aux_input_ptr,
|
|
|
+ aux_input_to_forget_weights_ptr, output_state_ptr,
|
|
|
+ recurrent_to_forget_weights_ptr, cell_state_ptr,
|
|
|
+ cell_to_forget_weights_ptr, forget_layer_norm_coefficients_ptr,
|
|
|
+ forget_gate_bias_ptr, n_batch, n_input, n_aux_input, n_output, n_cell,
|
|
|
+ /*activation=*/kTfLiteActSigmoid, forget_gate_scratch, is_input_all_zeros,
|
|
|
+ is_aux_input_all_zeros);
|
|
|
+ // Calculate the cell update gate.
|
|
|
+ CalculateLstmGateFloat(input_ptr, input_to_cell_weights_ptr, aux_input_ptr,
|
|
|
+ aux_input_to_cell_weights_ptr, output_state_ptr,
|
|
|
+ recurrent_to_cell_weights_ptr, /*cell_state=*/nullptr,
|
|
|
+ /*cell_to_gate_weights=*/nullptr,
|
|
|
+ cell_layer_norm_coefficients_ptr, cell_gate_bias_ptr,
|
|
|
+ n_batch, n_input, n_aux_input, n_output, n_cell,
|
|
|
+ params->activation, cell_gate_scratch,
|
|
|
+ is_input_all_zeros, is_aux_input_all_zeros);
|
|
|
+ // Update the cell state.
|
|
|
+ UpdateLstmCellFloat(n_batch, n_cell, cell_state_ptr, input_gate_scratch,
|
|
|
+ forget_gate_scratch, cell_gate_scratch, use_cifg,
|
|
|
+ params->cell_clip);
|
|
|
+ // Calculate output gate.
|
|
|
+ CalculateLstmGateFloat(
|
|
|
+ input_ptr, input_to_output_weights_ptr, aux_input_ptr,
|
|
|
+ aux_input_to_output_weights_ptr, output_state_ptr,
|
|
|
+ recurrent_to_output_weights_ptr, cell_state_ptr,
|
|
|
+ cell_to_output_weights_ptr, output_layer_norm_coefficients_ptr,
|
|
|
+ output_gate_bias_ptr, n_batch, n_input, n_aux_input, n_output, n_cell,
|
|
|
+ /*activation=*/kTfLiteActSigmoid, output_gate_scratch, is_input_all_zeros,
|
|
|
+ is_aux_input_all_zeros);
|
|
|
+ // Update the output state.
|
|
|
+ CalculateLstmOutputFloat(n_batch, n_cell, n_output, cell_state_ptr,
|
|
|
+ output_gate_scratch, params->activation,
|
|
|
+ projection_weights_ptr, projection_bias_ptr,
|
|
|
+ params->proj_clip, output_state_ptr, scratch2);
|
|
|
+ // Copy output state to the output. Note that the output's rows may not be
|
|
|
+ // contiguous (output_batch_leading_dim != n_output).
|
|
|
+ for (int b = 0; b < n_batch; b++) {
|
|
|
+ std::memcpy(output_ptr + b * output_batch_leading_dim,
|
|
|
+ output_state_ptr + b * n_output, n_output * sizeof(float));
|
|
|
+ }
|
|
|
+}
|
|
|
+
|
|
|
+// Same as above but with quantized weight matrices. In detail:
|
|
|
+// Input of size 'n_batch * n_input':
|
|
|
+// input_ptr
|
|
|
+// Input of size 'n_batch * n_aux_input':
|
|
|
+// aux_input_ptr - optional (can be nullptr)
|
|
|
+//
|
|
|
+// LSTM weights:
|
|
|
+// Quantized input weights of size 'n_cell * n_input':
|
|
|
+// input_to_input_weights - optional
|
|
|
+// input_to_forget_weights
|
|
|
+// input_to_cell_weights
|
|
|
+// input_to_input_weights
|
|
|
+// Quantized auxiliary input weights of size 'n_cell * n_aux_input':
|
|
|
+// aux_input_to_input_weights - optional
|
|
|
+// aux_input_to_forget_weights - optional
|
|
|
+// aux_input_to_cell_weights - optional
|
|
|
+// aux_input_to_output_weights - optional
|
|
|
+// Quantized recurrent weights of size 'n_cell * n_output':
|
|
|
+// recurrent_to_input_weights - optional
|
|
|
+// recurrent_to_forget_weights
|
|
|
+// recurrent_to_cell_weights
|
|
|
+// recurrent_to_input_weights
|
|
|
+// Quantized peephole weights of size 'n_cell', representing diagonal matrices.
|
|
|
+// cell_to_input_weights - optional
|
|
|
+// cell_to_cell_weights - optional
|
|
|
+// cell_to_output_weights - optional
|
|
|
+// Quantized projection weights of size 'n_output * n_cell'
|
|
|
+// projection_weights_ptr - optional
|
|
|
+// Weight scales (scalars) for each of the weights above.
|
|
|
+// input_to_input_weights_scale - optional
|
|
|
+// input_to_forget_weights_scale
|
|
|
+// input_to_cell_weights_scale
|
|
|
+// input_to_output_weights_scale
|
|
|
+// aux_input_to_input_weights_scale - optional
|
|
|
+// aux_input_to_forget_weights_scale - optional
|
|
|
+// aux_input_to_cell_weights_scale - optional
|
|
|
+// aux_input_to_output_weights_scale - optional
|
|
|
+// recurrent_to_input_weights_scale - optional
|
|
|
+// recurrent_to_forget_weights_scale
|
|
|
+// recurrent_to_cell_weights_scale
|
|
|
+// recurrent_to_output_weights_scale
|
|
|
+// cell_to_input_weights_scale,
|
|
|
+// cell_to_forget_weights_scale,
|
|
|
+// cell_to_output_weights_scale,
|
|
|
+// projection_weights_scale - optional
|
|
|
+// Gate biases of size 'n_cell':
|
|
|
+// input_gate_bias_ptr - optional
|
|
|
+// forget_gate_bias_ptr
|
|
|
+// cell_gate_bias_ptr
|
|
|
+// output_gate_bias_ptr
|
|
|
+//
|
|
|
+// Layer norm coefficients of size 'n_cell', representing diagonal matrices.
|
|
|
+// input_layer_norm_coefficients_ptr - optional
|
|
|
+// forget_layer_norm_coefficients_ptr - optional
|
|
|
+// cell_layer_norm_coefficients_ptr - optional
|
|
|
+// output_layer_norm_coefficients_ptr - optional
|
|
|
+//
|
|
|
+// Temporary pre-allocated storage for quantized values:
|
|
|
+// quantized_input_ptr (same size as input_ptr)
|
|
|
+// quantized_output_state_ptr (same size as output_state_ptr)
|
|
|
+// quantized_output_scratch (same size as cell_state_ptr)
|
|
|
+// Temporary pre-allocated storage for recovered values:
|
|
|
+// recovered_cell_weights (same size as cell_to_*_weights)
|
|
|
+//
|
|
|
+// Outputs:
|
|
|
+// output_state_ptr - size 'n_batch * n_output'
|
|
|
+// cell_state_ptr - size 'n_batch * n_cell'
|
|
|
+// output_ptr - size 'n_batch * output_batch_leading_dim'
|
|
|
+inline void LstmStepHybrid(
|
|
|
+ const float* input_ptr, const int8_t* input_to_input_weights_ptr,
|
|
|
+ const uint8_t* input_to_input_weights_ledger_ptr,
|
|
|
+ float input_to_input_weights_scale,
|
|
|
+ const int8_t* input_to_forget_weights_ptr,
|
|
|
+ const uint8_t* input_to_forget_weights_ledger_ptr,
|
|
|
+ float input_to_forget_weights_scale,
|
|
|
+ const int8_t* input_to_cell_weights_ptr,
|
|
|
+ const uint8_t* input_to_cell_weights_ledger_ptr,
|
|
|
+ float input_to_cell_weights_scale,
|
|
|
+ const int8_t* input_to_output_weights_ptr,
|
|
|
+ const uint8_t* input_to_output_weights_ledger_ptr,
|
|
|
+ float input_to_output_weights_scale, const float* aux_input_ptr,
|
|
|
+ const int8_t* aux_input_to_input_weights_ptr,
|
|
|
+ float aux_input_to_input_weights_scale,
|
|
|
+ const int8_t* aux_input_to_forget_weights_ptr,
|
|
|
+ float aux_input_to_forget_weights_scale,
|
|
|
+ const int8_t* aux_input_to_cell_weights_ptr,
|
|
|
+ float aux_input_to_cell_weights_scale,
|
|
|
+ const int8_t* aux_input_to_output_weights_ptr,
|
|
|
+ float aux_input_to_output_weights_scale,
|
|
|
+ const int8_t* recurrent_to_input_weights_ptr,
|
|
|
+ const uint8_t* recurrent_to_input_weights_ledger_ptr,
|
|
|
+ float recurrent_to_input_weights_scale,
|
|
|
+ const int8_t* recurrent_to_forget_weights_ptr,
|
|
|
+ const uint8_t* recurrent_to_forget_weights_ledger_ptr,
|
|
|
+ float recurrent_to_forget_weights_scale,
|
|
|
+ const int8_t* recurrent_to_cell_weights_ptr,
|
|
|
+ const uint8_t* recurrent_to_cell_weights_ledger_ptr,
|
|
|
+ float recurrent_to_cell_weights_scale,
|
|
|
+ const int8_t* recurrent_to_output_weights_ptr,
|
|
|
+ const uint8_t* recurrent_to_output_weights_ledger_ptr,
|
|
|
+ float recurrent_to_output_weights_scale,
|
|
|
+ const int8_t* cell_to_input_weights_ptr, float cell_to_input_weights_scale,
|
|
|
+ const int8_t* cell_to_forget_weights_ptr,
|
|
|
+ float cell_to_forget_weights_scale,
|
|
|
+ const int8_t* cell_to_output_weights_ptr,
|
|
|
+ float cell_to_output_weights_scale,
|
|
|
+ const float* input_layer_norm_coefficients_ptr,
|
|
|
+ const float* forget_layer_norm_coefficients_ptr,
|
|
|
+ const float* cell_layer_norm_coefficients_ptr,
|
|
|
+ const float* output_layer_norm_coefficients_ptr,
|
|
|
+ const float* input_gate_bias_ptr, const float* forget_gate_bias_ptr,
|
|
|
+ const float* cell_gate_bias_ptr, const float* output_gate_bias_ptr,
|
|
|
+ const int8_t* projection_weights_ptr,
|
|
|
+ const uint8_t* projection_weights_ledger_ptr,
|
|
|
+ float projection_weights_scale, const float* projection_bias_ptr,
|
|
|
+ const TfLiteLSTMParams* params, int n_batch, int n_cell, int n_input,
|
|
|
+ int n_aux_input, int n_output, int output_batch_leading_dim,
|
|
|
+ float* scratch0, float* scratch1, float* scratch2, float* scratch3,
|
|
|
+ float* scales, float* input_sf, float* aux_input_sf, float* output_state_sf,
|
|
|
+ float* scaling_factors_scratch, float* recovered_cell_weights,
|
|
|
+ int8_t* quantized_input_ptr, int8_t* quantized_aux_input_ptr,
|
|
|
+ int8_t* quantized_output_state_ptr, int8_t* quantized_output_scratch,
|
|
|
+ float* output_state_ptr, float* cell_state_ptr, int32_t* accum_scratch_ptr,
|
|
|
+ float* output_ptr, int32_t* input_zp, int32_t* aux_input_zp,
|
|
|
+ int32_t* output_state_zp, int32_t* row_sums, int row_sums_size,
|
|
|
+ bool* compute_row_sums, bool asymmetric_quantize_inputs) {
|
|
|
+ // Since we have already checked that weights are all there or none, we
|
|
|
+ // can check the existence of only one to the get the condition.
|
|
|
+ const bool use_cifg = (input_to_input_weights_ptr == nullptr);
|
|
|
+ // Make named scratch buffers for the different gates.
|
|
|
+ float* input_gate_scratch = scratch0;
|
|
|
+ float* forget_gate_scratch = scratch1;
|
|
|
+ float* cell_gate_scratch = scratch2;
|
|
|
+ float* output_gate_scratch = scratch3;
|
|
|
+
|
|
|
+ int32_t* input_to_input_row_sums = nullptr;
|
|
|
+ int32_t* input_to_forget_row_sums = nullptr;
|
|
|
+ int32_t* input_to_cell_row_sums = nullptr;
|
|
|
+ int32_t* input_to_output_row_sums = nullptr;
|
|
|
+ int32_t* aux_input_to_input_row_sums = nullptr;
|
|
|
+ int32_t* aux_input_to_forget_row_sums = nullptr;
|
|
|
+ int32_t* aux_input_to_cell_row_sums = nullptr;
|
|
|
+ int32_t* aux_input_to_output_row_sums = nullptr;
|
|
|
+ int32_t* recurrent_to_input_row_sums = nullptr;
|
|
|
+ int32_t* recurrent_to_forget_row_sums = nullptr;
|
|
|
+ int32_t* recurrent_to_cell_row_sums = nullptr;
|
|
|
+ int32_t* recurrent_to_output_row_sums = nullptr;
|
|
|
+ int32_t* projection_weights_row_sums = nullptr;
|
|
|
+
|
|
|
+ if (asymmetric_quantize_inputs) {
|
|
|
+ int num_row_sums = use_cifg ? 6 : 8;
|
|
|
+ if (aux_input_ptr != nullptr) {
|
|
|
+ num_row_sums += use_cifg ? 3 : 4;
|
|
|
+ }
|
|
|
+ if (projection_weights_ptr != nullptr) {
|
|
|
+ num_row_sums += ceil(static_cast<float>(n_output) / n_cell);
|
|
|
+ }
|
|
|
+ TFLITE_DCHECK(row_sums_size == num_row_sums);
|
|
|
+ input_to_input_row_sums = row_sums;
|
|
|
+ input_to_forget_row_sums =
|
|
|
+ use_cifg ? input_to_input_row_sums : input_to_input_row_sums + n_cell;
|
|
|
+ input_to_cell_row_sums = input_to_forget_row_sums + n_cell;
|
|
|
+ input_to_output_row_sums = input_to_cell_row_sums + n_cell;
|
|
|
+ if (aux_input_ptr != nullptr) {
|
|
|
+ aux_input_to_input_row_sums = input_to_output_row_sums + n_cell;
|
|
|
+ aux_input_to_forget_row_sums = use_cifg
|
|
|
+ ? aux_input_to_input_row_sums
|
|
|
+ : aux_input_to_input_row_sums + n_cell;
|
|
|
+ aux_input_to_cell_row_sums = aux_input_to_forget_row_sums + n_cell;
|
|
|
+ aux_input_to_output_row_sums = aux_input_to_cell_row_sums + n_cell;
|
|
|
+ }
|
|
|
+ recurrent_to_input_row_sums = aux_input_ptr
|
|
|
+ ? aux_input_to_output_row_sums + n_cell
|
|
|
+ : input_to_output_row_sums + n_cell;
|
|
|
+ recurrent_to_forget_row_sums = use_cifg
|
|
|
+ ? recurrent_to_input_row_sums
|
|
|
+ : recurrent_to_input_row_sums + n_cell;
|
|
|
+ recurrent_to_cell_row_sums = recurrent_to_forget_row_sums + n_cell;
|
|
|
+ recurrent_to_output_row_sums = recurrent_to_cell_row_sums + n_cell;
|
|
|
+ if (projection_weights_ptr != nullptr) {
|
|
|
+ projection_weights_row_sums = recurrent_to_output_row_sums + n_cell;
|
|
|
+ }
|
|
|
+ if (*compute_row_sums) {
|
|
|
+ ComputeRowSums(
|
|
|
+ input_to_input_row_sums, input_to_forget_row_sums,
|
|
|
+ input_to_cell_row_sums, input_to_output_row_sums,
|
|
|
+ aux_input_to_input_row_sums, aux_input_to_forget_row_sums,
|
|
|
+ aux_input_to_cell_row_sums, aux_input_to_output_row_sums,
|
|
|
+ recurrent_to_input_row_sums, recurrent_to_forget_row_sums,
|
|
|
+ recurrent_to_cell_row_sums, recurrent_to_output_row_sums,
|
|
|
+ projection_weights_row_sums, row_sums, n_cell, n_input, n_aux_input,
|
|
|
+ n_output, input_to_input_weights_ptr, input_to_forget_weights_ptr,
|
|
|
+ input_to_cell_weights_ptr, input_to_output_weights_ptr,
|
|
|
+ aux_input_to_input_weights_ptr, aux_input_to_forget_weights_ptr,
|
|
|
+ aux_input_to_cell_weights_ptr, aux_input_to_output_weights_ptr,
|
|
|
+ recurrent_to_input_weights_ptr, recurrent_to_forget_weights_ptr,
|
|
|
+ recurrent_to_cell_weights_ptr, recurrent_to_output_weights_ptr,
|
|
|
+ projection_weights_ptr, use_cifg, aux_input_ptr);
|
|
|
+ *compute_row_sums = false;
|
|
|
+ }
|
|
|
+ }
|
|
|
+
|
|
|
+ // Check if inputs are all zeros so we can skip some computations.
|
|
|
+ const bool is_input_all_zeros =
|
|
|
+ micro_tensor_utils::IsZeroVector(input_ptr, n_batch * n_input);
|
|
|
+ const bool is_aux_input_all_zeros =
|
|
|
+ (aux_input_ptr == nullptr ||
|
|
|
+ micro_tensor_utils::IsZeroVector(aux_input_ptr, n_batch * n_aux_input));
|
|
|
+ const bool is_output_state_all_zeros =
|
|
|
+ micro_tensor_utils::IsZeroVector(output_state_ptr, n_batch * n_output);
|
|
|
+ // Quantize inputs.
|
|
|
+ if (!is_input_all_zeros) {
|
|
|
+ micro_tensor_utils::BatchQuantizeFloats(
|
|
|
+ input_ptr, n_batch, n_input, quantized_input_ptr, input_sf, input_zp,
|
|
|
+ asymmetric_quantize_inputs);
|
|
|
+ }
|
|
|
+ if (!is_aux_input_all_zeros) {
|
|
|
+ micro_tensor_utils::BatchQuantizeFloats(
|
|
|
+ aux_input_ptr, n_batch, n_aux_input, quantized_aux_input_ptr,
|
|
|
+ aux_input_sf, aux_input_zp, asymmetric_quantize_inputs);
|
|
|
+ }
|
|
|
+ if (!is_output_state_all_zeros) {
|
|
|
+ micro_tensor_utils::BatchQuantizeFloats(
|
|
|
+ output_state_ptr, n_batch, n_output, quantized_output_state_ptr,
|
|
|
+ output_state_sf, output_state_zp, asymmetric_quantize_inputs);
|
|
|
+ }
|
|
|
+ if (!use_cifg) {
|
|
|
+ // Calculate the input gate. (If not CIFG.)
|
|
|
+ CalculateLstmGateHybrid(
|
|
|
+ quantized_input_ptr, input_sf, input_zp, input_to_input_weights_ptr,
|
|
|
+ input_to_input_weights_ledger_ptr, input_to_input_weights_scale,
|
|
|
+ input_to_input_row_sums, quantized_aux_input_ptr, aux_input_sf,
|
|
|
+ aux_input_zp, aux_input_to_input_weights_ptr,
|
|
|
+ aux_input_to_input_weights_scale, aux_input_to_input_row_sums,
|
|
|
+ quantized_output_state_ptr, output_state_sf, output_state_zp,
|
|
|
+ recurrent_to_input_weights_ptr, recurrent_to_input_weights_ledger_ptr,
|
|
|
+ recurrent_to_input_weights_scale, recurrent_to_input_row_sums,
|
|
|
+ cell_state_ptr, cell_to_input_weights_ptr, cell_to_input_weights_scale,
|
|
|
+ input_layer_norm_coefficients_ptr, input_gate_bias_ptr, n_batch,
|
|
|
+ n_input, n_aux_input, n_output, n_cell, kTfLiteActSigmoid,
|
|
|
+ input_gate_scratch, is_input_all_zeros, is_aux_input_all_zeros,
|
|
|
+ is_output_state_all_zeros, compute_row_sums, scaling_factors_scratch,
|
|
|
+ recovered_cell_weights, scales, accum_scratch_ptr);
|
|
|
+ }
|
|
|
+ // Calculate the forget gate.
|
|
|
+ CalculateLstmGateHybrid(
|
|
|
+ quantized_input_ptr, input_sf, input_zp, input_to_forget_weights_ptr,
|
|
|
+ input_to_forget_weights_ledger_ptr, input_to_forget_weights_scale,
|
|
|
+ input_to_forget_row_sums, quantized_aux_input_ptr, aux_input_sf,
|
|
|
+ aux_input_zp, aux_input_to_forget_weights_ptr,
|
|
|
+ aux_input_to_forget_weights_scale, aux_input_to_forget_row_sums,
|
|
|
+ quantized_output_state_ptr, output_state_sf, output_state_zp,
|
|
|
+ recurrent_to_forget_weights_ptr, recurrent_to_forget_weights_ledger_ptr,
|
|
|
+ recurrent_to_forget_weights_scale, recurrent_to_forget_row_sums,
|
|
|
+ cell_state_ptr, cell_to_forget_weights_ptr, cell_to_forget_weights_scale,
|
|
|
+ forget_layer_norm_coefficients_ptr, forget_gate_bias_ptr, n_batch,
|
|
|
+ n_input, n_aux_input, n_output, n_cell, kTfLiteActSigmoid,
|
|
|
+ forget_gate_scratch, is_input_all_zeros, is_aux_input_all_zeros,
|
|
|
+ is_output_state_all_zeros, compute_row_sums, scaling_factors_scratch,
|
|
|
+ recovered_cell_weights, scales, accum_scratch_ptr);
|
|
|
+ // Calculate the cell update gate.
|
|
|
+ CalculateLstmGateHybrid(
|
|
|
+ quantized_input_ptr, input_sf, input_zp, input_to_cell_weights_ptr,
|
|
|
+ input_to_cell_weights_ledger_ptr, input_to_cell_weights_scale,
|
|
|
+ input_to_cell_row_sums, quantized_aux_input_ptr, aux_input_sf,
|
|
|
+ aux_input_zp, aux_input_to_cell_weights_ptr,
|
|
|
+ aux_input_to_cell_weights_scale, aux_input_to_cell_row_sums,
|
|
|
+ quantized_output_state_ptr, output_state_sf, output_state_zp,
|
|
|
+ recurrent_to_cell_weights_ptr, recurrent_to_cell_weights_ledger_ptr,
|
|
|
+ recurrent_to_cell_weights_scale, recurrent_to_cell_row_sums,
|
|
|
+ /*cell_state=*/nullptr, /*cell_to_gate_weights=*/nullptr,
|
|
|
+ /*cell_to_gate_weights_scale=*/0.0f, cell_layer_norm_coefficients_ptr,
|
|
|
+ cell_gate_bias_ptr, n_batch, n_input, n_aux_input, n_output, n_cell,
|
|
|
+ params->activation, cell_gate_scratch, is_input_all_zeros,
|
|
|
+ is_aux_input_all_zeros, is_output_state_all_zeros, compute_row_sums,
|
|
|
+ scaling_factors_scratch, recovered_cell_weights, scales,
|
|
|
+ accum_scratch_ptr);
|
|
|
+ // Update the cell state.
|
|
|
+ UpdateLstmCellFloat(n_batch, n_cell, cell_state_ptr, input_gate_scratch,
|
|
|
+ forget_gate_scratch, cell_gate_scratch, use_cifg,
|
|
|
+ params->cell_clip);
|
|
|
+ // Calculate the output gate.
|
|
|
+ CalculateLstmGateHybrid(
|
|
|
+ quantized_input_ptr, input_sf, input_zp, input_to_output_weights_ptr,
|
|
|
+ input_to_output_weights_ledger_ptr, input_to_output_weights_scale,
|
|
|
+ input_to_output_row_sums, quantized_aux_input_ptr, aux_input_sf,
|
|
|
+ aux_input_zp, aux_input_to_output_weights_ptr,
|
|
|
+ aux_input_to_output_weights_scale, aux_input_to_output_row_sums,
|
|
|
+ quantized_output_state_ptr, output_state_sf, output_state_zp,
|
|
|
+ recurrent_to_output_weights_ptr, recurrent_to_output_weights_ledger_ptr,
|
|
|
+ recurrent_to_output_weights_scale, recurrent_to_output_row_sums,
|
|
|
+ cell_state_ptr, cell_to_output_weights_ptr, cell_to_output_weights_scale,
|
|
|
+ output_layer_norm_coefficients_ptr, output_gate_bias_ptr, n_batch,
|
|
|
+ n_input, n_aux_input, n_output, n_cell, kTfLiteActSigmoid,
|
|
|
+ output_gate_scratch, is_input_all_zeros, is_aux_input_all_zeros,
|
|
|
+ is_output_state_all_zeros, compute_row_sums, scaling_factors_scratch,
|
|
|
+ recovered_cell_weights, scales, accum_scratch_ptr);
|
|
|
+ // Update the output state.
|
|
|
+ CalculateLstmOutputHybrid(
|
|
|
+ n_batch, n_cell, n_output, cell_state_ptr, output_gate_scratch,
|
|
|
+ params->activation, projection_weights_ptr, projection_weights_ledger_ptr,
|
|
|
+ projection_weights_scale, projection_bias_ptr, params->proj_clip,
|
|
|
+ output_state_ptr, asymmetric_quantize_inputs, projection_weights_row_sums,
|
|
|
+ compute_row_sums, scratch2, quantized_output_scratch, input_sf, input_zp,
|
|
|
+ accum_scratch_ptr, scales);
|
|
|
+ // Copy output state to the output. Note that the output's rows may not be
|
|
|
+ // contiguous (output_batch_leading_dim != n_output).
|
|
|
+ for (int b = 0; b < n_batch; b++) {
|
|
|
+ std::memcpy(output_ptr + b * output_batch_leading_dim,
|
|
|
+ output_state_ptr + b * n_output, n_output * sizeof(float));
|
|
|
+ }
|
|
|
+}
|
|
|
+
|
|
|
+// Fully quantized lstm kernel for 16 bit gate matmul output.
|
|
|
+//
|
|
|
+// Input tensor of size n_batch * n_input:
|
|
|
+// input_ptr
|
|
|
+//
|
|
|
+// LSTM weights:
|
|
|
+// Quantized input weights of size 'n_cell * n_input':
|
|
|
+// input_to_input_weight_ptr - optional
|
|
|
+// input_to_forget_weight_ptr - optional
|
|
|
+// input_to_cell_weight_ptr - optional
|
|
|
+// input_to_output_weight_ptr - optional
|
|
|
+//
|
|
|
+// Quantized recurrent weights of size 'n_cell * n_output':
|
|
|
+// recurrent_to_input_weight_ptr - optional
|
|
|
+// recurrent_to_forget_weights_ptr
|
|
|
+// recurrent_to_cell_weights_ptr
|
|
|
+// recurrent_to_input_weights_ptr
|
|
|
+//
|
|
|
+// Quantized peephole weights of size 'n_cell', representing diagonal matrices.
|
|
|
+// cell_to_input_weights - optional
|
|
|
+// cell_to_cell_weights - optional
|
|
|
+// cell_to_output_weights - optional
|
|
|
+//
|
|
|
+// Quantized projection weights of size 'n_output * n_cell'
|
|
|
+// projection_weight_ptr - optional
|
|
|
+//
|
|
|
+// Weight scales (scalars) for each of the weights above.
|
|
|
+// effective_input_to_input_scale_a - optional
|
|
|
+// effective_input_to_input_scale_b - optional
|
|
|
+// effective_input_to_forget_scale_a
|
|
|
+// effective_input_to_forget_scale_b
|
|
|
+// effective_input_to_cell_scale_a
|
|
|
+// effective_input_to_cell_scale_b
|
|
|
+// effective_input_to_output_scale_a
|
|
|
+// effective_input_to_output_scale_b
|
|
|
+// effective_recurrent_to_input_scale_a - optional
|
|
|
+// effective_recurrent_to_input_scale_b - optional
|
|
|
+// effective_recurrent_to_forget_scale_a
|
|
|
+// effective_recurrent_to_forget_scale_b
|
|
|
+// effective_recurrent_to_cell_scale_a
|
|
|
+// effective_recurrent_to_cell_scale_b
|
|
|
+// effective_recurrent_to_output_scale_a
|
|
|
+// effective_recurrent_to_output_scale_b
|
|
|
+// effective_proj_scale_a - optional
|
|
|
+// effective_proj_scale_b - optional
|
|
|
+//
|
|
|
+// Gate biases of size 'n_cell':
|
|
|
+// input_gate_bias_ptr - optional
|
|
|
+// forget_gate_bias_ptr
|
|
|
+// cell_gate_bias_ptr
|
|
|
+// output_gate_bias_ptr
|
|
|
+//
|
|
|
+// Layer norm coefficients of size 'n_cell', representing diagonal matrices.
|
|
|
+// layer_norm_input_weight_ptr - optional
|
|
|
+// layer_norm_forget_weight_ptr - optional
|
|
|
+// layer_norm_cell_weight_ptr - optional
|
|
|
+// layer_norm_output_weight_ptr - optional
|
|
|
+//
|
|
|
+// Layer norm scales of size 'n_cell'.
|
|
|
+// layer_norm_input_scale_a - optional
|
|
|
+// layer_norm_input_scale_b - optional
|
|
|
+// layer_norm_forget_scale_a - optional
|
|
|
+// layer_norm_forget_scale_b - optional
|
|
|
+// layer_norm_cell_scale_a - optional
|
|
|
+// layer_norm_cell_scale_b - optional
|
|
|
+// layer_norm_output_scale_a - optional
|
|
|
+// layer_norm_output_scale_b - optional
|
|
|
+//
|
|
|
+// Scalar values:
|
|
|
+// quantized_cell_clip: quantized clip value for cell.
|
|
|
+// quantized_proj_clip: quantized clip value for projection.
|
|
|
+// cell_state_scale: the power of two scale for cell state.
|
|
|
+//
|
|
|
+// Zero points:
|
|
|
+// output_state_zp: zero point of output state
|
|
|
+// hidden_zp: zero point for hidden state.
|
|
|
+//
|
|
|
+// Temporary pre-allocated storage for the calculation. Each is of size n_cell *
|
|
|
+// n_batch.
|
|
|
+// scratch0
|
|
|
+// scratch1
|
|
|
+// scratch2
|
|
|
+// scratch3
|
|
|
+// scratch4
|
|
|
+// scratch5: this scratch buffer is created purely for optimizing the
|
|
|
+// MatrixBatchVectorMultiplyAccumulate.
|
|
|
+//
|
|
|
+// Outputs:
|
|
|
+// output_state_ptr - size 'n_batch * n_output'
|
|
|
+// cell_state_ptr - size 'n_batch * n_cell'
|
|
|
+// output_ptr - size 'n_batch * n_output'
|
|
|
+// TODO(b/159947023): scratch0 is not used if (!cifg). Don't allocate then.
|
|
|
+inline void LstmStepInteger8x8_16(
|
|
|
+ const int8_t* input_ptr, const int8_t* input_to_input_weight_ptr,
|
|
|
+ int32_t effective_input_to_input_scale_a,
|
|
|
+ int32_t effective_input_to_input_scale_b,
|
|
|
+ const int8_t* input_to_forget_weight_ptr,
|
|
|
+ int32_t effective_input_to_forget_scale_a,
|
|
|
+ int32_t effective_input_to_forget_scale_b,
|
|
|
+ const int8_t* input_to_cell_weight_ptr,
|
|
|
+ int32_t effective_input_to_cell_scale_a,
|
|
|
+ int32_t effective_input_to_cell_scale_b,
|
|
|
+ const int8_t* input_to_output_weight_ptr,
|
|
|
+ int32_t effective_input_to_output_scale_a,
|
|
|
+ int32_t effective_input_to_output_scale_b,
|
|
|
+ const int8_t* recurrent_to_input_weight_ptr,
|
|
|
+ int32_t effective_recurrent_to_input_scale_a,
|
|
|
+ int32_t effective_recurrent_to_input_scale_b,
|
|
|
+ const int8_t* recurrent_to_forget_weight_ptr,
|
|
|
+ int32_t effective_recurrent_to_forget_scale_a,
|
|
|
+ int32_t effective_recurrent_to_forget_scale_b,
|
|
|
+ const int8_t* recurrent_to_cell_weight_ptr,
|
|
|
+ int32_t effective_recurrent_to_cell_scale_a,
|
|
|
+ int32_t effective_recurrent_to_cell_scale_b,
|
|
|
+ const int8_t* recurrent_to_output_weight_ptr,
|
|
|
+ int32_t effective_recurrent_to_output_scale_a,
|
|
|
+ int32_t effective_recurrent_to_output_scale_b,
|
|
|
+ const int16_t* cell_to_input_weight_ptr,
|
|
|
+ int32_t effective_cell_to_input_scale_a,
|
|
|
+ int32_t effective_cell_to_input_scale_b,
|
|
|
+ const int16_t* cell_to_forget_weight_ptr,
|
|
|
+ int32_t effective_cell_to_forget_scale_a,
|
|
|
+ int32_t effective_cell_to_forget_scale_b,
|
|
|
+ const int16_t* cell_to_output_weight_ptr,
|
|
|
+ int32_t effective_cell_to_output_scale_a,
|
|
|
+ int32_t effective_cell_to_output_scale_b,
|
|
|
+ const int8_t* projection_weight_ptr, int32_t effective_proj_scale_a,
|
|
|
+ int32_t effective_proj_scale_b, int32_t hidden_zp,
|
|
|
+ int32_t effective_hidden_scale_a, int32_t effective_hidden_scale_b,
|
|
|
+ const int16_t* layer_norm_input_weight_ptr,
|
|
|
+ int32_t layer_norm_input_scale_a, int32_t layer_norm_input_scale_b,
|
|
|
+ const int16_t* layer_norm_forget_weight_ptr,
|
|
|
+ int32_t layer_norm_forget_scale_a, int32_t layer_norm_forget_scale_b,
|
|
|
+ const int16_t* layer_norm_cell_weight_ptr, int32_t layer_norm_cell_scale_a,
|
|
|
+ int32_t layer_norm_cell_scale_b,
|
|
|
+ const int16_t* layer_norm_output_weight_ptr,
|
|
|
+ int32_t layer_norm_output_scale_a, int32_t layer_norm_output_scale_b,
|
|
|
+ const int32_t* input_gate_bias_ptr, const int32_t* forget_gate_bias_ptr,
|
|
|
+ const int32_t* cell_gate_bias_ptr, const int32_t* output_gate_bias_ptr,
|
|
|
+ int16_t quantized_cell_clip, int8_t quantized_proj_clip,
|
|
|
+ int32_t cell_state_scale, int32_t input_variance_guard,
|
|
|
+ int32_t forget_variance_guard, int32_t cell_variance_guard,
|
|
|
+ int32_t output_variance_guard,
|
|
|
+ const int32_t* input_to_forget_effective_bias,
|
|
|
+ const int32_t* recurrent_to_forget_effective_bias,
|
|
|
+ const int32_t* input_to_cell_effective_bias,
|
|
|
+ const int32_t* recurrent_to_cell_effective_bias,
|
|
|
+ const int32_t* input_to_output_effective_bias,
|
|
|
+ const int32_t* recurrent_to_output_effective_bias,
|
|
|
+ const int32_t* input_to_input_effective_bias,
|
|
|
+ const int32_t* recurrent_to_input_effective_bias,
|
|
|
+ const int32_t* projection_effective_bias, int n_batch, int n_cell,
|
|
|
+ int n_input, int n_output, int8_t* output_state_ptr,
|
|
|
+ int32_t output_state_zp, int16_t* cell_state_ptr, int8_t* output_ptr,
|
|
|
+ int16_t* scratch0, int16_t* scratch1, int16_t* scratch2, int16_t* scratch3,
|
|
|
+ int8_t* scratch4, int32_t* scratch5) {
|
|
|
+ // Make named scratch buffers for the different gates.
|
|
|
+ int16_t* input_gate_scratch = scratch0;
|
|
|
+ int16_t* forget_gate_scratch = scratch1;
|
|
|
+ int16_t* cell_gate_scratch = scratch2;
|
|
|
+ int16_t* output_gate_scratch = scratch3;
|
|
|
+
|
|
|
+ // Since we have already checked that weights are all there or none, we
|
|
|
+ // can check the existence of only one to the get the condition.
|
|
|
+ const bool use_cifg = (input_to_input_weight_ptr == nullptr);
|
|
|
+
|
|
|
+ // Check for nullptrs.
|
|
|
+ TFLITE_DCHECK(input_to_forget_effective_bias);
|
|
|
+ TFLITE_DCHECK(recurrent_to_forget_effective_bias);
|
|
|
+ TFLITE_DCHECK(input_to_cell_effective_bias);
|
|
|
+ TFLITE_DCHECK(recurrent_to_cell_effective_bias);
|
|
|
+ TFLITE_DCHECK(input_to_output_effective_bias);
|
|
|
+ TFLITE_DCHECK(recurrent_to_output_effective_bias);
|
|
|
+ if (!use_cifg) {
|
|
|
+ TFLITE_DCHECK(input_to_input_effective_bias);
|
|
|
+ TFLITE_DCHECK(recurrent_to_input_effective_bias);
|
|
|
+ }
|
|
|
+ const bool use_projection = (projection_weight_ptr != nullptr);
|
|
|
+ if (use_projection) {
|
|
|
+ TFLITE_DCHECK(projection_effective_bias);
|
|
|
+ }
|
|
|
+ if (!use_cifg) {
|
|
|
+ // Calculate the input gate. (If not CIFG.)
|
|
|
+ CalculateLstmGateInteger8x8_16(
|
|
|
+ input_ptr, input_to_input_weight_ptr, input_to_input_effective_bias,
|
|
|
+ effective_input_to_input_scale_a, effective_input_to_input_scale_b,
|
|
|
+ output_state_ptr, recurrent_to_input_weight_ptr,
|
|
|
+ recurrent_to_input_effective_bias, effective_recurrent_to_input_scale_a,
|
|
|
+ effective_recurrent_to_input_scale_b, cell_state_ptr,
|
|
|
+ cell_to_input_weight_ptr, effective_cell_to_input_scale_a,
|
|
|
+ effective_cell_to_input_scale_b, layer_norm_input_weight_ptr,
|
|
|
+ input_gate_bias_ptr, layer_norm_input_scale_a, layer_norm_input_scale_b,
|
|
|
+ input_variance_guard, n_batch, n_input, n_output, n_cell,
|
|
|
+ kTfLiteActSigmoid, input_gate_scratch, scratch5);
|
|
|
+ }
|
|
|
+ // Calculate the forget gate.
|
|
|
+ CalculateLstmGateInteger8x8_16(
|
|
|
+ input_ptr, input_to_forget_weight_ptr, input_to_forget_effective_bias,
|
|
|
+ effective_input_to_forget_scale_a, effective_input_to_forget_scale_b,
|
|
|
+ output_state_ptr, recurrent_to_forget_weight_ptr,
|
|
|
+ recurrent_to_forget_effective_bias, effective_recurrent_to_forget_scale_a,
|
|
|
+ effective_recurrent_to_forget_scale_b, cell_state_ptr,
|
|
|
+ cell_to_forget_weight_ptr, effective_cell_to_forget_scale_a,
|
|
|
+ effective_cell_to_forget_scale_b, layer_norm_forget_weight_ptr,
|
|
|
+ forget_gate_bias_ptr, layer_norm_forget_scale_a,
|
|
|
+ layer_norm_forget_scale_b, forget_variance_guard, n_batch, n_input,
|
|
|
+ n_output, n_cell, kTfLiteActSigmoid, forget_gate_scratch, scratch5);
|
|
|
+ // Calculate the cell update gate.
|
|
|
+ CalculateLstmGateInteger8x8_16(
|
|
|
+ input_ptr, input_to_cell_weight_ptr, input_to_cell_effective_bias,
|
|
|
+ effective_input_to_cell_scale_a, effective_input_to_cell_scale_b,
|
|
|
+ output_state_ptr, recurrent_to_cell_weight_ptr,
|
|
|
+ recurrent_to_cell_effective_bias, effective_recurrent_to_cell_scale_a,
|
|
|
+ effective_recurrent_to_cell_scale_b, cell_state_ptr,
|
|
|
+ /*cell_to_gate_weights=*/nullptr, /*cell_to_gate_scale_a=*/0,
|
|
|
+ /*cell_to_gate_scale_b=*/0, layer_norm_cell_weight_ptr,
|
|
|
+ cell_gate_bias_ptr, layer_norm_cell_scale_a, layer_norm_cell_scale_b,
|
|
|
+ cell_variance_guard, n_batch, n_input, n_output, n_cell, kTfLiteActTanh,
|
|
|
+ cell_gate_scratch, scratch5);
|
|
|
+ // Update the cell state.
|
|
|
+ UpdateLstmCellInteger(n_batch, n_cell, cell_state_ptr, cell_state_scale,
|
|
|
+ input_gate_scratch, forget_gate_scratch,
|
|
|
+ cell_gate_scratch, use_cifg, quantized_cell_clip);
|
|
|
+ // Calculate the output gate.
|
|
|
+ CalculateLstmGateInteger8x8_16(
|
|
|
+ input_ptr, input_to_output_weight_ptr, input_to_output_effective_bias,
|
|
|
+ effective_input_to_output_scale_a, effective_input_to_output_scale_b,
|
|
|
+ output_state_ptr, recurrent_to_output_weight_ptr,
|
|
|
+ recurrent_to_output_effective_bias, effective_recurrent_to_output_scale_a,
|
|
|
+ effective_recurrent_to_output_scale_b, cell_state_ptr,
|
|
|
+ cell_to_output_weight_ptr, effective_cell_to_output_scale_a,
|
|
|
+ effective_cell_to_output_scale_b, layer_norm_output_weight_ptr,
|
|
|
+ output_gate_bias_ptr, layer_norm_output_scale_a,
|
|
|
+ layer_norm_output_scale_b, output_variance_guard, n_batch, n_input,
|
|
|
+ n_output, n_cell, kTfLiteActSigmoid, output_gate_scratch, scratch5);
|
|
|
+ // Update the output state.
|
|
|
+ CalculateLstmOutputInteger8x8_16(
|
|
|
+ n_batch, n_cell, n_output, cell_state_ptr, cell_state_scale,
|
|
|
+ output_gate_scratch, effective_hidden_scale_a, effective_hidden_scale_b,
|
|
|
+ hidden_zp, projection_weight_ptr, effective_proj_scale_a,
|
|
|
+ effective_proj_scale_b, projection_effective_bias, output_state_zp,
|
|
|
+ quantized_proj_clip, output_state_ptr, scratch0, scratch4, scratch5);
|
|
|
+ // Copy output state to the output. Note that unlike float or hybrid, output
|
|
|
+ // is always contiguous.
|
|
|
+ std::memcpy(output_ptr, output_state_ptr,
|
|
|
+ n_batch * n_output * sizeof(int8_t));
|
|
|
+}
|
|
|
+
|
|
|
+// Fully quantized lstm kernel for 8 bit gate matmul output.
|
|
|
+//
|
|
|
+// Input tensor of size n_batch * n_input:
|
|
|
+// input_ptr
|
|
|
+//
|
|
|
+// LSTM weights:
|
|
|
+// Quantized input weights of size 'n_cell * n_input':
|
|
|
+// input_to_input_weight_ptr - optional
|
|
|
+// input_to_forget_weight_ptr - optional
|
|
|
+// input_to_cell_weight_ptr - optional
|
|
|
+// input_to_output_weight_ptr - optional
|
|
|
+//
|
|
|
+// Quantized recurrent weights of size 'n_cell * n_output':
|
|
|
+// recurrent_to_input_weight_ptr - optional
|
|
|
+// recurrent_to_forget_weights_ptr
|
|
|
+// recurrent_to_cell_weights_ptr
|
|
|
+// recurrent_to_input_weights_ptr
|
|
|
+//
|
|
|
+// Quantized peephole weights of size 'n_cell', representing diagonal matrices.
|
|
|
+// cell_to_input_weights - optional
|
|
|
+// cell_to_cell_weights - optional
|
|
|
+// cell_to_output_weights - optional
|
|
|
+//
|
|
|
+// Quantized projection weights of size 'n_output * n_cell'
|
|
|
+// projection_weight_ptr - optional
|
|
|
+//
|
|
|
+// Weight scales (scalars) for each of the weights above.
|
|
|
+// effective_input_to_input_scale_a - optional
|
|
|
+// effective_input_to_input_scale_b - optional
|
|
|
+// effective_input_to_forget_scale_a
|
|
|
+// effective_input_to_forget_scale_b
|
|
|
+// effective_input_to_cell_scale_a
|
|
|
+// effective_input_to_cell_scale_b
|
|
|
+// effective_input_to_output_scale_a
|
|
|
+// effective_input_to_output_scale_b
|
|
|
+// effective_recurrent_to_input_scale_a - optional
|
|
|
+// effective_recurrent_to_input_scale_b - optional
|
|
|
+// effective_recurrent_to_forget_scale_a
|
|
|
+// effective_recurrent_to_forget_scale_b
|
|
|
+// effective_recurrent_to_cell_scale_a
|
|
|
+// effective_recurrent_to_cell_scale_b
|
|
|
+// effective_recurrent_to_output_scale_a
|
|
|
+// effective_recurrent_to_output_scale_b
|
|
|
+// effective_proj_scale_a - optional
|
|
|
+// effective_proj_scale_b - optional
|
|
|
+//
|
|
|
+// Gate biases of size 'n_cell':
|
|
|
+// input_gate_bias_ptr - optional
|
|
|
+// forget_gate_bias_ptr
|
|
|
+// cell_gate_bias_ptr
|
|
|
+// output_gate_bias_ptr
|
|
|
+//
|
|
|
+// Layer norm coefficients of size 'n_cell', representing diagonal matrices.
|
|
|
+// layer_norm_input_weight_ptr - optional
|
|
|
+// layer_norm_forget_weight_ptr - optional
|
|
|
+// layer_norm_cell_weight_ptr - optional
|
|
|
+// layer_norm_output_weight_ptr - optional
|
|
|
+//
|
|
|
+// Layer norm scales of size 'n_cell'.
|
|
|
+// layer_norm_input_scale_a - optional
|
|
|
+// layer_norm_input_scale_b - optional
|
|
|
+// layer_norm_forget_scale_a - optional
|
|
|
+// layer_norm_forget_scale_b - optional
|
|
|
+// layer_norm_cell_scale_a - optional
|
|
|
+// layer_norm_cell_scale_b - optional
|
|
|
+// layer_norm_output_scale_a - optional
|
|
|
+// layer_norm_output_scale_b - optional
|
|
|
+//
|
|
|
+// Scalar values:
|
|
|
+// quantized_cell_clip: quantized clip value for cell.
|
|
|
+// quantized_proj_clip: quantized clip value for projection.
|
|
|
+// cell_state_scale: the power of two scale for cell state.
|
|
|
+//
|
|
|
+// Zero points:
|
|
|
+// input_zp: zero point for input tensor.
|
|
|
+// output_state_zp: zero point of output state.
|
|
|
+// hidden_zp: zero point for hidden state.
|
|
|
+//
|
|
|
+// Temporary pre-allocated storage for the calculation. Each is of size n_cell *
|
|
|
+// n_batch.
|
|
|
+// scratch0
|
|
|
+// scratch1
|
|
|
+// scratch2
|
|
|
+// scratch3
|
|
|
+// scratch4
|
|
|
+// scratch5
|
|
|
+// scratch6
|
|
|
+// scratch7
|
|
|
+//
|
|
|
+// Outputs:
|
|
|
+// output_state_ptr - size 'n_batch * n_output'
|
|
|
+// cell_state_ptr - size 'n_batch * n_cell'
|
|
|
+// output_ptr - size 'n_batch * n_output'
|
|
|
+//
|
|
|
+// Can move zero point calculation into Prepare() for better perfomance.
|
|
|
+// TODO(b/159947023): scratch5 is unused, remove.
|
|
|
+inline void LstmStepInteger8x8_8(
|
|
|
+ const int8_t* input_ptr, int32_t input_zp,
|
|
|
+ const int8_t* input_to_input_weight_ptr,
|
|
|
+ int32_t effective_input_to_input_scale_a,
|
|
|
+ int32_t effective_input_to_input_scale_b,
|
|
|
+ const int8_t* input_to_forget_weight_ptr,
|
|
|
+ int32_t effective_input_to_forget_scale_a,
|
|
|
+ int32_t effective_input_to_forget_scale_b,
|
|
|
+ const int8_t* input_to_cell_weight_ptr,
|
|
|
+ int32_t effective_input_to_cell_scale_a,
|
|
|
+ int32_t effective_input_to_cell_scale_b,
|
|
|
+ const int8_t* input_to_output_weight_ptr,
|
|
|
+ int32_t effective_input_to_output_scale_a,
|
|
|
+ int32_t effective_input_to_output_scale_b,
|
|
|
+ const int8_t* recurrent_to_input_weight_ptr,
|
|
|
+ int32_t effective_recurrent_to_input_scale_a,
|
|
|
+ int32_t effective_recurrent_to_input_scale_b,
|
|
|
+ const int8_t* recurrent_to_forget_weight_ptr,
|
|
|
+ int32_t effective_recurrent_to_forget_scale_a,
|
|
|
+ int32_t effective_recurrent_to_forget_scale_b,
|
|
|
+ const int8_t* recurrent_to_cell_weight_ptr,
|
|
|
+ int32_t effective_recurrent_to_cell_scale_a,
|
|
|
+ int32_t effective_recurrent_to_cell_scale_b,
|
|
|
+ const int8_t* recurrent_to_output_weight_ptr,
|
|
|
+ int32_t effective_recurrent_to_output_scale_a,
|
|
|
+ int32_t effective_recurrent_to_output_scale_b,
|
|
|
+ const int8_t* cell_to_input_weight_ptr,
|
|
|
+ int32_t effective_cell_to_input_scale_a,
|
|
|
+ int32_t effective_cell_to_input_scale_b,
|
|
|
+ const int8_t* cell_to_forget_weight_ptr,
|
|
|
+ int32_t effective_cell_to_forget_scale_a,
|
|
|
+ int32_t effective_cell_to_forget_scale_b,
|
|
|
+ const int8_t* cell_to_output_weight_ptr,
|
|
|
+ int32_t effective_cell_to_output_scale_a,
|
|
|
+ int32_t effective_cell_to_output_scale_b,
|
|
|
+ const int8_t* projection_weight_ptr, int32_t effective_proj_scale_a,
|
|
|
+ int32_t effective_proj_scale_b, const int16_t* layer_norm_input_weight_ptr,
|
|
|
+ int32_t layer_norm_input_scale_a, int32_t layer_norm_input_scale_b,
|
|
|
+ const int16_t* layer_norm_forget_weight_ptr,
|
|
|
+ int32_t layer_norm_forget_scale_a, int32_t layer_norm_forget_scale_b,
|
|
|
+ const int16_t* layer_norm_cell_weight_ptr, int32_t layer_norm_cell_scale_a,
|
|
|
+ int32_t layer_norm_cell_scale_b,
|
|
|
+ const int16_t* layer_norm_output_weight_ptr,
|
|
|
+ int32_t layer_norm_output_scale_a, int32_t layer_norm_output_scale_b,
|
|
|
+ const int32_t* input_gate_bias_ptr, const int32_t* forget_gate_bias_ptr,
|
|
|
+ const int32_t* cell_gate_bias_ptr, const int32_t* output_gate_bias_ptr,
|
|
|
+ const int32_t* projection_bias_ptr, const TfLiteLSTMParams* params,
|
|
|
+ const int32_t* intermediate_scale_a, const int32_t* intermediate_scale_b,
|
|
|
+ const int32_t* intermediate_zp, int16_t quantized_cell_clip,
|
|
|
+ int8_t quantized_proj_clip, int n_batch, int n_cell, int n_input,
|
|
|
+ int n_output, int output_batch_leading_dim, int8_t* output_state_ptr,
|
|
|
+ int32_t output_state_zp, int16_t* cell_state_ptr, int8_t* output_ptr,
|
|
|
+ int8_t* scratch0, int8_t* scratch1, int16_t* scratch2, int16_t* scratch3,
|
|
|
+ int16_t* scratch4, int16_t* scratch5, int16_t* scratch6,
|
|
|
+ int16_t* scratch7) {
|
|
|
+ // TODO(b/159066113): scratch5 is unused, remove.
|
|
|
+
|
|
|
+ // Make named scratch buffers for the different gates.
|
|
|
+ int16_t* forget_gate_scratch = scratch2;
|
|
|
+ int16_t* cell_gate_scratch = scratch3;
|
|
|
+ int16_t* output_gate_scratch = scratch4;
|
|
|
+ // no-CIFG is not supported here
|
|
|
+
|
|
|
+ // Calculate the forget gate.
|
|
|
+ CalculateLstmGateInteger8x8_8(
|
|
|
+ input_ptr, input_zp, input_to_forget_weight_ptr,
|
|
|
+ effective_input_to_forget_scale_a, effective_input_to_forget_scale_b,
|
|
|
+ intermediate_scale_a[2], intermediate_scale_b[2], intermediate_zp[4],
|
|
|
+ output_state_ptr, output_state_zp, recurrent_to_forget_weight_ptr,
|
|
|
+ effective_recurrent_to_forget_scale_a,
|
|
|
+ effective_recurrent_to_forget_scale_b, intermediate_scale_a[3],
|
|
|
+ intermediate_scale_b[3], intermediate_zp[5], layer_norm_forget_weight_ptr,
|
|
|
+ layer_norm_forget_scale_a, layer_norm_forget_scale_b,
|
|
|
+ forget_gate_bias_ptr, n_batch, n_input, n_output, n_cell,
|
|
|
+ kTfLiteActSigmoid, forget_gate_scratch, scratch0, scratch1);
|
|
|
+ // Calculate the cell update gate.
|
|
|
+ CalculateLstmGateInteger8x8_8(
|
|
|
+ input_ptr, input_zp, input_to_cell_weight_ptr,
|
|
|
+ effective_input_to_cell_scale_a, effective_input_to_cell_scale_b,
|
|
|
+ intermediate_scale_a[4], intermediate_scale_b[4], intermediate_zp[7],
|
|
|
+ output_state_ptr, output_state_zp, recurrent_to_cell_weight_ptr,
|
|
|
+ effective_recurrent_to_cell_scale_a, effective_recurrent_to_cell_scale_b,
|
|
|
+ intermediate_scale_a[5], intermediate_scale_b[5], intermediate_zp[8],
|
|
|
+ layer_norm_cell_weight_ptr, layer_norm_cell_scale_a,
|
|
|
+ layer_norm_cell_scale_b, cell_gate_bias_ptr, n_batch, n_input, n_output,
|
|
|
+ n_cell, kTfLiteActTanh, cell_gate_scratch, scratch0, scratch1);
|
|
|
+ // Update the cell state.
|
|
|
+ UpdateLstmCellInteger(n_batch, n_cell, cell_state_ptr,
|
|
|
+ /*cell_state_scale=*/-15, /*input_gate=*/nullptr,
|
|
|
+ forget_gate_scratch, cell_gate_scratch,
|
|
|
+ /*use_cifg=*/true, quantized_cell_clip);
|
|
|
+ // Calculate the output gate.
|
|
|
+ CalculateLstmGateInteger8x8_8(
|
|
|
+ input_ptr, input_zp, input_to_output_weight_ptr,
|
|
|
+ effective_input_to_output_scale_a, effective_input_to_output_scale_b,
|
|
|
+ intermediate_scale_a[6], intermediate_scale_b[6], intermediate_zp[10],
|
|
|
+ output_state_ptr, output_state_zp, recurrent_to_output_weight_ptr,
|
|
|
+ effective_recurrent_to_output_scale_a,
|
|
|
+ effective_recurrent_to_output_scale_b, intermediate_scale_a[11],
|
|
|
+ intermediate_scale_b[7], intermediate_zp[7], layer_norm_output_weight_ptr,
|
|
|
+ layer_norm_output_scale_a, layer_norm_output_scale_b,
|
|
|
+ output_gate_bias_ptr, n_batch, n_input, n_output, n_cell,
|
|
|
+ kTfLiteActSigmoid, output_gate_scratch, scratch0, scratch1);
|
|
|
+ // Update the output state.
|
|
|
+ CalculateLstmOutputInteger8x8_8(
|
|
|
+ n_batch, n_cell, n_output, cell_state_ptr, output_gate_scratch,
|
|
|
+ projection_weight_ptr, effective_proj_scale_a, effective_proj_scale_b,
|
|
|
+ projection_bias_ptr, output_state_zp, quantized_proj_clip,
|
|
|
+ output_state_ptr, scratch2);
|
|
|
+ // Copy output state to the output. Note that unlike float or hybrid, output
|
|
|
+ // is always contigous.
|
|
|
+ std::memcpy(output_ptr, output_state_ptr,
|
|
|
+ n_batch * n_output * sizeof(int8_t));
|
|
|
+}
|
|
|
+
|
|
|
+} // namespace
|
|
|
+
|
|
|
+TfLiteStatus EvalFloatLstm(
|
|
|
+ const TfLiteEvalTensor* input,
|
|
|
+ const TfLiteEvalTensor* input_to_input_weights,
|
|
|
+ const TfLiteEvalTensor* input_to_forget_weights,
|
|
|
+ const TfLiteEvalTensor* input_to_cell_weights,
|
|
|
+ const TfLiteEvalTensor* input_to_output_weights,
|
|
|
+ const TfLiteEvalTensor* recurrent_to_input_weights,
|
|
|
+ const TfLiteEvalTensor* recurrent_to_forget_weights,
|
|
|
+ const TfLiteEvalTensor* recurrent_to_cell_weights,
|
|
|
+ const TfLiteEvalTensor* recurrent_to_output_weights,
|
|
|
+ const TfLiteEvalTensor* cell_to_input_weights,
|
|
|
+ const TfLiteEvalTensor* cell_to_forget_weights,
|
|
|
+ const TfLiteEvalTensor* cell_to_output_weights,
|
|
|
+ const TfLiteEvalTensor* input_layer_norm_coefficients,
|
|
|
+ const TfLiteEvalTensor* forget_layer_norm_coefficients,
|
|
|
+ const TfLiteEvalTensor* cell_layer_norm_coefficients,
|
|
|
+ const TfLiteEvalTensor* output_layer_norm_coefficients,
|
|
|
+ const TfLiteEvalTensor* aux_input,
|
|
|
+ const TfLiteEvalTensor* aux_input_to_input_weights,
|
|
|
+ const TfLiteEvalTensor* aux_input_to_forget_weights,
|
|
|
+ const TfLiteEvalTensor* aux_input_to_cell_weights,
|
|
|
+ const TfLiteEvalTensor* aux_input_to_output_weights,
|
|
|
+ const TfLiteEvalTensor* input_gate_bias,
|
|
|
+ const TfLiteEvalTensor* forget_gate_bias,
|
|
|
+ const TfLiteEvalTensor* cell_gate_bias,
|
|
|
+ const TfLiteEvalTensor* output_gate_bias,
|
|
|
+ const TfLiteEvalTensor* projection_weights,
|
|
|
+ const TfLiteEvalTensor* projection_bias, const TfLiteLSTMParams* params,
|
|
|
+ bool forward_sequence, bool time_major, int output_offset,
|
|
|
+ float* scratch_buffer, TfLiteEvalTensor* output_state,
|
|
|
+ TfLiteEvalTensor* cell_state, TfLiteEvalTensor* output) {
|
|
|
+ TFLITE_DCHECK(input->dims->size >= 2 && input->dims->size <= 3);
|
|
|
+ int max_time, n_batch;
|
|
|
+ if (input->dims->size == 3) {
|
|
|
+ max_time = (time_major) ? input->dims->data[0] : input->dims->data[1];
|
|
|
+ n_batch = (time_major) ? input->dims->data[1] : input->dims->data[0];
|
|
|
+ } else {
|
|
|
+ max_time = 1;
|
|
|
+ n_batch = input->dims->data[0];
|
|
|
+ }
|
|
|
+ const int n_input = input->dims->data[input->dims->size - 1];
|
|
|
+ const int aux_input_size =
|
|
|
+ (aux_input) ? aux_input->dims->data[aux_input->dims->size - 1] : 0;
|
|
|
+
|
|
|
+ // n_cell and n_output will be the same size when there is no projection.
|
|
|
+ const int n_cell = input_to_output_weights->dims->data[0];
|
|
|
+ const int n_output = recurrent_to_output_weights->dims->data[1];
|
|
|
+
|
|
|
+ // Since we have already checked that weights are all there or none, we can
|
|
|
+ // check the existence of only one to the get the condition.
|
|
|
+ const bool use_cifg = (input_to_input_weights == nullptr);
|
|
|
+
|
|
|
+ // Index the scratch buffers pointers to the global scratch buffer.
|
|
|
+ float* input_gate_scratch = nullptr;
|
|
|
+ float* cell_gate_scratch = nullptr;
|
|
|
+ float* forget_gate_scratch = nullptr;
|
|
|
+ float* output_gate_scratch = nullptr;
|
|
|
+ if (use_cifg) {
|
|
|
+ cell_gate_scratch = scratch_buffer;
|
|
|
+ forget_gate_scratch = scratch_buffer + n_cell * n_batch;
|
|
|
+ output_gate_scratch = scratch_buffer + 2 * n_cell * n_batch;
|
|
|
+ } else {
|
|
|
+ input_gate_scratch = scratch_buffer;
|
|
|
+ cell_gate_scratch = scratch_buffer + n_cell * n_batch;
|
|
|
+ forget_gate_scratch = scratch_buffer + 2 * n_cell * n_batch;
|
|
|
+ output_gate_scratch = scratch_buffer + 3 * n_cell * n_batch;
|
|
|
+ }
|
|
|
+
|
|
|
+ const int output_batch_leading_dim =
|
|
|
+ output->dims->data[output->dims->size - 1];
|
|
|
+ if (time_major) {
|
|
|
+ // Loop through the sequence.
|
|
|
+ const int input_step = n_batch * n_input;
|
|
|
+ const int output_step = n_batch * output_batch_leading_dim;
|
|
|
+ for (int t = 0; t < max_time; t++) {
|
|
|
+ // If this is the forward_sequence, step forward, otherwise step
|
|
|
+ // backwards.
|
|
|
+ const int t_rel = forward_sequence ? t : max_time - t - 1;
|
|
|
+ const float* input_ptr =
|
|
|
+ tflite::micro::GetTensorData<float>(input) + t_rel * input_step;
|
|
|
+ const float* aux_input_ptr = nullptr;
|
|
|
+ if (aux_input) {
|
|
|
+ aux_input_ptr =
|
|
|
+ tflite::micro::GetTensorData<float>(aux_input) + t_rel * input_step;
|
|
|
+ }
|
|
|
+ float* output_ptr = tflite::micro::GetTensorData<float>(output) +
|
|
|
+ t_rel * output_step + output_offset;
|
|
|
+
|
|
|
+ LstmStepFloat(
|
|
|
+ input_ptr,
|
|
|
+ input_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(input_to_input_weights),
|
|
|
+ input_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(input_to_forget_weights),
|
|
|
+ input_to_cell_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(input_to_cell_weights),
|
|
|
+ input_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(input_to_output_weights),
|
|
|
+ aux_input_ptr,
|
|
|
+ aux_input_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(aux_input_to_input_weights),
|
|
|
+ aux_input_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ aux_input_to_forget_weights),
|
|
|
+ aux_input_to_cell_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(aux_input_to_cell_weights),
|
|
|
+ aux_input_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ aux_input_to_output_weights),
|
|
|
+ recurrent_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(recurrent_to_input_weights),
|
|
|
+ recurrent_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ recurrent_to_forget_weights),
|
|
|
+ recurrent_to_cell_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(recurrent_to_cell_weights),
|
|
|
+ recurrent_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ recurrent_to_output_weights),
|
|
|
+ cell_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(cell_to_input_weights),
|
|
|
+ cell_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(cell_to_forget_weights),
|
|
|
+ cell_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(cell_to_output_weights),
|
|
|
+ input_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ input_layer_norm_coefficients),
|
|
|
+ forget_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ forget_layer_norm_coefficients),
|
|
|
+ cell_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ cell_layer_norm_coefficients),
|
|
|
+ output_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ output_layer_norm_coefficients),
|
|
|
+ input_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(input_gate_bias),
|
|
|
+ forget_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(forget_gate_bias),
|
|
|
+ cell_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(cell_gate_bias),
|
|
|
+ output_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(output_gate_bias),
|
|
|
+ projection_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(projection_weights),
|
|
|
+ projection_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(projection_bias),
|
|
|
+ params, n_batch, n_cell, n_input, aux_input_size, n_output,
|
|
|
+ output_batch_leading_dim,
|
|
|
+ tflite::micro::GetTensorData<float>(output_state),
|
|
|
+ tflite::micro::GetTensorData<float>(cell_state), input_gate_scratch,
|
|
|
+ forget_gate_scratch, cell_gate_scratch, output_gate_scratch,
|
|
|
+ output_ptr);
|
|
|
+ }
|
|
|
+ } else {
|
|
|
+ for (int b = 0; b < n_batch; b++) {
|
|
|
+ const int input_step = n_input;
|
|
|
+ const int output_step = output_batch_leading_dim;
|
|
|
+ for (int t = 0; t < max_time; t++) {
|
|
|
+ // If this is the forward_sequence, step forward, otherwise step
|
|
|
+ // backwards.
|
|
|
+ const int t_rel = forward_sequence ? t : max_time - t - 1;
|
|
|
+ const int time_offset = b * max_time + t_rel;
|
|
|
+ const float* input_ptr = tflite::micro::GetTensorData<float>(input) +
|
|
|
+ time_offset * input_step;
|
|
|
+ const float* aux_input_ptr = nullptr;
|
|
|
+ if (aux_input) {
|
|
|
+ aux_input_ptr = tflite::micro::GetTensorData<float>(aux_input) +
|
|
|
+ time_offset * input_step;
|
|
|
+ }
|
|
|
+ float* output_ptr = tflite::micro::GetTensorData<float>(output) +
|
|
|
+ time_offset * output_step + output_offset;
|
|
|
+
|
|
|
+ // Offset the {output,cell}_state pointers to the right batch.
|
|
|
+ float* output_state_ptr =
|
|
|
+ tflite::micro::GetTensorData<float>(output_state) +
|
|
|
+ b * output_batch_leading_dim;
|
|
|
+ float* cell_state_ptr =
|
|
|
+ tflite::micro::GetTensorData<float>(cell_state) + b * n_cell;
|
|
|
+ // Offset the scratch pointers to the right batch.
|
|
|
+ float* input_gate_scratch_ptr =
|
|
|
+ input_gate_scratch ? input_gate_scratch + b * n_cell : nullptr;
|
|
|
+ float* forget_gate_scratch_ptr = forget_gate_scratch + b * n_cell;
|
|
|
+ float* cell_gate_scratch_ptr = cell_gate_scratch + b * n_cell;
|
|
|
+ float* output_gate_scratch_ptr = output_gate_scratch + b * n_cell;
|
|
|
+
|
|
|
+ LstmStepFloat(
|
|
|
+ input_ptr,
|
|
|
+ input_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(input_to_input_weights),
|
|
|
+ input_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(input_to_forget_weights),
|
|
|
+ input_to_cell_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(input_to_cell_weights),
|
|
|
+ input_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(input_to_output_weights),
|
|
|
+ aux_input_ptr,
|
|
|
+ aux_input_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ aux_input_to_input_weights),
|
|
|
+ aux_input_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ aux_input_to_forget_weights),
|
|
|
+ aux_input_to_cell_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ aux_input_to_cell_weights),
|
|
|
+ aux_input_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ aux_input_to_output_weights),
|
|
|
+ recurrent_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ recurrent_to_input_weights),
|
|
|
+ recurrent_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ recurrent_to_forget_weights),
|
|
|
+ recurrent_to_cell_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ recurrent_to_cell_weights),
|
|
|
+ recurrent_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ recurrent_to_output_weights),
|
|
|
+ cell_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(cell_to_input_weights),
|
|
|
+ cell_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(cell_to_forget_weights),
|
|
|
+ cell_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(cell_to_output_weights),
|
|
|
+ input_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ input_layer_norm_coefficients),
|
|
|
+ forget_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ forget_layer_norm_coefficients),
|
|
|
+ cell_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ cell_layer_norm_coefficients),
|
|
|
+ output_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ output_layer_norm_coefficients),
|
|
|
+ input_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(input_gate_bias),
|
|
|
+ forget_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(forget_gate_bias),
|
|
|
+ cell_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(cell_gate_bias),
|
|
|
+ output_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(output_gate_bias),
|
|
|
+ projection_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(projection_weights),
|
|
|
+ projection_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(projection_bias),
|
|
|
+ params,
|
|
|
+ /*n_batch=*/1, n_cell, n_input, aux_input_size, n_output,
|
|
|
+ output_batch_leading_dim, output_state_ptr, cell_state_ptr,
|
|
|
+ input_gate_scratch_ptr, forget_gate_scratch_ptr,
|
|
|
+ cell_gate_scratch_ptr, output_gate_scratch_ptr, output_ptr);
|
|
|
+ }
|
|
|
+ }
|
|
|
+ }
|
|
|
+ return kTfLiteOk;
|
|
|
+}
|
|
|
+
|
|
|
+TfLiteStatus EvalHybridLstm(
|
|
|
+ const HybridLstmScales* hybrid_lstm_scales, const TfLiteEvalTensor* input,
|
|
|
+ const TfLiteEvalTensor* input_to_input_weights,
|
|
|
+ const TfLiteEvalTensor* input_to_input_weights_ledger,
|
|
|
+ const TfLiteEvalTensor* input_to_forget_weights,
|
|
|
+ const TfLiteEvalTensor* input_to_forget_weights_ledger,
|
|
|
+ const TfLiteEvalTensor* input_to_cell_weights,
|
|
|
+ const TfLiteEvalTensor* input_to_cell_weights_ledger,
|
|
|
+ const TfLiteEvalTensor* input_to_output_weights,
|
|
|
+ const TfLiteEvalTensor* input_to_output_weights_ledger,
|
|
|
+ const TfLiteEvalTensor* recurrent_to_input_weights,
|
|
|
+ const TfLiteEvalTensor* recurrent_to_input_weights_ledger,
|
|
|
+ const TfLiteEvalTensor* recurrent_to_forget_weights,
|
|
|
+ const TfLiteEvalTensor* recurrent_to_forget_weights_ledger,
|
|
|
+ const TfLiteEvalTensor* recurrent_to_cell_weights,
|
|
|
+ const TfLiteEvalTensor* recurrent_to_cell_weights_ledger,
|
|
|
+ const TfLiteEvalTensor* recurrent_to_output_weights,
|
|
|
+ const TfLiteEvalTensor* recurrent_to_output_weights_ledger,
|
|
|
+ const TfLiteEvalTensor* cell_to_input_weights,
|
|
|
+ const TfLiteEvalTensor* cell_to_forget_weights,
|
|
|
+ const TfLiteEvalTensor* cell_to_output_weights,
|
|
|
+ const TfLiteEvalTensor* input_layer_norm_coefficients,
|
|
|
+ const TfLiteEvalTensor* forget_layer_norm_coefficients,
|
|
|
+ const TfLiteEvalTensor* cell_layer_norm_coefficients,
|
|
|
+ const TfLiteEvalTensor* output_layer_norm_coefficients,
|
|
|
+ const TfLiteEvalTensor* aux_input,
|
|
|
+ const TfLiteEvalTensor* aux_input_to_input_weights,
|
|
|
+ const TfLiteEvalTensor* aux_input_to_forget_weights,
|
|
|
+ const TfLiteEvalTensor* aux_input_to_cell_weights,
|
|
|
+ const TfLiteEvalTensor* aux_input_to_output_weights,
|
|
|
+ const TfLiteEvalTensor* input_gate_bias,
|
|
|
+ const TfLiteEvalTensor* forget_gate_bias,
|
|
|
+ const TfLiteEvalTensor* cell_gate_bias,
|
|
|
+ const TfLiteEvalTensor* output_gate_bias,
|
|
|
+ const TfLiteEvalTensor* projection_weights,
|
|
|
+ const TfLiteEvalTensor* projection_weights_ledger,
|
|
|
+ const TfLiteEvalTensor* projection_bias, const TfLiteLSTMParams* params,
|
|
|
+ bool forward_sequence, bool time_major, int output_offset,
|
|
|
+ float* scratch_buffer, float* input_sf, float* aux_input_sf,
|
|
|
+ float* output_state_sf, float* prod_scaling_factors,
|
|
|
+ float* recovered_cell_weights, int8_t* input_quantized,
|
|
|
+ int8_t* aux_input_quantized, int8_t* output_state_quantized,
|
|
|
+ int8_t* cell_state_quantized, float* scales, TfLiteEvalTensor* output_state,
|
|
|
+ TfLiteEvalTensor* cell_state, int32_t* output_scratch_buffer,
|
|
|
+ TfLiteEvalTensor* output, int32_t* input_zp, int32_t* aux_input_zp,
|
|
|
+ int32_t* output_state_zp, int32_t* row_sums, int row_sums_size,
|
|
|
+ bool* compute_row_sums) {
|
|
|
+ TFLITE_DCHECK(input->dims->size >= 2 && input->dims->size <= 3);
|
|
|
+ const int n_input = input->dims->data[input->dims->size - 1];
|
|
|
+ int max_time, n_batch;
|
|
|
+ if (input->dims->size == 2) {
|
|
|
+ max_time = 1;
|
|
|
+ n_batch = input->dims->data[0];
|
|
|
+ } else {
|
|
|
+ max_time = (time_major) ? input->dims->data[0] : input->dims->data[1];
|
|
|
+ n_batch = (time_major) ? input->dims->data[1] : input->dims->data[0];
|
|
|
+ }
|
|
|
+ const int aux_input_size =
|
|
|
+ (aux_input) ? aux_input->dims->data[aux_input->dims->size - 1] : 0;
|
|
|
+ // n_cell and n_output will be the same size when there is no projection.
|
|
|
+ const int n_cell = input_to_output_weights->dims->data[0];
|
|
|
+ const int n_output = recurrent_to_output_weights->dims->data[1];
|
|
|
+
|
|
|
+ // Since we have already checked that weights are all there or none, we can
|
|
|
+ // check the existence of only one to get the condition.
|
|
|
+ const bool use_cifg = (input_to_input_weights == nullptr);
|
|
|
+
|
|
|
+ float* input_gate_scratch = nullptr;
|
|
|
+ float* cell_gate_scratch = nullptr;
|
|
|
+ float* forget_gate_scratch = nullptr;
|
|
|
+ float* output_gate_scratch = nullptr;
|
|
|
+ if (use_cifg) {
|
|
|
+ cell_gate_scratch = scratch_buffer;
|
|
|
+ forget_gate_scratch = scratch_buffer + n_cell * n_batch;
|
|
|
+ output_gate_scratch = scratch_buffer + 2 * n_cell * n_batch;
|
|
|
+ } else {
|
|
|
+ input_gate_scratch = scratch_buffer;
|
|
|
+ cell_gate_scratch = scratch_buffer + n_cell * n_batch;
|
|
|
+ forget_gate_scratch = scratch_buffer + 2 * n_cell * n_batch;
|
|
|
+ output_gate_scratch = scratch_buffer + 3 * n_cell * n_batch;
|
|
|
+ }
|
|
|
+
|
|
|
+ const int output_batch_leading_dim =
|
|
|
+ output->dims->data[output->dims->size - 1];
|
|
|
+
|
|
|
+ int32_t* input_zp_ptr = nullptr;
|
|
|
+ int32_t* aux_input_zp_ptr = nullptr;
|
|
|
+ int32_t* output_state_zp_ptr = nullptr;
|
|
|
+ int32_t* row_sums_ptr = nullptr;
|
|
|
+ if (params->asymmetric_quantize_inputs) {
|
|
|
+ input_zp_ptr = input_zp;
|
|
|
+ aux_input_zp_ptr = aux_input_zp;
|
|
|
+ output_state_zp_ptr = output_state_zp;
|
|
|
+ row_sums_ptr = row_sums;
|
|
|
+ }
|
|
|
+
|
|
|
+ if (time_major) {
|
|
|
+ // Feed the sequence into the LSTM step-by-step.
|
|
|
+ const int input_step = n_batch * n_input;
|
|
|
+ const int output_step = n_batch * output_batch_leading_dim;
|
|
|
+ for (int t = 0; t < max_time; t++) {
|
|
|
+ // If this is the forward_sequence, step forward, otherwise step
|
|
|
+ // backwards.
|
|
|
+ const int t_rel = forward_sequence ? t : max_time - t - 1;
|
|
|
+ const float* input_ptr =
|
|
|
+ tflite::micro::GetTensorData<float>(input) + t_rel * input_step;
|
|
|
+ const float* aux_input_ptr = nullptr;
|
|
|
+ if (aux_input) {
|
|
|
+ aux_input_ptr =
|
|
|
+ tflite::micro::GetTensorData<float>(aux_input) + t_rel * input_step;
|
|
|
+ }
|
|
|
+ float* output_ptr = tflite::micro::GetTensorData<float>(output) +
|
|
|
+ t_rel * output_step + output_offset;
|
|
|
+ LstmStepHybrid(
|
|
|
+ input_ptr,
|
|
|
+ input_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(input_to_input_weights),
|
|
|
+ input_to_input_weights_ledger == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<uint8_t>(
|
|
|
+ input_to_input_weights_ledger),
|
|
|
+ hybrid_lstm_scales->input_to_input_weights_scale,
|
|
|
+ input_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(input_to_forget_weights),
|
|
|
+ input_to_forget_weights_ledger == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<uint8_t>(
|
|
|
+ input_to_forget_weights_ledger),
|
|
|
+ hybrid_lstm_scales->input_to_forget_weights_scale,
|
|
|
+ input_to_cell_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(input_to_cell_weights),
|
|
|
+ input_to_cell_weights_ledger == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<uint8_t>(
|
|
|
+ input_to_cell_weights_ledger),
|
|
|
+ hybrid_lstm_scales->input_to_cell_weights_scale,
|
|
|
+ input_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(input_to_output_weights),
|
|
|
+ input_to_output_weights_ledger == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<uint8_t>(
|
|
|
+ input_to_output_weights_ledger),
|
|
|
+ hybrid_lstm_scales->input_to_output_weights_scale, aux_input_ptr,
|
|
|
+ aux_input_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(
|
|
|
+ aux_input_to_input_weights),
|
|
|
+ hybrid_lstm_scales->aux_input_to_input_weights_scale,
|
|
|
+ aux_input_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(
|
|
|
+ aux_input_to_forget_weights),
|
|
|
+ hybrid_lstm_scales->aux_input_to_forget_weights_scale,
|
|
|
+ aux_input_to_cell_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(aux_input_to_cell_weights),
|
|
|
+ hybrid_lstm_scales->aux_input_to_cell_weights_scale,
|
|
|
+ aux_input_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(
|
|
|
+ aux_input_to_output_weights),
|
|
|
+ hybrid_lstm_scales->aux_input_to_output_weights_scale,
|
|
|
+ recurrent_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(
|
|
|
+ recurrent_to_input_weights),
|
|
|
+ recurrent_to_input_weights_ledger == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<uint8_t>(
|
|
|
+ recurrent_to_input_weights_ledger),
|
|
|
+ hybrid_lstm_scales->recurrent_to_input_weights_scale,
|
|
|
+ recurrent_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(
|
|
|
+ recurrent_to_forget_weights),
|
|
|
+ recurrent_to_forget_weights_ledger == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<uint8_t>(
|
|
|
+ recurrent_to_forget_weights_ledger),
|
|
|
+ hybrid_lstm_scales->recurrent_to_forget_weights_scale,
|
|
|
+ recurrent_to_cell_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(recurrent_to_cell_weights),
|
|
|
+ recurrent_to_cell_weights_ledger == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<uint8_t>(
|
|
|
+ recurrent_to_cell_weights_ledger),
|
|
|
+ hybrid_lstm_scales->recurrent_to_cell_weights_scale,
|
|
|
+ recurrent_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(
|
|
|
+ recurrent_to_output_weights),
|
|
|
+ recurrent_to_output_weights_ledger == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<uint8_t>(
|
|
|
+ recurrent_to_output_weights_ledger),
|
|
|
+ hybrid_lstm_scales->recurrent_to_output_weights_scale,
|
|
|
+ cell_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(cell_to_input_weights),
|
|
|
+ hybrid_lstm_scales->cell_to_input_weights_scale,
|
|
|
+ cell_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(cell_to_forget_weights),
|
|
|
+ hybrid_lstm_scales->cell_to_forget_weights_scale,
|
|
|
+ cell_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(cell_to_output_weights),
|
|
|
+ hybrid_lstm_scales->cell_to_output_weights_scale,
|
|
|
+ input_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ input_layer_norm_coefficients),
|
|
|
+ forget_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ forget_layer_norm_coefficients),
|
|
|
+ cell_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ cell_layer_norm_coefficients),
|
|
|
+ output_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ output_layer_norm_coefficients),
|
|
|
+ input_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(input_gate_bias),
|
|
|
+ forget_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(forget_gate_bias),
|
|
|
+ cell_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(cell_gate_bias),
|
|
|
+ output_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(output_gate_bias),
|
|
|
+ projection_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(projection_weights),
|
|
|
+ projection_weights_ledger == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<uint8_t>(
|
|
|
+ projection_weights_ledger),
|
|
|
+ hybrid_lstm_scales->projection_weights_scale,
|
|
|
+ projection_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(projection_bias),
|
|
|
+ params, n_batch, n_cell, n_input, aux_input_size, n_output,
|
|
|
+ output_batch_leading_dim, input_gate_scratch, forget_gate_scratch,
|
|
|
+ cell_gate_scratch, output_gate_scratch, scales, input_sf,
|
|
|
+ aux_input_sf, output_state_sf, prod_scaling_factors,
|
|
|
+ recovered_cell_weights, input_quantized, aux_input_quantized,
|
|
|
+ output_state_quantized, cell_state_quantized,
|
|
|
+ tflite::micro::GetTensorData<float>(output_state),
|
|
|
+ tflite::micro::GetTensorData<float>(cell_state),
|
|
|
+ output_scratch_buffer, output_ptr, input_zp_ptr, aux_input_zp_ptr,
|
|
|
+ output_state_zp_ptr, row_sums_ptr, row_sums_size, compute_row_sums,
|
|
|
+ params->asymmetric_quantize_inputs);
|
|
|
+ }
|
|
|
+ } else {
|
|
|
+ for (int b = 0; b < n_batch; b++) {
|
|
|
+ const int input_step = n_input;
|
|
|
+ const int output_step = output_batch_leading_dim;
|
|
|
+ for (int t = 0; t < max_time; t++) {
|
|
|
+ // If this is the forward_sequence, step forward, otherwise step
|
|
|
+ // backwards.
|
|
|
+ const int t_rel = forward_sequence ? t : max_time - t - 1;
|
|
|
+ const int time_offset = b * max_time + t_rel;
|
|
|
+ const float* input_ptr = tflite::micro::GetTensorData<float>(input) +
|
|
|
+ time_offset * input_step;
|
|
|
+ const float* aux_input_ptr = nullptr;
|
|
|
+ if (aux_input) {
|
|
|
+ aux_input_ptr = tflite::micro::GetTensorData<float>(aux_input) +
|
|
|
+ time_offset * input_step;
|
|
|
+ }
|
|
|
+ float* output_ptr = tflite::micro::GetTensorData<float>(output) +
|
|
|
+ time_offset * output_step + output_offset;
|
|
|
+
|
|
|
+ // Offset the {output,cell}_state pointers to the right batch.
|
|
|
+ float* output_state_ptr =
|
|
|
+ tflite::micro::GetTensorData<float>(output_state) +
|
|
|
+ b * output_batch_leading_dim;
|
|
|
+ float* cell_state_ptr =
|
|
|
+ tflite::micro::GetTensorData<float>(cell_state) + b * n_cell;
|
|
|
+ // Offset the scratch pointers to the right batch.
|
|
|
+ float* input_gate_scratch_ptr =
|
|
|
+ input_gate_scratch ? input_gate_scratch + b * n_cell : nullptr;
|
|
|
+ float* forget_gate_scratch_ptr = forget_gate_scratch + b * n_cell;
|
|
|
+ float* cell_gate_scratch_ptr = cell_gate_scratch + b * n_cell;
|
|
|
+ float* output_gate_scratch_ptr = output_gate_scratch + b * n_cell;
|
|
|
+
|
|
|
+ LstmStepHybrid(
|
|
|
+ input_ptr,
|
|
|
+ input_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(input_to_input_weights),
|
|
|
+ input_to_input_weights_ledger == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<uint8_t>(
|
|
|
+ input_to_input_weights_ledger),
|
|
|
+ hybrid_lstm_scales->input_to_input_weights_scale,
|
|
|
+ input_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(input_to_forget_weights),
|
|
|
+ input_to_forget_weights_ledger == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<uint8_t>(
|
|
|
+ input_to_forget_weights_ledger),
|
|
|
+ hybrid_lstm_scales->input_to_forget_weights_scale,
|
|
|
+ input_to_cell_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(input_to_cell_weights),
|
|
|
+ input_to_cell_weights_ledger == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<uint8_t>(
|
|
|
+ input_to_cell_weights_ledger),
|
|
|
+ hybrid_lstm_scales->input_to_cell_weights_scale,
|
|
|
+ input_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(input_to_output_weights),
|
|
|
+ input_to_output_weights_ledger == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<uint8_t>(
|
|
|
+ input_to_output_weights_ledger),
|
|
|
+ hybrid_lstm_scales->input_to_output_weights_scale, aux_input_ptr,
|
|
|
+ aux_input_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(
|
|
|
+ aux_input_to_input_weights),
|
|
|
+ hybrid_lstm_scales->aux_input_to_input_weights_scale,
|
|
|
+ aux_input_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(
|
|
|
+ aux_input_to_forget_weights),
|
|
|
+ hybrid_lstm_scales->aux_input_to_forget_weights_scale,
|
|
|
+ aux_input_to_cell_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(
|
|
|
+ aux_input_to_cell_weights),
|
|
|
+ hybrid_lstm_scales->aux_input_to_cell_weights_scale,
|
|
|
+ aux_input_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(
|
|
|
+ aux_input_to_output_weights),
|
|
|
+ hybrid_lstm_scales->aux_input_to_output_weights_scale,
|
|
|
+ recurrent_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(
|
|
|
+ recurrent_to_input_weights),
|
|
|
+ recurrent_to_input_weights_ledger == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<uint8_t>(
|
|
|
+ recurrent_to_input_weights_ledger),
|
|
|
+ hybrid_lstm_scales->recurrent_to_input_weights_scale,
|
|
|
+ recurrent_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(
|
|
|
+ recurrent_to_forget_weights),
|
|
|
+ recurrent_to_forget_weights_ledger == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<uint8_t>(
|
|
|
+ recurrent_to_forget_weights_ledger),
|
|
|
+ hybrid_lstm_scales->recurrent_to_forget_weights_scale,
|
|
|
+ recurrent_to_cell_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(
|
|
|
+ recurrent_to_cell_weights),
|
|
|
+ recurrent_to_cell_weights_ledger == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<uint8_t>(
|
|
|
+ recurrent_to_cell_weights_ledger),
|
|
|
+ hybrid_lstm_scales->recurrent_to_cell_weights_scale,
|
|
|
+ recurrent_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(
|
|
|
+ recurrent_to_output_weights),
|
|
|
+ recurrent_to_output_weights_ledger == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<uint8_t>(
|
|
|
+ recurrent_to_output_weights_ledger),
|
|
|
+ hybrid_lstm_scales->recurrent_to_output_weights_scale,
|
|
|
+ cell_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(cell_to_input_weights),
|
|
|
+ hybrid_lstm_scales->cell_to_input_weights_scale,
|
|
|
+ cell_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(cell_to_forget_weights),
|
|
|
+ hybrid_lstm_scales->cell_to_forget_weights_scale,
|
|
|
+ cell_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(cell_to_output_weights),
|
|
|
+ hybrid_lstm_scales->cell_to_output_weights_scale,
|
|
|
+ input_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ input_layer_norm_coefficients),
|
|
|
+ forget_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ forget_layer_norm_coefficients),
|
|
|
+ cell_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ cell_layer_norm_coefficients),
|
|
|
+ output_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(
|
|
|
+ output_layer_norm_coefficients),
|
|
|
+ input_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(input_gate_bias),
|
|
|
+ forget_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(forget_gate_bias),
|
|
|
+ cell_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(cell_gate_bias),
|
|
|
+ output_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(output_gate_bias),
|
|
|
+ projection_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(projection_weights),
|
|
|
+ projection_weights_ledger == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<uint8_t>(
|
|
|
+ projection_weights_ledger),
|
|
|
+ hybrid_lstm_scales->projection_weights_scale,
|
|
|
+ projection_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<float>(projection_bias),
|
|
|
+ params,
|
|
|
+ /*n_batch=*/1, n_cell, n_input, aux_input_size, n_output,
|
|
|
+ output_batch_leading_dim, input_gate_scratch_ptr,
|
|
|
+ forget_gate_scratch_ptr, cell_gate_scratch_ptr,
|
|
|
+ output_gate_scratch_ptr, scales, input_sf, aux_input_sf,
|
|
|
+ output_state_sf, prod_scaling_factors, recovered_cell_weights,
|
|
|
+ input_quantized, aux_input_quantized, output_state_quantized,
|
|
|
+ cell_state_quantized, output_state_ptr, cell_state_ptr,
|
|
|
+ output_scratch_buffer, output_ptr, input_zp_ptr, aux_input_zp_ptr,
|
|
|
+ output_state_zp_ptr, row_sums_ptr, row_sums_size, compute_row_sums,
|
|
|
+ params->asymmetric_quantize_inputs);
|
|
|
+ }
|
|
|
+ }
|
|
|
+ }
|
|
|
+
|
|
|
+ return kTfLiteOk;
|
|
|
+}
|
|
|
+
|
|
|
+TfLiteStatus EvalInteger8x8_16Lstm(
|
|
|
+ const TfLiteEvalTensor* input,
|
|
|
+ const TfLiteEvalTensor* input_to_input_weights,
|
|
|
+ const TfLiteEvalTensor* input_to_forget_weights,
|
|
|
+ const TfLiteEvalTensor* input_to_cell_weights,
|
|
|
+ const TfLiteEvalTensor* input_to_output_weights,
|
|
|
+ const TfLiteEvalTensor* recurrent_to_input_weights,
|
|
|
+ const TfLiteEvalTensor* recurrent_to_forget_weights,
|
|
|
+ const TfLiteEvalTensor* recurrent_to_cell_weights,
|
|
|
+ const TfLiteEvalTensor* recurrent_to_output_weights,
|
|
|
+ const TfLiteEvalTensor* cell_to_input_weights,
|
|
|
+ const TfLiteEvalTensor* cell_to_forget_weights,
|
|
|
+ const TfLiteEvalTensor* cell_to_output_weights,
|
|
|
+ const TfLiteEvalTensor* input_layer_norm_coefficients,
|
|
|
+ const TfLiteEvalTensor* forget_layer_norm_coefficients,
|
|
|
+ const TfLiteEvalTensor* cell_layer_norm_coefficients,
|
|
|
+ const TfLiteEvalTensor* output_layer_norm_coefficients,
|
|
|
+ const TfLiteEvalTensor* input_gate_bias,
|
|
|
+ const TfLiteEvalTensor* forget_gate_bias,
|
|
|
+ const TfLiteEvalTensor* cell_gate_bias,
|
|
|
+ const TfLiteEvalTensor* output_gate_bias,
|
|
|
+ const TfLiteEvalTensor* projection_weights,
|
|
|
+ const TfLiteEvalTensor* projection_bias, const TfLiteLSTMParams* params,
|
|
|
+ bool forward_sequence, bool time_major,
|
|
|
+ const IntegerLstmParameter* integer_lstm_param, int32_t output_state_zp,
|
|
|
+ TfLiteEvalTensor* output_state, TfLiteEvalTensor* cell_state,
|
|
|
+ TfLiteEvalTensor* output, int16_t* scratch0, int16_t* scratch1,
|
|
|
+ int16_t* scratch2, int16_t* scratch3, int8_t* scratch4, int32_t* scratch5) {
|
|
|
+ TFLITE_DCHECK(input->dims->size >= 2 && input->dims->size <= 3);
|
|
|
+ const int n_input = input->dims->data[input->dims->size - 1];
|
|
|
+ int max_time, n_batch;
|
|
|
+ if (input->dims->size == 2) {
|
|
|
+ max_time = 1;
|
|
|
+ n_batch = input->dims->data[0];
|
|
|
+ } else {
|
|
|
+ max_time = (time_major) ? input->dims->data[0] : input->dims->data[1];
|
|
|
+ n_batch = (time_major) ? input->dims->data[1] : input->dims->data[0];
|
|
|
+ }
|
|
|
+
|
|
|
+ // n_cell and n_output will be the same size when there is no projection.
|
|
|
+ const int n_cell = input_to_output_weights->dims->data[0];
|
|
|
+ const int n_output = recurrent_to_output_weights->dims->data[1];
|
|
|
+
|
|
|
+ // Get params for time/batch/sequence.
|
|
|
+ const int output_batch_leading_dim =
|
|
|
+ output->dims->data[output->dims->size - 1];
|
|
|
+
|
|
|
+ if (time_major) {
|
|
|
+ const int input_step = n_batch * n_input;
|
|
|
+ const int output_step = n_batch * output_batch_leading_dim;
|
|
|
+ for (int t = 0; t < max_time; t++) {
|
|
|
+ const int t_rel = t;
|
|
|
+ int8_t* output_ptr =
|
|
|
+ tflite::micro::GetTensorData<int8_t>(output) + t_rel * output_step;
|
|
|
+ const int8_t* input_ptr =
|
|
|
+ tflite::micro::GetTensorData<int8_t>(input) + t_rel * input_step;
|
|
|
+ LstmStepInteger8x8_16(
|
|
|
+ input_ptr,
|
|
|
+ input_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(input_to_input_weights),
|
|
|
+ integer_lstm_param->effective_input_to_input_scale_a,
|
|
|
+ integer_lstm_param->effective_input_to_input_scale_b,
|
|
|
+ input_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(input_to_forget_weights),
|
|
|
+ integer_lstm_param->effective_input_to_forget_scale_a,
|
|
|
+ integer_lstm_param->effective_input_to_forget_scale_b,
|
|
|
+ input_to_cell_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(input_to_cell_weights),
|
|
|
+ integer_lstm_param->effective_input_to_cell_scale_a,
|
|
|
+ integer_lstm_param->effective_input_to_cell_scale_b,
|
|
|
+ input_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(input_to_output_weights),
|
|
|
+ integer_lstm_param->effective_input_to_output_scale_a,
|
|
|
+ integer_lstm_param->effective_input_to_output_scale_b,
|
|
|
+ recurrent_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(
|
|
|
+ recurrent_to_input_weights),
|
|
|
+ integer_lstm_param->effective_recurrent_to_input_scale_a,
|
|
|
+ integer_lstm_param->effective_recurrent_to_input_scale_b,
|
|
|
+ recurrent_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(
|
|
|
+ recurrent_to_forget_weights),
|
|
|
+ integer_lstm_param->effective_recurrent_to_forget_scale_a,
|
|
|
+ integer_lstm_param->effective_recurrent_to_forget_scale_b,
|
|
|
+ recurrent_to_cell_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(recurrent_to_cell_weights),
|
|
|
+ integer_lstm_param->effective_recurrent_to_cell_scale_a,
|
|
|
+ integer_lstm_param->effective_recurrent_to_cell_scale_b,
|
|
|
+ recurrent_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(
|
|
|
+ recurrent_to_output_weights),
|
|
|
+ integer_lstm_param->effective_recurrent_to_output_scale_a,
|
|
|
+ integer_lstm_param->effective_recurrent_to_output_scale_b,
|
|
|
+ cell_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int16_t>(cell_to_input_weights),
|
|
|
+ integer_lstm_param->effective_cell_to_input_scale_a,
|
|
|
+ integer_lstm_param->effective_cell_to_input_scale_b,
|
|
|
+ cell_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int16_t>(cell_to_forget_weights),
|
|
|
+ integer_lstm_param->effective_cell_to_forget_scale_a,
|
|
|
+ integer_lstm_param->effective_cell_to_forget_scale_b,
|
|
|
+ cell_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int16_t>(cell_to_output_weights),
|
|
|
+ integer_lstm_param->effective_cell_to_output_scale_a,
|
|
|
+ integer_lstm_param->effective_cell_to_output_scale_b,
|
|
|
+ projection_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(projection_weights),
|
|
|
+ integer_lstm_param->effective_proj_scale_a,
|
|
|
+ integer_lstm_param->effective_proj_scale_b,
|
|
|
+ integer_lstm_param->hidden_zp,
|
|
|
+ integer_lstm_param->effective_hidden_scale_a,
|
|
|
+ integer_lstm_param->effective_hidden_scale_b,
|
|
|
+ input_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int16_t>(
|
|
|
+ input_layer_norm_coefficients),
|
|
|
+ integer_lstm_param->layer_norm_input_scale_a,
|
|
|
+ integer_lstm_param->layer_norm_input_scale_b,
|
|
|
+ forget_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int16_t>(
|
|
|
+ forget_layer_norm_coefficients),
|
|
|
+ integer_lstm_param->layer_norm_forget_scale_a,
|
|
|
+ integer_lstm_param->layer_norm_forget_scale_b,
|
|
|
+ cell_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int16_t>(
|
|
|
+ cell_layer_norm_coefficients),
|
|
|
+ integer_lstm_param->layer_norm_cell_scale_a,
|
|
|
+ integer_lstm_param->layer_norm_cell_scale_b,
|
|
|
+ output_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int16_t>(
|
|
|
+ output_layer_norm_coefficients),
|
|
|
+ integer_lstm_param->layer_norm_output_scale_a,
|
|
|
+ integer_lstm_param->layer_norm_output_scale_b,
|
|
|
+ input_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int32_t>(input_gate_bias),
|
|
|
+ forget_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int32_t>(forget_gate_bias),
|
|
|
+ cell_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int32_t>(cell_gate_bias),
|
|
|
+ output_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int32_t>(output_gate_bias),
|
|
|
+ integer_lstm_param->quantized_cell_clip,
|
|
|
+ integer_lstm_param->quantized_proj_clip,
|
|
|
+ integer_lstm_param->cell_scale,
|
|
|
+ integer_lstm_param->input_variance_guard,
|
|
|
+ integer_lstm_param->forget_variance_guard,
|
|
|
+ integer_lstm_param->cell_variance_guard,
|
|
|
+ integer_lstm_param->output_variance_guard,
|
|
|
+ integer_lstm_param->input_to_forget_effective_bias,
|
|
|
+ integer_lstm_param->recurrent_to_forget_effective_bias,
|
|
|
+ integer_lstm_param->input_to_cell_effective_bias,
|
|
|
+ integer_lstm_param->recurrent_to_cell_effective_bias,
|
|
|
+ integer_lstm_param->input_to_output_effective_bias,
|
|
|
+ integer_lstm_param->recurrent_to_output_effective_bias,
|
|
|
+ integer_lstm_param->input_to_input_effective_bias,
|
|
|
+ integer_lstm_param->recurrent_to_input_effective_bias,
|
|
|
+ integer_lstm_param->projection_effective_bias, n_batch, n_cell,
|
|
|
+ n_input, n_output, tflite::micro::GetTensorData<int8_t>(output_state),
|
|
|
+ output_state_zp, tflite::micro::GetTensorData<int16_t>(cell_state),
|
|
|
+ output_ptr, scratch0, scratch1, scratch2, scratch3, scratch4,
|
|
|
+ scratch5);
|
|
|
+ }
|
|
|
+ } else {
|
|
|
+ for (int b = 0; b < n_batch; b++) {
|
|
|
+ const int input_step = n_input;
|
|
|
+ const int output_step = output_batch_leading_dim;
|
|
|
+ for (int t = 0; t < max_time; t++) {
|
|
|
+ // If this is the forward_sequence, step forward, otherwise step
|
|
|
+ // backwards.
|
|
|
+ const int t_rel = forward_sequence ? t : max_time - t - 1;
|
|
|
+ const int time_offset = b * max_time + t_rel;
|
|
|
+ const int8_t* input_ptr = tflite::micro::GetTensorData<int8_t>(input) +
|
|
|
+ time_offset * input_step;
|
|
|
+ int8_t* output_ptr = tflite::micro::GetTensorData<int8_t>(output) +
|
|
|
+ time_offset * output_step;
|
|
|
+
|
|
|
+ // Offset the {output,cell}_state pointers to the right batch.
|
|
|
+ int8_t* output_state_ptr =
|
|
|
+ tflite::micro::GetTensorData<int8_t>(output_state) +
|
|
|
+ b * output_batch_leading_dim;
|
|
|
+ int16_t* cell_state_ptr =
|
|
|
+ tflite::micro::GetTensorData<int16_t>(cell_state) + b * n_cell;
|
|
|
+
|
|
|
+ LstmStepInteger8x8_16(
|
|
|
+ input_ptr,
|
|
|
+ input_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(input_to_input_weights),
|
|
|
+ integer_lstm_param->effective_input_to_input_scale_a,
|
|
|
+ integer_lstm_param->effective_input_to_input_scale_b,
|
|
|
+ input_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(input_to_forget_weights),
|
|
|
+ integer_lstm_param->effective_input_to_forget_scale_a,
|
|
|
+ integer_lstm_param->effective_input_to_forget_scale_b,
|
|
|
+ input_to_cell_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(input_to_cell_weights),
|
|
|
+ integer_lstm_param->effective_input_to_cell_scale_a,
|
|
|
+ integer_lstm_param->effective_input_to_cell_scale_b,
|
|
|
+ input_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(input_to_output_weights),
|
|
|
+ integer_lstm_param->effective_input_to_output_scale_a,
|
|
|
+ integer_lstm_param->effective_input_to_output_scale_b,
|
|
|
+ recurrent_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(
|
|
|
+ recurrent_to_input_weights),
|
|
|
+ integer_lstm_param->effective_recurrent_to_input_scale_a,
|
|
|
+ integer_lstm_param->effective_recurrent_to_input_scale_b,
|
|
|
+ recurrent_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(
|
|
|
+ recurrent_to_forget_weights),
|
|
|
+ integer_lstm_param->effective_recurrent_to_forget_scale_a,
|
|
|
+ integer_lstm_param->effective_recurrent_to_forget_scale_b,
|
|
|
+ recurrent_to_cell_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(
|
|
|
+ recurrent_to_cell_weights),
|
|
|
+ integer_lstm_param->effective_recurrent_to_cell_scale_a,
|
|
|
+ integer_lstm_param->effective_recurrent_to_cell_scale_b,
|
|
|
+ recurrent_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(
|
|
|
+ recurrent_to_output_weights),
|
|
|
+ integer_lstm_param->effective_recurrent_to_output_scale_a,
|
|
|
+ integer_lstm_param->effective_recurrent_to_output_scale_b,
|
|
|
+ cell_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int16_t>(cell_to_input_weights),
|
|
|
+ integer_lstm_param->effective_cell_to_input_scale_a,
|
|
|
+ integer_lstm_param->effective_cell_to_input_scale_b,
|
|
|
+ cell_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int16_t>(cell_to_forget_weights),
|
|
|
+ integer_lstm_param->effective_cell_to_forget_scale_a,
|
|
|
+ integer_lstm_param->effective_cell_to_forget_scale_b,
|
|
|
+ cell_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int16_t>(cell_to_output_weights),
|
|
|
+ integer_lstm_param->effective_cell_to_output_scale_a,
|
|
|
+ integer_lstm_param->effective_cell_to_output_scale_b,
|
|
|
+ projection_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(projection_weights),
|
|
|
+ integer_lstm_param->effective_proj_scale_a,
|
|
|
+ integer_lstm_param->effective_proj_scale_b,
|
|
|
+ integer_lstm_param->hidden_zp,
|
|
|
+ integer_lstm_param->effective_hidden_scale_a,
|
|
|
+ integer_lstm_param->effective_hidden_scale_b,
|
|
|
+ input_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int16_t>(
|
|
|
+ input_layer_norm_coefficients),
|
|
|
+ integer_lstm_param->layer_norm_input_scale_a,
|
|
|
+ integer_lstm_param->layer_norm_input_scale_b,
|
|
|
+ forget_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int16_t>(
|
|
|
+ forget_layer_norm_coefficients),
|
|
|
+ integer_lstm_param->layer_norm_forget_scale_a,
|
|
|
+ integer_lstm_param->layer_norm_forget_scale_b,
|
|
|
+ cell_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int16_t>(
|
|
|
+ cell_layer_norm_coefficients),
|
|
|
+ integer_lstm_param->layer_norm_cell_scale_a,
|
|
|
+ integer_lstm_param->layer_norm_cell_scale_b,
|
|
|
+ output_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int16_t>(
|
|
|
+ output_layer_norm_coefficients),
|
|
|
+ integer_lstm_param->layer_norm_output_scale_a,
|
|
|
+ integer_lstm_param->layer_norm_output_scale_b,
|
|
|
+ input_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int32_t>(input_gate_bias),
|
|
|
+ forget_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int32_t>(forget_gate_bias),
|
|
|
+ cell_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int32_t>(cell_gate_bias),
|
|
|
+ output_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int32_t>(output_gate_bias),
|
|
|
+ integer_lstm_param->quantized_cell_clip,
|
|
|
+ integer_lstm_param->quantized_proj_clip,
|
|
|
+ integer_lstm_param->cell_scale,
|
|
|
+ integer_lstm_param->input_variance_guard,
|
|
|
+ integer_lstm_param->forget_variance_guard,
|
|
|
+ integer_lstm_param->cell_variance_guard,
|
|
|
+ integer_lstm_param->output_variance_guard,
|
|
|
+ integer_lstm_param->input_to_forget_effective_bias,
|
|
|
+ integer_lstm_param->recurrent_to_forget_effective_bias,
|
|
|
+ integer_lstm_param->input_to_cell_effective_bias,
|
|
|
+ integer_lstm_param->recurrent_to_cell_effective_bias,
|
|
|
+ integer_lstm_param->input_to_output_effective_bias,
|
|
|
+ integer_lstm_param->recurrent_to_output_effective_bias,
|
|
|
+ integer_lstm_param->input_to_input_effective_bias,
|
|
|
+ integer_lstm_param->recurrent_to_input_effective_bias,
|
|
|
+ integer_lstm_param->projection_effective_bias, /*n_batch=*/1,
|
|
|
+ n_cell, n_input, n_output, output_state_ptr, output_state_zp,
|
|
|
+ cell_state_ptr, output_ptr, scratch0, scratch1, scratch2, scratch3,
|
|
|
+ scratch4, scratch5);
|
|
|
+ }
|
|
|
+ }
|
|
|
+ }
|
|
|
+
|
|
|
+ return kTfLiteOk;
|
|
|
+}
|
|
|
+
|
|
|
+TfLiteStatus EvalInteger8x8_8Lstm(
|
|
|
+ const TfLiteEvalTensor* input,
|
|
|
+ const TfLiteEvalTensor* input_to_input_weights,
|
|
|
+ const TfLiteEvalTensor* input_to_forget_weights,
|
|
|
+ const TfLiteEvalTensor* input_to_cell_weights,
|
|
|
+ const TfLiteEvalTensor* input_to_output_weights,
|
|
|
+ const TfLiteEvalTensor* recurrent_to_input_weights,
|
|
|
+ const TfLiteEvalTensor* recurrent_to_forget_weights,
|
|
|
+ const TfLiteEvalTensor* recurrent_to_cell_weights,
|
|
|
+ const TfLiteEvalTensor* recurrent_to_output_weights,
|
|
|
+ const TfLiteEvalTensor* cell_to_input_weights,
|
|
|
+ const TfLiteEvalTensor* cell_to_forget_weights,
|
|
|
+ const TfLiteEvalTensor* cell_to_output_weights,
|
|
|
+ const TfLiteEvalTensor* input_layer_norm_coefficients,
|
|
|
+ const TfLiteEvalTensor* forget_layer_norm_coefficients,
|
|
|
+ const TfLiteEvalTensor* cell_layer_norm_coefficients,
|
|
|
+ const TfLiteEvalTensor* output_layer_norm_coefficients,
|
|
|
+ const TfLiteEvalTensor* input_gate_bias,
|
|
|
+ const TfLiteEvalTensor* forget_gate_bias,
|
|
|
+ const TfLiteEvalTensor* cell_gate_bias,
|
|
|
+ const TfLiteEvalTensor* output_gate_bias,
|
|
|
+ const TfLiteEvalTensor* projection_weights,
|
|
|
+ const TfLiteEvalTensor* projection_bias, const TfLiteLSTMParams* params,
|
|
|
+ TfLiteEvalTensor* output_state, TfLiteEvalTensor* cell_state,
|
|
|
+ TfLiteEvalTensor* output, const IntegerLstmParameter* integer_lstm_param,
|
|
|
+ int32_t input_zp, int32_t output_state_zp, int8_t* scratch0,
|
|
|
+ int8_t* scratch1, int16_t* scratch2, int16_t* scratch3, int16_t* scratch4,
|
|
|
+ int16_t* scratch5, int16_t* scratch6, int16_t* scratch7) {
|
|
|
+ TFLITE_DCHECK(input->dims->size >= 2 && input->dims->size <= 3);
|
|
|
+ const int n_input = input->dims->data[input->dims->size - 1];
|
|
|
+ int max_time, n_batch;
|
|
|
+ if (input->dims->size == 2) {
|
|
|
+ max_time = 1;
|
|
|
+ n_batch = input->dims->data[0];
|
|
|
+ } else {
|
|
|
+ max_time = input->dims->data[0];
|
|
|
+ n_batch = input->dims->data[1];
|
|
|
+ }
|
|
|
+
|
|
|
+ // n_cell and n_output will be the same size when there is no projection.
|
|
|
+ const int n_cell = input_to_output_weights->dims->data[0];
|
|
|
+ const int n_output = recurrent_to_output_weights->dims->data[1];
|
|
|
+
|
|
|
+ // Get params for time/batch/sequence.
|
|
|
+ const int output_batch_leading_dim =
|
|
|
+ output->dims->data[output->dims->size - 1];
|
|
|
+ const int input_step = n_batch * n_input;
|
|
|
+ const int output_step = n_batch * output_batch_leading_dim;
|
|
|
+
|
|
|
+ for (int t = 0; t < max_time; t++) {
|
|
|
+ const int t_rel = t;
|
|
|
+ int8_t* output_ptr =
|
|
|
+ tflite::micro::GetTensorData<int8_t>(output) + t_rel * output_step;
|
|
|
+ // Input can be int8 asymmetric or int16 symmetric.
|
|
|
+ const int8_t* input_ptr =
|
|
|
+ tflite::micro::GetTensorData<int8_t>(input) + t_rel * input_step;
|
|
|
+ LstmStepInteger8x8_8(
|
|
|
+ input_ptr, input_zp,
|
|
|
+
|
|
|
+ input_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(input_to_input_weights),
|
|
|
+ integer_lstm_param->effective_input_to_input_scale_a,
|
|
|
+ integer_lstm_param->effective_input_to_input_scale_b,
|
|
|
+
|
|
|
+ input_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(input_to_forget_weights),
|
|
|
+ integer_lstm_param->effective_input_to_forget_scale_a,
|
|
|
+ integer_lstm_param->effective_input_to_forget_scale_b,
|
|
|
+
|
|
|
+ input_to_cell_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(input_to_cell_weights),
|
|
|
+ integer_lstm_param->effective_input_to_cell_scale_a,
|
|
|
+ integer_lstm_param->effective_input_to_cell_scale_b,
|
|
|
+
|
|
|
+ input_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(input_to_output_weights),
|
|
|
+ integer_lstm_param->effective_input_to_output_scale_a,
|
|
|
+ integer_lstm_param->effective_input_to_output_scale_b,
|
|
|
+
|
|
|
+ recurrent_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(recurrent_to_input_weights),
|
|
|
+ integer_lstm_param->effective_recurrent_to_input_scale_a,
|
|
|
+ integer_lstm_param->effective_recurrent_to_input_scale_b,
|
|
|
+
|
|
|
+ recurrent_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(recurrent_to_forget_weights),
|
|
|
+ integer_lstm_param->effective_recurrent_to_forget_scale_a,
|
|
|
+ integer_lstm_param->effective_recurrent_to_forget_scale_b,
|
|
|
+
|
|
|
+ recurrent_to_cell_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(recurrent_to_cell_weights),
|
|
|
+ integer_lstm_param->effective_recurrent_to_cell_scale_a,
|
|
|
+ integer_lstm_param->effective_recurrent_to_cell_scale_b,
|
|
|
+
|
|
|
+ recurrent_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(recurrent_to_output_weights),
|
|
|
+ integer_lstm_param->effective_recurrent_to_output_scale_a,
|
|
|
+ integer_lstm_param->effective_recurrent_to_output_scale_b,
|
|
|
+
|
|
|
+ cell_to_input_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(cell_to_input_weights),
|
|
|
+ integer_lstm_param->effective_cell_to_input_scale_a,
|
|
|
+ integer_lstm_param->effective_cell_to_input_scale_b,
|
|
|
+
|
|
|
+ cell_to_forget_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(cell_to_forget_weights),
|
|
|
+ integer_lstm_param->effective_cell_to_forget_scale_a,
|
|
|
+ integer_lstm_param->effective_cell_to_forget_scale_b,
|
|
|
+
|
|
|
+ cell_to_output_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(cell_to_output_weights),
|
|
|
+ integer_lstm_param->effective_cell_to_output_scale_a,
|
|
|
+ integer_lstm_param->effective_cell_to_output_scale_b,
|
|
|
+
|
|
|
+ projection_weights == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int8_t>(projection_weights),
|
|
|
+ integer_lstm_param->effective_proj_scale_a,
|
|
|
+ integer_lstm_param->effective_proj_scale_b,
|
|
|
+
|
|
|
+ input_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int16_t>(
|
|
|
+ input_layer_norm_coefficients),
|
|
|
+ integer_lstm_param->layer_norm_input_scale_a,
|
|
|
+ integer_lstm_param->layer_norm_input_scale_b,
|
|
|
+
|
|
|
+ forget_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int16_t>(
|
|
|
+ forget_layer_norm_coefficients),
|
|
|
+ integer_lstm_param->layer_norm_forget_scale_a,
|
|
|
+ integer_lstm_param->layer_norm_forget_scale_b,
|
|
|
+
|
|
|
+ cell_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int16_t>(
|
|
|
+ cell_layer_norm_coefficients),
|
|
|
+ integer_lstm_param->layer_norm_cell_scale_a,
|
|
|
+ integer_lstm_param->layer_norm_cell_scale_b,
|
|
|
+
|
|
|
+ output_layer_norm_coefficients == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int16_t>(
|
|
|
+ output_layer_norm_coefficients),
|
|
|
+ integer_lstm_param->layer_norm_output_scale_a,
|
|
|
+ integer_lstm_param->layer_norm_output_scale_b,
|
|
|
+
|
|
|
+ input_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int32_t>(input_gate_bias),
|
|
|
+ forget_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int32_t>(forget_gate_bias),
|
|
|
+ cell_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int32_t>(cell_gate_bias),
|
|
|
+ output_gate_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int32_t>(output_gate_bias),
|
|
|
+ projection_bias == nullptr
|
|
|
+ ? nullptr
|
|
|
+ : tflite::micro::GetTensorData<int32_t>(projection_bias),
|
|
|
+
|
|
|
+ params, integer_lstm_param->intermediate_scale_a,
|
|
|
+ integer_lstm_param->intermediate_scale_b,
|
|
|
+ integer_lstm_param->intermediate_zp,
|
|
|
+ integer_lstm_param->quantized_cell_clip,
|
|
|
+ integer_lstm_param->quantized_proj_clip, n_batch, n_cell, n_input,
|
|
|
+ n_output, output_batch_leading_dim,
|
|
|
+ tflite::micro::GetTensorData<int8_t>(output_state), output_state_zp,
|
|
|
+ tflite::micro::GetTensorData<int16_t>(cell_state), output_ptr, scratch0,
|
|
|
+ scratch1, scratch2, scratch3, scratch4, scratch5, scratch6, scratch7);
|
|
|
+ }
|
|
|
+
|
|
|
+ return kTfLiteOk;
|
|
|
+}
|
|
|
+
|
|
|
+} // namespace tflite
|
jomjol