path: root/tensorflow/contrib/lite/kernels/lstm.cc

/* Copyright 2017 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 <cassert>
#include <cmath>
#include <cstdio>
#include <cstdlib>
#include <iostream>
#include <limits>

#include "tensorflow/contrib/lite/builtin_op_data.h"
#include "tensorflow/contrib/lite/context.h"
#include "tensorflow/contrib/lite/kernels/activation_functor.h"
#include "tensorflow/contrib/lite/kernels/gemm_support.h"
#include "tensorflow/contrib/lite/kernels/internal/kernel_utils.h"
#include "tensorflow/contrib/lite/kernels/internal/optimized/optimized_ops.h"
#include "tensorflow/contrib/lite/kernels/internal/tensor.h"
#include "tensorflow/contrib/lite/kernels/internal/tensor_utils.h"
#include "tensorflow/contrib/lite/kernels/kernel_util.h"
#include "tensorflow/contrib/lite/kernels/op_macros.h"

namespace tflite {
namespace ops {
namespace builtin {
namespace lstm {

struct OpData {
  // Which kernel type to use. Full kernel (18 or 20 inputs) or basic kernel
  // (5 inputs).
  TfLiteLSTMKernelType kernel_type;

  // These fields are only used by full kernel.
  int activation_state_tensor_index;
  int cell_state_tensor_index;
  int scratch_tensor_index;
};

// For full inputs kernel (18 or 20 inputs).
namespace full {

// Input Tensors of size {n_batch, n_input}
constexpr int kInputTensor = 0;

// Input weight tensors of size: {n_cell, n_input}
constexpr int kInputToInputWeightsTensor = 1;  // Optional
constexpr int kInputToForgetWeightsTensor = 2;
constexpr int kInputToCellWeightsTensor = 3;
constexpr int kInputToOutputWeightsTensor = 4;

// Recurrent weight tensors of size {n_cell, n_output}
constexpr int kRecurrentToInputWeightsTensor = 5;  // Optional
constexpr int kRecurrentToForgetWeightsTensor = 6;
constexpr int kRecurrentToCellWeightsTensor = 7;
constexpr int kRecurrentToOutputWeightsTensor = 8;

// Peephole weights tensors of size {n_cell}, representing a diagonal matrix.
constexpr int kCellToInputWeightsTensor = 9;    // Optional
constexpr int kCellToForgetWeightsTensor = 10;  // Optional
constexpr int kCellToOutputWeightsTensor = 11;  // Optional

// Gates bias tensors of size {n_cell}
constexpr int kInputGateBiasTensor = 12;  // Optional
constexpr int kForgetGateBiasTensor = 13;
constexpr int kCellGateBiasTensor = 14;
constexpr int kOutputGateBiasTensor = 15;

// Projection weight tensor of size {n_output, n_cell}
constexpr int kProjectionWeightsTensor = 16;  // Optional
// Projection bias tensor of size {n_output}
constexpr int kProjectionBiasTensor = 17;  // Optional

// If the node has 20 inputs, the following 2 tensors are used as state tensors.
// These are defined as variable tensors, and will be modified by this op.
constexpr int kInputActivationStateTensor = 18;
constexpr int kInputCellStateTensor = 19;

// Output tensors.
// * If the node has 18 inputs, these 2 tensors are used as state tensors.
// * If the node has 20 inputs, these 2 tensors are ignored.
// TODO(ycling): Make the 2 output state tensors optional, and propagate the
// state to output tensors when the 2 tensors present.
constexpr int kOutputStateTensor = 0;
constexpr int kCellStateTensor = 1;
constexpr int kOutputTensor = 2;

void* Init(TfLiteContext* context, const char* buffer, size_t length) {
  auto* op_data = new OpData();
  op_data->kernel_type = kTfLiteLSTMFullKernel;
  context->AddTensors(context, /*tensors_to_add=*/7,
                      &op_data->scratch_tensor_index);
  return op_data;
}

// Check that input tensor dimensions matches with each other.
TfLiteStatus CheckInputTensorDimensions(TfLiteContext* context,
                                        TfLiteNode* node, int n_input,
                                        int n_output, int n_cell) {
  const auto* params = reinterpret_cast<TfLiteLSTMParams*>(node->builtin_data);

  // Making sure clipping parameters have valid values.
  // == 0 means no clipping
  //  > 0 means clipping
  TF_LITE_ENSURE(context, params->cell_clip >= 0);
  TF_LITE_ENSURE(context, params->proj_clip >= 0);

  const TfLiteTensor* input_to_input_weights =
      GetOptionalInputTensor(context, node, kInputToInputWeightsTensor);
  if (input_to_input_weights != nullptr) {
    TF_LITE_ENSURE_EQ(context, input_to_input_weights->dims->size, 2);
    TF_LITE_ENSURE_EQ(context, input_to_input_weights->dims->data[0], n_cell);
    TF_LITE_ENSURE_EQ(context, input_to_input_weights->dims->data[1], n_input);
  }

  const TfLiteTensor* input_to_forget_weights =
      GetInput(context, node, kInputToForgetWeightsTensor);
  TF_LITE_ENSURE_EQ(context, input_to_forget_weights->dims->size, 2);
  TF_LITE_ENSURE_EQ(context, input_to_forget_weights->dims->data[0], n_cell);
  TF_LITE_ENSURE_EQ(context, input_to_forget_weights->dims->data[1], n_input);

  const TfLiteTensor* input_to_cell_weights =
      GetInput(context, node, kInputToCellWeightsTensor);
  TF_LITE_ENSURE_EQ(context, input_to_cell_weights->dims->size, 2);
  TF_LITE_ENSURE_EQ(context, input_to_cell_weights->dims->data[0], n_cell);
  TF_LITE_ENSURE_EQ(context, input_to_cell_weights->dims->data[1], n_input);

  const TfLiteTensor* recurrent_to_input_weights =
      GetOptionalInputTensor(context, node, kRecurrentToInputWeightsTensor);
  if (recurrent_to_input_weights != nullptr) {
    TF_LITE_ENSURE_EQ(context, recurrent_to_input_weights->dims->size, 2);
    TF_LITE_ENSURE_EQ(context, recurrent_to_input_weights->dims->data[0],
                      n_cell);
    TF_LITE_ENSURE_EQ(context, recurrent_to_input_weights->dims->data[1],
                      n_output);
  }

  const TfLiteTensor* recurrent_to_forget_weights =
      GetInput(context, node, kRecurrentToForgetWeightsTensor);
  TF_LITE_ENSURE_EQ(context, recurrent_to_forget_weights->dims->size, 2);
  TF_LITE_ENSURE_EQ(context, recurrent_to_forget_weights->dims->data[0],
                    n_cell);
  TF_LITE_ENSURE_EQ(context, recurrent_to_forget_weights->dims->data[1],
                    n_output);

  const TfLiteTensor* recurrent_to_cell_weights =
      GetInput(context, node, kRecurrentToCellWeightsTensor);
  TF_LITE_ENSURE_EQ(context, recurrent_to_cell_weights->dims->size, 2);
  TF_LITE_ENSURE_EQ(context, recurrent_to_cell_weights->dims->data[0], n_cell);
  TF_LITE_ENSURE_EQ(context, recurrent_to_cell_weights->dims->data[1],
                    n_output);

  // We make sure the input-gate's parameters are either both present (regular
  // LSTM) or not at all (CIFG-LSTM).
  const bool cifg_weights_all_or_none =
      ((input_to_input_weights != nullptr) &&
       (recurrent_to_input_weights != nullptr)) ||
      ((input_to_input_weights == nullptr) &&
       (recurrent_to_input_weights == nullptr));
  TF_LITE_ENSURE(context, cifg_weights_all_or_none == true);

  const TfLiteTensor* cell_to_input_weights =
      GetOptionalInputTensor(context, node, kCellToInputWeightsTensor);
  if (cell_to_input_weights) {
    TF_LITE_ENSURE_EQ(context, cell_to_input_weights->dims->size, 1);
    TF_LITE_ENSURE_EQ(context, cell_to_input_weights->dims->data[0], n_cell);
  }

  const TfLiteTensor* cell_to_forget_weights =
      GetOptionalInputTensor(context, node, kCellToForgetWeightsTensor);
  if (cell_to_forget_weights) {
    TF_LITE_ENSURE_EQ(context, cell_to_forget_weights->dims->size, 1);
    TF_LITE_ENSURE_EQ(context, cell_to_forget_weights->dims->data[0], n_cell);
  }

  const TfLiteTensor* cell_to_output_weights =
      GetOptionalInputTensor(context, node, kCellToOutputWeightsTensor);
  if (cell_to_output_weights) {
    TF_LITE_ENSURE_EQ(context, cell_to_output_weights->dims->size, 1);
    TF_LITE_ENSURE_EQ(context, cell_to_output_weights->dims->data[0], n_cell);
  }

  // Making sure the peephole weights are there all or none.
  const bool use_cifg = (input_to_input_weights == nullptr);
  const bool peephole_weights_all_or_none =
      ((cell_to_input_weights != nullptr || use_cifg) &&
       (cell_to_forget_weights != nullptr) &&
       (cell_to_output_weights != nullptr)) ||
      ((cell_to_input_weights == nullptr) &&
       (cell_to_forget_weights == nullptr) &&
       (cell_to_output_weights == nullptr));
  TF_LITE_ENSURE(context, peephole_weights_all_or_none == true);

  // Make sure the input gate bias is present only when not a CIFG-LSTM.
  const TfLiteTensor* input_gate_bias =
      GetOptionalInputTensor(context, node, kInputGateBiasTensor);
  if (use_cifg) {
    TF_LITE_ENSURE_EQ(context, input_gate_bias, nullptr);
  } else {
    TF_LITE_ENSURE_EQ(context, input_gate_bias->dims->size, 1);
    TF_LITE_ENSURE_EQ(context, input_gate_bias->dims->data[0], n_cell);
  }

  const TfLiteTensor* forget_gate_bias =
      GetInput(context, node, kForgetGateBiasTensor);
  TF_LITE_ENSURE_EQ(context, forget_gate_bias->dims->size, 1);
  TF_LITE_ENSURE_EQ(context, forget_gate_bias->dims->data[0], n_cell);

  const TfLiteTensor* cell_bias = GetInput(context, node, kCellGateBiasTensor);
  TF_LITE_ENSURE_EQ(context, cell_bias->dims->size, 1);
  TF_LITE_ENSURE_EQ(context, cell_bias->dims->data[0], n_cell);

  const TfLiteTensor* output_gate_bias =
      GetInput(context, node, kOutputGateBiasTensor);
  TF_LITE_ENSURE_EQ(context, output_gate_bias->dims->size, 1);
  TF_LITE_ENSURE_EQ(context, output_gate_bias->dims->data[0], n_cell);

  const TfLiteTensor* projection_weights =
      GetOptionalInputTensor(context, node, kProjectionWeightsTensor);
  if (projection_weights != nullptr) {
    TF_LITE_ENSURE_EQ(context, projection_weights->dims->size, 2);
    TF_LITE_ENSURE_EQ(context, projection_weights->dims->data[0], n_output);
    TF_LITE_ENSURE_EQ(context, projection_weights->dims->data[1], n_cell);
  }

  const TfLiteTensor* projection_bias =
      GetOptionalInputTensor(context, node, kProjectionBiasTensor);
  if (projection_bias != nullptr) {
    TF_LITE_ENSURE_EQ(context, projection_bias->dims->size, 1);
    TF_LITE_ENSURE_EQ(context, projection_bias->dims->data[0], n_output);
  }

  // Making sure the projection tensors are consistent:
  // 1) If projection weight is not present, then projection bias should not be
  // present.
  // 2) If projection weight is present, then projection bias is optional.
  // TODO(ghodrat): make sure this is correct.
  const bool projection_tensors_consistent =
      ((projection_weights != nullptr) || (projection_bias == nullptr));
  TF_LITE_ENSURE(context, projection_tensors_consistent == true);

  return kTfLiteOk;
}

// Resize the output, state tensors based on the sizes of the input tensors.
// Allocate a temporary scratch tensor. Also check that the sizes of the input
// tensors match each other.
TfLiteStatus Prepare(TfLiteContext* context, TfLiteNode* node) {
  OpData* op_data = reinterpret_cast<OpData*>(node->user_data);

  TF_LITE_ENSURE_EQ(context, node->outputs->size, 3);

  // True if the node is using input variable state tensors. It means:
  // * The state tensors are defined as inputs. In this case it would be the
  //   19th and 20th input tensors.
  // * Otherwise, the output tensors are used to store states.
  bool use_input_variable_states;
  if (node->inputs->size == 20) {
    use_input_variable_states = true;
    op_data->activation_state_tensor_index =
        node->inputs->data[kInputActivationStateTensor];
    op_data->cell_state_tensor_index =
        node->inputs->data[kInputCellStateTensor];
  } else if (node->inputs->size == 18) {
    use_input_variable_states = false;
    op_data->activation_state_tensor_index =
        node->outputs->data[kOutputStateTensor];
    op_data->cell_state_tensor_index = node->outputs->data[kCellStateTensor];
  } else {
    context->ReportError(
        context, "The LSTM Full kernel expects 18 or 20 inputs. Got %d inputs",
        node->inputs->size);
    return kTfLiteError;
  }

  // Inferring batch size, number of outputs and number of cells from the
  // input tensors.
  const TfLiteTensor* input = GetInput(context, node, kInputTensor);
  TF_LITE_ENSURE_EQ(context, input->type, kTfLiteFloat32);
  TF_LITE_ENSURE(context, input->dims->size > 1);
  const int n_batch = input->dims->data[0];
  const int n_input = input->dims->data[1];

  const TfLiteTensor* input_to_output_weights =
      GetInput(context, node, kInputToOutputWeightsTensor);
  const int n_cell = input_to_output_weights->dims->data[0];
  TF_LITE_ENSURE_EQ(context, input_to_output_weights->dims->size, 2);
  TF_LITE_ENSURE_EQ(context, input_to_output_weights->dims->data[1], n_input);

  const TfLiteTensor* recurrent_to_output_weights =
      GetInput(context, node, kRecurrentToOutputWeightsTensor);
  TF_LITE_ENSURE_EQ(context, recurrent_to_output_weights->dims->size, 2);
  TF_LITE_ENSURE_EQ(context, recurrent_to_output_weights->dims->data[0],
                    n_cell);
  const int n_output = recurrent_to_output_weights->dims->data[1];

  // Check that input tensor dimensions matches with each other.
  TF_LITE_ENSURE_OK(context, CheckInputTensorDimensions(context, node, n_input,
                                                        n_output, n_cell));

  // Get the pointer to output, activation_state and cell_state tensors.
  TfLiteTensor* output = GetOutput(context, node, kOutputTensor);

  TfLiteTensor* activation_state =
      &context->tensors[op_data->activation_state_tensor_index];
  TfLiteTensor* cell_state =
      &context->tensors[op_data->cell_state_tensor_index];

  if (use_input_variable_states) {
    // Check the shape of input state tensors.
    // These tensor may be 1D or 2D. It's fine as long as the total size is
    // correct.
    TF_LITE_ENSURE_EQ(context, NumElements(activation_state),
                      n_batch * n_output);
    TF_LITE_ENSURE_EQ(context, NumElements(cell_state), n_batch * n_cell);
  } else {
    // If the state tensors are outputs, this function takes the
    // responsibility to resize the state tensors.
    TfLiteIntArray* activation_state_size = TfLiteIntArrayCreate(2);
    activation_state_size->data[0] = n_batch;
    activation_state_size->data[1] = n_output;
    TF_LITE_ENSURE_OK(context, context->ResizeTensor(context, activation_state,
                                                     activation_state_size));

    TfLiteIntArray* cell_size = TfLiteIntArrayCreate(2);
    cell_size->data[0] = n_batch;
    cell_size->data[1] = n_cell;
    TF_LITE_ENSURE_OK(context,
                      context->ResizeTensor(context, cell_state, cell_size));
    // Mark state tensors as persistent tensors.
    activation_state->allocation_type = kTfLiteArenaRwPersistent;
    cell_state->allocation_type = kTfLiteArenaRwPersistent;
  }

  // Resize the output tensors.
  TfLiteIntArray* output_size = TfLiteIntArrayCreate(2);
  output_size->data[0] = n_batch;
  output_size->data[1] = n_output;
  TF_LITE_ENSURE_OK(context,
                    context->ResizeTensor(context, output, output_size));

  // The weights are of consistent type, so it suffices to check one.
  // TODO(mirkov): create a utility/macro for this check, so all Ops can use it.
  const bool is_hybrid_op = (input_to_output_weights->type == kTfLiteUInt8 &&
                             input->type == kTfLiteFloat32);

  TfLiteIntArrayFree(node->temporaries);
  if (is_hybrid_op) {
    node->temporaries = TfLiteIntArrayCreate(7);
  } else {
    node->temporaries = TfLiteIntArrayCreate(1);
  }
  node->temporaries->data[0] = op_data->scratch_tensor_index;

  // Create a scratch buffer tensor.
  TfLiteTensor* scratch_buffer = GetTemporary(context, node, /*index=*/0);
  scratch_buffer->type = input->type;
  scratch_buffer->allocation_type = kTfLiteArenaRw;

  const TfLiteTensor* input_to_input_weights =
      GetOptionalInputTensor(context, node, kInputToInputWeightsTensor);
  const bool use_cifg = (input_to_input_weights == nullptr);
  TfLiteIntArray* scratch_buffer_size = TfLiteIntArrayCreate(2);
  scratch_buffer_size->data[0] = n_batch;
  if (use_cifg) {
    // Reserving space for Cell, Forget, Output gates
    scratch_buffer_size->data[1] = n_cell * 3;
  } else {
    // Reserving space for Input, Cell, Forget, Output gates
    scratch_buffer_size->data[1] = n_cell * 4;
  }
  TF_LITE_ENSURE_OK(context, context->ResizeTensor(context, scratch_buffer,
                                                   scratch_buffer_size));

  if (is_hybrid_op) {
    // Allocate temporary tensors to store quantized values of input,
    // activation_state and cell_state tensors.
    node->temporaries->data[1] = op_data->scratch_tensor_index + 1;
    TfLiteTensor* input_quantized = GetTemporary(context, node, /*index=*/1);
    input_quantized->type = kTfLiteUInt8;
    input_quantized->allocation_type = kTfLiteArenaRw;
    if (!TfLiteIntArrayEqual(input_quantized->dims, input->dims)) {
      TfLiteIntArray* input_quantized_size = TfLiteIntArrayCopy(input->dims);
      TF_LITE_ENSURE_OK(context, context->ResizeTensor(context, input_quantized,
                                                       input_quantized_size));
    }
    node->temporaries->data[2] = op_data->scratch_tensor_index + 2;
    TfLiteTensor* activation_state_quantized =
        GetTemporary(context, node, /*index=*/2);
    activation_state_quantized->type = kTfLiteUInt8;
    activation_state_quantized->allocation_type = kTfLiteArenaRw;
    if (!TfLiteIntArrayEqual(activation_state_quantized->dims,
                             activation_state->dims)) {
      TfLiteIntArray* activation_state_quantized_size =
          TfLiteIntArrayCopy(activation_state->dims);
      TF_LITE_ENSURE_OK(
          context, context->ResizeTensor(context, activation_state_quantized,
                                         activation_state_quantized_size));
    }
    node->temporaries->data[3] = op_data->scratch_tensor_index + 3;
    TfLiteTensor* cell_state_quantized =
        GetTemporary(context, node, /*index=*/3);
    cell_state_quantized->type = kTfLiteUInt8;
    cell_state_quantized->allocation_type = kTfLiteArenaRw;
    if (!TfLiteIntArrayEqual(cell_state_quantized->dims, cell_state->dims)) {
      TfLiteIntArray* cell_state_quantized_size =
          TfLiteIntArrayCopy(cell_state->dims);
      TF_LITE_ENSURE_OK(context,
                        context->ResizeTensor(context, cell_state_quantized,
                                              cell_state_quantized_size));
    }

    // Allocate temporary tensors to store scaling factors and product scaling
    // factors. The latter is a convenience storage which allows to quantize
    // a vector once (which produces the scaling factors) and multiply it with
    // different matrices (which requires multiplying the scaling factors with
    // the scaling factor of the matrix).
    node->temporaries->data[4] = op_data->scratch_tensor_index + 4;
    TfLiteTensor* scaling_factors = GetTemporary(context, node, /*index=*/4);
    scaling_factors->type = kTfLiteFloat32;
    scaling_factors->allocation_type = kTfLiteArenaRw;
    TfLiteIntArray* scaling_factors_size = TfLiteIntArrayCreate(1);
    scaling_factors_size->data[0] = n_batch;
    if (!TfLiteIntArrayEqual(scaling_factors->dims, scaling_factors_size)) {
      TF_LITE_ENSURE_OK(context, context->ResizeTensor(context, scaling_factors,
                                                       scaling_factors_size));
    }
    node->temporaries->data[5] = op_data->scratch_tensor_index + 5;
    TfLiteTensor* prod_scaling_factors =
        GetTemporary(context, node, /*index=*/5);
    prod_scaling_factors->type = kTfLiteFloat32;
    prod_scaling_factors->allocation_type = kTfLiteArenaRw;
    TfLiteIntArray* prod_scaling_factors_size = TfLiteIntArrayCreate(1);
    prod_scaling_factors_size->data[0] = n_batch;
    if (!TfLiteIntArrayEqual(prod_scaling_factors->dims,
                             prod_scaling_factors_size)) {
      TF_LITE_ENSURE_OK(context,
                        context->ResizeTensor(context, prod_scaling_factors,
                                              prod_scaling_factors_size));
    }

    // Allocate a temporary tensor to store the recovered cell weights. Since
    // this is used for diagonal matrices, only need to store n_cell values.
    node->temporaries->data[6] = op_data->scratch_tensor_index + 6;
    TfLiteTensor* recovered_cell_weights =
        GetTemporary(context, node, /*index=*/6);
    recovered_cell_weights->type = kTfLiteFloat32;
    recovered_cell_weights->allocation_type = kTfLiteArenaRw;
    TfLiteIntArray* recovered_cell_weights_size = TfLiteIntArrayCreate(1);
    recovered_cell_weights_size->data[0] = n_cell;
    if (!TfLiteIntArrayEqual(recovered_cell_weights->dims,
                             recovered_cell_weights_size)) {
      TF_LITE_ENSURE_OK(context,
                        context->ResizeTensor(context, recovered_cell_weights,
                                              recovered_cell_weights_size));
    }
  }
  return kTfLiteOk;
}

// The LSTM Op engine.
TfLiteStatus EvalFloat(
    const TfLiteTensor* input, const TfLiteTensor* input_to_input_weights,
    const TfLiteTensor* input_to_forget_weights,
    const TfLiteTensor* input_to_cell_weights,
    const TfLiteTensor* input_to_output_weights,
    const TfLiteTensor* recurrent_to_input_weights,
    const TfLiteTensor* recurrent_to_forget_weights,
    const TfLiteTensor* recurrent_to_cell_weights,
    const TfLiteTensor* recurrent_to_output_weights,
    const TfLiteTensor* cell_to_input_weights,
    const TfLiteTensor* cell_to_forget_weights,
    const TfLiteTensor* cell_to_output_weights,
    const TfLiteTensor* input_gate_bias, const TfLiteTensor* forget_gate_bias,
    const TfLiteTensor* cell_bias, const TfLiteTensor* output_gate_bias,
    const TfLiteTensor* projection_weights, const TfLiteTensor* projection_bias,
    const TfLiteLSTMParams* params, TfLiteTensor* scratch_buffer,
    TfLiteTensor* activation_state, TfLiteTensor* cell_state,
    TfLiteTensor* output) {
  const int n_batch = input->dims->data[0];
  const int n_input = 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];

  // 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);
  const bool use_peephole = (cell_to_output_weights != nullptr);

  float* input_gate_scratch = nullptr;
  float* cell_scratch = nullptr;
  float* forget_gate_scratch = nullptr;
  float* output_gate_scratch = nullptr;
  if (use_cifg) {
    cell_scratch = scratch_buffer->data.f;
    forget_gate_scratch = scratch_buffer->data.f + n_cell * n_batch;
    output_gate_scratch = scratch_buffer->data.f + 2 * n_cell * n_batch;
  } else {
    input_gate_scratch = scratch_buffer->data.f;
    cell_scratch = scratch_buffer->data.f + n_cell * n_batch;
    forget_gate_scratch = scratch_buffer->data.f + 2 * n_cell * n_batch;
    output_gate_scratch = scratch_buffer->data.f + 3 * n_cell * n_batch;
  }

  // Check optional tensors, the respective pointers can be null.
  const float* input_to_input_weights_ptr =
      (use_cifg) ? nullptr : input_to_input_weights->data.f;
  const float* recurrent_to_input_weights_ptr =
      (use_cifg) ? nullptr : recurrent_to_input_weights->data.f;
  const float* input_gate_bias_ptr =
      (use_cifg) ? nullptr : input_gate_bias->data.f;
  const float* cell_to_input_weights_ptr =
      (use_peephole && !use_cifg) ? cell_to_input_weights->data.f : nullptr;
  const float* cell_to_forget_weights_ptr =
      (use_peephole) ? cell_to_forget_weights->data.f : nullptr;
  const float* cell_to_output_weights_ptr =
      (use_peephole) ? cell_to_output_weights->data.f : nullptr;
  const float* projection_weights_ptr =
      (projection_weights == nullptr) ? nullptr : projection_weights->data.f;
  const float* projection_bias_ptr =
      (projection_bias == nullptr) ? nullptr : projection_bias->data.f;

  // Required tensors, pointers are non-null.
  const float* input_ptr_batch = input->data.f;
  const float* input_to_forget_weights_ptr = input_to_forget_weights->data.f;
  const float* input_to_cell_weights_ptr = input_to_cell_weights->data.f;
  const float* input_to_output_weights_ptr = input_to_output_weights->data.f;
  const float* recurrent_to_forget_weights_ptr =
      recurrent_to_forget_weights->data.f;
  const float* recurrent_to_cell_weights_ptr =
      recurrent_to_cell_weights->data.f;
  const float* recurrent_to_output_weights_ptr =
      recurrent_to_output_weights->data.f;
  const float* forget_gate_bias_ptr = forget_gate_bias->data.f;
  const float* cell_bias_ptr = cell_bias->data.f;
  const float* output_gate_bias_ptr = output_gate_bias->data.f;

  float* activation_state_ptr = activation_state->data.f;
  float* cell_state_ptr = cell_state->data.f;
  float* output_ptr_batch = output->data.f;

  kernel_utils::LstmStep(
      input_ptr_batch, input_to_input_weights_ptr, input_to_forget_weights_ptr,
      input_to_cell_weights_ptr, 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,
      cell_to_input_weights_ptr, cell_to_forget_weights_ptr,
      cell_to_output_weights_ptr, input_gate_bias_ptr, forget_gate_bias_ptr,
      cell_bias_ptr, output_gate_bias_ptr, projection_weights_ptr,
      projection_bias_ptr, params, n_batch, n_cell, n_input, n_output,
      activation_state_ptr, cell_state_ptr, input_gate_scratch,
      forget_gate_scratch, cell_scratch, output_gate_scratch, output_ptr_batch);

  return kTfLiteOk;
}

TfLiteStatus EvalHybrid(
    const TfLiteTensor* input, const TfLiteTensor* input_to_input_weights,
    const TfLiteTensor* input_to_forget_weights,
    const TfLiteTensor* input_to_cell_weights,
    const TfLiteTensor* input_to_output_weights,
    const TfLiteTensor* recurrent_to_input_weights,
    const TfLiteTensor* recurrent_to_forget_weights,
    const TfLiteTensor* recurrent_to_cell_weights,
    const TfLiteTensor* recurrent_to_output_weights,
    const TfLiteTensor* cell_to_input_weights,
    const TfLiteTensor* cell_to_forget_weights,
    const TfLiteTensor* cell_to_output_weights,
    const TfLiteTensor* input_gate_bias, const TfLiteTensor* forget_gate_bias,
    const TfLiteTensor* cell_bias, const TfLiteTensor* output_gate_bias,
    const TfLiteTensor* projection_weights, const TfLiteTensor* projection_bias,
    const TfLiteLSTMParams* params, TfLiteTensor* scratch_buffer,
    TfLiteTensor* scaling_factors, TfLiteTensor* prod_scaling_factors,
    TfLiteTensor* recovered_cell_weights, TfLiteTensor* input_quantized,
    TfLiteTensor* activation_state_quantized,
    TfLiteTensor* cell_state_quantized, TfLiteTensor* activation_state,
    TfLiteTensor* cell_state, TfLiteTensor* output) {
  const int n_batch = input->dims->data[0];
  const int n_input = 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];

  // 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);
  const bool use_peephole = (cell_to_output_weights != nullptr);

  float* input_gate_scratch = nullptr;
  float* cell_scratch = nullptr;
  float* forget_gate_scratch = nullptr;
  float* output_gate_scratch = nullptr;
  if (use_cifg) {
    cell_scratch = scratch_buffer->data.f;
    forget_gate_scratch = scratch_buffer->data.f + n_cell * n_batch;
    output_gate_scratch = scratch_buffer->data.f + 2 * n_cell * n_batch;
  } else {
    input_gate_scratch = scratch_buffer->data.f;
    cell_scratch = scratch_buffer->data.f + n_cell * n_batch;
    forget_gate_scratch = scratch_buffer->data.f + 2 * n_cell * n_batch;
    output_gate_scratch = scratch_buffer->data.f + 3 * n_cell * n_batch;
  }

  // Check optional tensors, the respective pointers can be null.
  int8_t* input_to_input_weights_ptr = nullptr;
  float input_to_input_weights_scale = 1.0f;
  int8_t* recurrent_to_input_weights_ptr = nullptr;
  float recurrent_to_input_weights_scale = 1.0f;
  float* input_gate_bias_ptr = nullptr;
  if (!use_cifg) {
    input_to_input_weights_ptr =
        reinterpret_cast<int8_t*>(input_to_input_weights->data.uint8);
    recurrent_to_input_weights_ptr =
        reinterpret_cast<int8_t*>(recurrent_to_input_weights->data.uint8);
    input_gate_bias_ptr = input_gate_bias->data.f;
    input_to_input_weights_scale = input_to_input_weights->params.scale;
    recurrent_to_input_weights_scale = recurrent_to_input_weights->params.scale;
  }

  int8_t* cell_to_input_weights_ptr = nullptr;
  int8_t* cell_to_forget_weights_ptr = nullptr;
  int8_t* cell_to_output_weights_ptr = nullptr;
  float cell_to_input_weights_scale = 1.0f;
  float cell_to_forget_weights_scale = 1.0f;
  float cell_to_output_weights_scale = 1.0f;
  if (use_peephole) {
    if (!use_cifg) {
      cell_to_input_weights_ptr =
          reinterpret_cast<int8_t*>(cell_to_input_weights->data.uint8);
      cell_to_input_weights_scale = cell_to_input_weights->params.scale;
    }
    cell_to_forget_weights_ptr =
        reinterpret_cast<int8_t*>(cell_to_forget_weights->data.uint8);
    cell_to_output_weights_ptr =
        reinterpret_cast<int8_t*>(cell_to_output_weights->data.uint8);
    cell_to_forget_weights_scale = cell_to_forget_weights->params.scale;
    cell_to_output_weights_scale = cell_to_output_weights->params.scale;
  }

  const int8_t* projection_weights_ptr =
      (projection_weights == nullptr)
          ? nullptr
          : reinterpret_cast<int8_t*>(projection_weights->data.uint8);
  const float projection_weights_scale =
      (projection_weights == nullptr) ? 1.0f : projection_weights->params.scale;
  const float* projection_bias_ptr =
      (projection_bias == nullptr) ? nullptr : projection_bias->data.f;

  // Required tensors, pointers are non-null.
  const float* input_ptr_batch = input->data.f;
  const int8_t* input_to_forget_weights_ptr =
      reinterpret_cast<int8_t*>(input_to_forget_weights->data.uint8);
  const float input_to_forget_weights_scale =
      input_to_forget_weights->params.scale;
  const int8_t* input_to_cell_weights_ptr =
      reinterpret_cast<int8_t*>(input_to_cell_weights->data.uint8);
  const float input_to_cell_weights_scale = input_to_cell_weights->params.scale;
  const int8_t* input_to_output_weights_ptr =
      reinterpret_cast<int8_t*>(input_to_output_weights->data.uint8);
  const float input_to_output_weights_scale =
      input_to_output_weights->params.scale;
  const int8_t* recurrent_to_forget_weights_ptr =
      reinterpret_cast<int8_t*>(recurrent_to_forget_weights->data.uint8);
  const float recurrent_to_forget_weights_scale =
      recurrent_to_forget_weights->params.scale;
  const int8_t* recurrent_to_cell_weights_ptr =
      reinterpret_cast<int8_t*>(recurrent_to_cell_weights->data.uint8);
  const float recurrent_to_cell_weights_scale =
      recurrent_to_cell_weights->params.scale;
  const int8_t* recurrent_to_output_weights_ptr =
      reinterpret_cast<int8_t*>(recurrent_to_output_weights->data.uint8);
  const float recurrent_to_output_weights_scale =
      recurrent_to_output_weights->params.scale;
  const float* forget_gate_bias_ptr = forget_gate_bias->data.f;
  const float* cell_bias_ptr = cell_bias->data.f;
  const float* output_gate_bias_ptr = output_gate_bias->data.f;

  float* activation_state_ptr = activation_state->data.f;
  float* cell_state_ptr = cell_state->data.f;
  float* output_ptr_batch = output->data.f;

  // Temporary storage for quantized values and scaling factors.
  int8_t* quantized_input_ptr =
      reinterpret_cast<int8_t*>(input_quantized->data.uint8);
  int8_t* quantized_activation_state_ptr =
      reinterpret_cast<int8_t*>(activation_state_quantized->data.uint8);
  int8_t* quantized_cell_state_ptr =
      reinterpret_cast<int8_t*>(cell_state_quantized->data.uint8);
  float* scaling_factors_ptr = scaling_factors->data.f;
  float* prod_scaling_factors_ptr = prod_scaling_factors->data.f;
  float* recovered_cell_weights_ptr = recovered_cell_weights->data.f;

  kernel_utils::LstmStep(
      input_ptr_batch, input_to_input_weights_ptr, input_to_input_weights_scale,
      input_to_forget_weights_ptr, input_to_forget_weights_scale,
      input_to_cell_weights_ptr, input_to_cell_weights_scale,
      input_to_output_weights_ptr, input_to_output_weights_scale,
      recurrent_to_input_weights_ptr, recurrent_to_input_weights_scale,
      recurrent_to_forget_weights_ptr, recurrent_to_forget_weights_scale,
      recurrent_to_cell_weights_ptr, recurrent_to_cell_weights_scale,
      recurrent_to_output_weights_ptr, recurrent_to_output_weights_scale,
      cell_to_input_weights_ptr, cell_to_input_weights_scale,
      cell_to_forget_weights_ptr, cell_to_forget_weights_scale,
      cell_to_output_weights_ptr, cell_to_output_weights_scale,
      input_gate_bias_ptr, forget_gate_bias_ptr, cell_bias_ptr,
      output_gate_bias_ptr, projection_weights_ptr, projection_weights_scale,
      projection_bias_ptr, params, n_batch, n_cell, n_input, n_output,
      input_gate_scratch, forget_gate_scratch, cell_scratch,
      output_gate_scratch, scaling_factors_ptr, prod_scaling_factors_ptr,
      recovered_cell_weights_ptr, quantized_input_ptr,
      quantized_activation_state_ptr, quantized_cell_state_ptr,
      activation_state_ptr, cell_state_ptr, output_ptr_batch);

  return kTfLiteOk;
}

TfLiteStatus Eval(TfLiteContext* context, TfLiteNode* node) {
  const auto* params = reinterpret_cast<TfLiteLSTMParams*>(node->builtin_data);
  OpData* op_data = reinterpret_cast<OpData*>(node->user_data);

  const TfLiteTensor* input = GetInput(context, node, kInputTensor);

  const TfLiteTensor* input_to_input_weights =
      GetOptionalInputTensor(context, node, kInputToInputWeightsTensor);
  const TfLiteTensor* input_to_forget_weights =
      GetInput(context, node, kInputToForgetWeightsTensor);
  const TfLiteTensor* input_to_cell_weights =
      GetInput(context, node, kInputToCellWeightsTensor);
  const TfLiteTensor* input_to_output_weights =
      GetInput(context, node, kInputToOutputWeightsTensor);

  const TfLiteTensor* recurrent_to_input_weights =
      GetOptionalInputTensor(context, node, kRecurrentToInputWeightsTensor);
  const TfLiteTensor* recurrent_to_forget_weights =
      GetInput(context, node, kRecurrentToForgetWeightsTensor);
  const TfLiteTensor* recurrent_to_cell_weights =
      GetInput(context, node, kRecurrentToCellWeightsTensor);
  const TfLiteTensor* recurrent_to_output_weights =
      GetInput(context, node, kRecurrentToOutputWeightsTensor);

  const TfLiteTensor* cell_to_input_weights =
      GetOptionalInputTensor(context, node, kCellToInputWeightsTensor);
  const TfLiteTensor* cell_to_forget_weights =
      GetOptionalInputTensor(context, node, kCellToForgetWeightsTensor);
  const TfLiteTensor* cell_to_output_weights =
      GetOptionalInputTensor(context, node, kCellToOutputWeightsTensor);

  const TfLiteTensor* input_gate_bias =
      GetOptionalInputTensor(context, node, kInputGateBiasTensor);
  const TfLiteTensor* forget_gate_bias =
      GetInput(context, node, kForgetGateBiasTensor);
  const TfLiteTensor* cell_bias = GetInput(context, node, kCellGateBiasTensor);
  const TfLiteTensor* output_gate_bias =
      GetInput(context, node, kOutputGateBiasTensor);

  const TfLiteTensor* projection_weights =
      GetOptionalInputTensor(context, node, kProjectionWeightsTensor);
  const TfLiteTensor* projection_bias =
      GetOptionalInputTensor(context, node, kProjectionBiasTensor);

  // Index the scratch buffers pointers to the global scratch buffer.
  TfLiteTensor* scratch_buffer = GetTemporary(context, node, /*index=*/0);

  TfLiteTensor* activation_state =
      &context->tensors[op_data->activation_state_tensor_index];
  TfLiteTensor* cell_state =
      &context->tensors[op_data->cell_state_tensor_index];

  TfLiteTensor* output = GetOutput(context, node, kOutputTensor);

  // TODO(mirkov): add a check that weights are all uint8s or all floats.
  switch (input_to_output_weights->type) {
    case kTfLiteFloat32: {
      return EvalFloat(input, input_to_input_weights, input_to_forget_weights,
                       input_to_cell_weights, input_to_output_weights,
                       recurrent_to_input_weights, recurrent_to_forget_weights,
                       recurrent_to_cell_weights, recurrent_to_output_weights,
                       cell_to_input_weights, cell_to_forget_weights,
                       cell_to_output_weights, input_gate_bias,
                       forget_gate_bias, cell_bias, output_gate_bias,
                       projection_weights, projection_bias, params,
                       scratch_buffer, activation_state, cell_state, output);
    }
    case kTfLiteUInt8: {
      TfLiteTensor* input_quantized = GetTemporary(context, node, /*index=*/1);
      TfLiteTensor* activation_state_quantized =
          GetTemporary(context, node, /*index=*/2);
      TfLiteTensor* cell_state_quantized =
          GetTemporary(context, node, /*index=*/3);
      TfLiteTensor* scaling_factors = GetTemporary(context, node, /*index=*/4);
      TfLiteTensor* prod_scaling_factors =
          GetTemporary(context, node, /*index=*/5);
      TfLiteTensor* recovered_cell_weights =
          GetTemporary(context, node, /*index=*/6);
      return EvalHybrid(
          input, input_to_input_weights, input_to_forget_weights,
          input_to_cell_weights, input_to_output_weights,
          recurrent_to_input_weights, recurrent_to_forget_weights,
          recurrent_to_cell_weights, recurrent_to_output_weights,
          cell_to_input_weights, cell_to_forget_weights, cell_to_output_weights,
          input_gate_bias, forget_gate_bias, cell_bias, output_gate_bias,
          projection_weights, projection_bias, params, scratch_buffer,
          scaling_factors, prod_scaling_factors, recovered_cell_weights,
          input_quantized, activation_state_quantized, cell_state_quantized,
          activation_state, cell_state, output);
    }
    default:
      context->ReportError(context, "Type %d is not currently supported.",
                           input_to_output_weights->type);
      return kTfLiteError;
  }
  return kTfLiteOk;
}

}  // namespace full

// For basic kernel (5-inputs).
namespace basic {

enum InputTensor {
  kInputData = 0,
  kInputPrevActivation = 1,
  kInputWeights = 2,
  kInputBiases = 3,
  kInputPrevState = 4,
  kInputNum = 5,
};

enum OutputTensor {
  kOutputActivation = 0,
  kOutputState = 1,
  kOutputConcatTemp = 2,
  kOutputActivationTemp = 3,
  kOutputNum = 4,
};

void* Init(TfLiteContext* context, const char* buffer, size_t length) {
  auto* op_data = new OpData();
  op_data->kernel_type = kTfLiteLSTMBasicKernel;
  // `scratch_tensor_index` is unused in this kernel.
  op_data->scratch_tensor_index = -1;
  return op_data;
}

TfLiteStatus Prepare(TfLiteContext* context, TfLiteNode* node) {
  TF_LITE_ENSURE(context, node->inputs->size == kInputNum);
  TF_LITE_ENSURE(context, node->outputs->size == kOutputNum);

  const TfLiteTensor* input = GetInput(context, node, kInputData);
  const TfLiteTensor* prev_activation =
      GetInput(context, node, kInputPrevActivation);
  const TfLiteTensor* weights = GetInput(context, node, kInputWeights);
  const TfLiteTensor* bias = GetInput(context, node, kInputBiases);
  const TfLiteTensor* prev_state = GetInput(context, node, kInputPrevState);

  TF_LITE_ENSURE_EQ(context, input->dims->size, 2);
  const int num_batches = input->dims->data[0];
  const int input_depth = input->dims->data[1];

  TF_LITE_ENSURE_EQ(context, prev_activation->dims->size, 2);
  TF_LITE_ENSURE_EQ(context, prev_activation->dims->data[0], num_batches);
  const int activation_depth = prev_activation->dims->data[1];
  const int total_depth = input_depth + activation_depth;

  TF_LITE_ENSURE_EQ(context, weights->dims->size, 2);
  TF_LITE_ENSURE_EQ(context, weights->dims->data[0], 4 * activation_depth);
  TF_LITE_ENSURE_EQ(context, weights->dims->data[1], total_depth);

  TF_LITE_ENSURE_EQ(context, bias->dims->size, 1);
  TF_LITE_ENSURE_EQ(context, bias->dims->data[0], 4 * activation_depth);

  TF_LITE_ENSURE_EQ(context, prev_state->dims->size, 2);
  TF_LITE_ENSURE_EQ(context, prev_state->dims->data[0], num_batches);
  TF_LITE_ENSURE_EQ(context, prev_state->dims->data[1], activation_depth);

  TfLiteTensor* activation_out = GetOutput(context, node, kOutputActivation);
  TfLiteTensor* state_out = GetOutput(context, node, kOutputState);
  TfLiteTensor* concat_temp = GetOutput(context, node, kOutputConcatTemp);
  TfLiteTensor* activation_temp =
      GetOutput(context, node, kOutputActivationTemp);

  TF_LITE_ENSURE_OK(context, context->ResizeTensor(
                                 context, activation_out,
                                 TfLiteIntArrayCopy(prev_activation->dims)));
  TF_LITE_ENSURE_OK(
      context, context->ResizeTensor(context, state_out,
                                     TfLiteIntArrayCopy(prev_state->dims)));

  TfLiteIntArray* concat_temp_size = TfLiteIntArrayCreate(2);
  concat_temp_size->data[0] = num_batches;
  concat_temp_size->data[1] = total_depth;
  TF_LITE_ENSURE_OK(
      context, context->ResizeTensor(context, concat_temp, concat_temp_size));
  TfLiteIntArray* activation_temp_size = TfLiteIntArrayCreate(2);
  activation_temp_size->data[0] = num_batches;
  activation_temp_size->data[1] = 4 * activation_depth;
  TF_LITE_ENSURE_OK(context, context->ResizeTensor(context, activation_temp,
                                                   activation_temp_size));

  // Set the state tensors as persistent.
  for (auto index : {kInputPrevActivation, kInputPrevState}) {
    TfLiteTensor* tensor = &context->tensors[node->inputs->data[index]];
    tensor->allocation_type = kTfLiteArenaRwPersistent;
  }
  return kTfLiteOk;
}

TfLiteStatus Eval(TfLiteContext* context, TfLiteNode* node) {
  const TfLiteTensor* input = GetInput(context, node, kInputData);
  const TfLiteTensor* prev_activation =
      GetInput(context, node, kInputPrevActivation);
  const TfLiteTensor* weights = GetInput(context, node, kInputWeights);
  const TfLiteTensor* bias = GetInput(context, node, kInputBiases);
  const TfLiteTensor* prev_state = GetInput(context, node, kInputPrevState);

  TfLiteTensor* activation_out = GetOutput(context, node, kOutputActivation);
  TfLiteTensor* state_out = GetOutput(context, node, kOutputState);
  TfLiteTensor* concat_temp = GetOutput(context, node, kOutputConcatTemp);
  TfLiteTensor* activation_temp =
      GetOutput(context, node, kOutputActivationTemp);

  if (input->type == kTfLiteFloat32 &&
      prev_activation->type == kTfLiteFloat32 &&
      weights->type == kTfLiteFloat32 && bias->type == kTfLiteFloat32 &&
      prev_state->type == kTfLiteFloat32 && state_out->type == kTfLiteFloat32 &&
      activation_out->type == kTfLiteFloat32 &&
      concat_temp->type == kTfLiteFloat32 &&
      activation_temp->type == kTfLiteFloat32) {
    optimized_ops::LstmCell(
        // Inputs.
        GetTensorData<float>(input), GetTensorDims(input),
        GetTensorData<float>(prev_activation), GetTensorDims(prev_activation),
        GetTensorData<float>(weights), GetTensorDims(weights),
        GetTensorData<float>(bias), GetTensorDims(bias),
        GetTensorData<float>(prev_state), GetTensorDims(prev_state),
        // Outputs.
        GetTensorData<float>(state_out), GetTensorDims(state_out),
        GetTensorData<float>(activation_out), GetTensorDims(activation_out),
        GetTensorData<float>(concat_temp), GetTensorDims(concat_temp),
        GetTensorData<float>(activation_temp), GetTensorDims(activation_temp));
  } else if (input->type == kTfLiteUInt8 &&
             prev_activation->type == kTfLiteUInt8 &&
             weights->type == kTfLiteUInt8 && bias->type == kTfLiteInt32 &&
             prev_state->type == kTfLiteInt16 &&
             state_out->type == kTfLiteInt16 &&
             activation_out->type == kTfLiteUInt8 &&
             concat_temp->type == kTfLiteUInt8 &&
             activation_temp->type == kTfLiteInt16) {
    gemmlowp::GemmContext* gemm_context = gemm_support::GetFromContext(context);
    int state_scale_log2_rounded;
    if (!CheckedLog2(state_out->params.scale, &state_scale_log2_rounded)) {
      context->ReportError(
          context,
          "The internal state of a LSTM cell must have a power-of-two scale.");
      return kTfLiteError;
    }
    const int state_integer_bits = 15 + state_scale_log2_rounded;
    if (state_integer_bits != 4) {
      context->ReportError(context,
                           "The only case of quantized LstmCell currently "
                           "supported is with StateIntegerBits==4");
      return kTfLiteError;
    }

    double real_accum_multiplier = 4096 * bias->params.scale;
    int32 accum_multiplier;
    int accum_shift;
    tflite::QuantizeMultiplier(real_accum_multiplier, &accum_multiplier,
                               &accum_shift);
    optimized_ops::LstmCell<4>(
        // Inputs.
        GetTensorData<uint8_t>(input), GetTensorDims(input),
        GetTensorData<uint8_t>(prev_activation), GetTensorDims(prev_activation),
        GetTensorData<uint8_t>(weights), GetTensorDims(weights),
        GetTensorData<int32_t>(bias), GetTensorDims(bias),
        GetTensorData<int16_t>(prev_state), GetTensorDims(prev_state),
        // Outputs.
        GetTensorData<int16_t>(state_out), GetTensorDims(state_out),
        GetTensorData<uint8_t>(activation_out), GetTensorDims(activation_out),
        GetTensorData<uint8_t>(concat_temp), GetTensorDims(concat_temp),
        GetTensorData<int16_t>(activation_temp), GetTensorDims(activation_temp),
        weights->params.zero_point, accum_multiplier, accum_shift,
        gemm_context);
  } else {
    context->ReportError(context,
                         "Unsupported combination of data types for LstmCell");
    return kTfLiteError;
  }

  // TODO(ycling): Investigate if this copy can be avoided with the 5-inputs
  // LSTM kernel.
  memcpy(prev_activation->data.raw, activation_out->data.raw,
         activation_out->bytes);
  memcpy(prev_state->data.raw, state_out->data.raw, state_out->bytes);

  return kTfLiteOk;
}

}  // namespace basic

void* Init(TfLiteContext* context, const char* buffer, size_t length) {
  gemm_support::IncrementUsageCounter(context);

  const auto* params = reinterpret_cast<const TfLiteLSTMParams*>(buffer);
  switch (params->kernel_type) {
    case kTfLiteLSTMFullKernel:
      return full::Init(context, buffer, length);
    case kTfLiteLSTMBasicKernel:
      return basic::Init(context, buffer, length);
  }
}
void Free(TfLiteContext* context, void* buffer) {
  gemm_support::DecrementUsageCounter(context);

  delete reinterpret_cast<OpData*>(buffer);
}

TfLiteStatus Prepare(TfLiteContext* context, TfLiteNode* node) {
  const auto* op_data = reinterpret_cast<const OpData*>(node->user_data);
  switch (op_data->kernel_type) {
    case kTfLiteLSTMFullKernel:
      return full::Prepare(context, node);
    case kTfLiteLSTMBasicKernel:
      return basic::Prepare(context, node);
  }
}

TfLiteStatus Eval(TfLiteContext* context, TfLiteNode* node) {
  const auto* op_data = reinterpret_cast<const OpData*>(node->user_data);
  switch (op_data->kernel_type) {
    case kTfLiteLSTMFullKernel:
      return full::Eval(context, node);
    case kTfLiteLSTMBasicKernel:
      return basic::Eval(context, node);
  }
}

}  // namespace lstm

TfLiteRegistration* Register_LSTM() {
  static TfLiteRegistration r = {lstm::Init, lstm::Free, lstm::Prepare,
                                 lstm::Eval};
  return &r;
}

}  // namespace builtin
}  // namespace ops
}  // namespace tflite