core/graph/gradients.cc

/* Copyright 2015 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 <deque>
#include <vector>

#include "tensorflow/core/common_runtime/device.h"
#include "tensorflow/core/common_runtime/executor.h"
#include "tensorflow/core/common_runtime/graph_optimizer.h"
#include "tensorflow/core/framework/function.h"
#include "tensorflow/core/framework/node_def.pb.h"
#include "tensorflow/core/framework/node_def_util.h"
#include "tensorflow/core/framework/op.h"
#include "tensorflow/core/framework/op_kernel.h"
#include "tensorflow/core/graph/algorithm.h"
#include "tensorflow/core/graph/gradients.h"
#include "tensorflow/core/graph/graph_constructor.h"
#include "tensorflow/core/graph/optimizer_cse.h"
#include "tensorflow/core/lib/gtl/map_util.h"
#include "tensorflow/core/platform/macros.h"

namespace tensorflow {

// TODO(andydavis) Remove some of the code duplicated between this module
// and that in 'common_runtime/function.cc'.
// A few string constant used throughout this module.
static const char* const kGradientOp = "SymbolicGradient";
static const char* const kNodeLabel = "Func";

string NodeOut::name() const {
  if (index == 0) {
    return node->name();
  } else {
    return strings::StrCat(node->name(), ":", index);
  }
}

DataType NodeOut::dtype() const { return node->output_type(index); }

struct NodeOutHash {
  uint64 operator()(const NodeOut& x) const {
    return Hash64(reinterpret_cast<const char*>(&x.node), sizeof(Node*),
                  x.index);
  }
};

struct NodeOutEq {
  bool operator()(const NodeOut& x, const NodeOut& y) const {
    return (x.node == y.node) && (x.index == y.index);
  }
};

static Node* AddZerosLike(Graph* g, NodeOut input) {
  DCHECK_LT(0, input.dtype());
  DCHECK_LT(input.dtype(), DT_FLOAT_REF);
  NodeDef ndef;
  ndef.set_name(g->NewName(kNodeLabel));
  ndef.set_op("ZerosLike");
  ndef.add_input(input.name());
  AddNodeAttr("T", input.dtype(), &ndef);
  Status s;
  Node* ret = g->AddNode(ndef, &s);
  TF_CHECK_OK(s);
  g->AddEdge(input.node, input.index, ret, 0);
  return ret;
}

static Node* AddSymGrad(Graph* g, Node* n, gtl::ArraySlice<NodeOut> grads) {
  const int num_x = n->num_inputs();
  const int num_y = n->num_outputs();
  CHECK_EQ(num_y, grads.size());

  NodeDef ndef;
  ndef.set_name(g->NewName(kNodeLabel));
  ndef.set_op(kGradientOp);

  // The gradient node should have num_x + num_y inputs.
  std::vector<NodeOut> n_inputs(num_x);
  for (const Edge* e : n->in_edges()) {
    if (e->IsControlEdge()) continue;
    n_inputs[e->dst_input()] = {e->src(), e->src_output()};
  }
  DataTypeVector in_types;
  for (const NodeOut& nout : n_inputs) {
    ndef.add_input(nout.name());
    in_types.push_back(nout.dtype());
  }
  for (const NodeOut& nout : grads) {
    ndef.add_input(nout.name());
    in_types.push_back(nout.dtype());
  }
  CHECK_EQ(ndef.input_size(), num_x + num_y);

  AddNodeAttr("Tin", in_types, &ndef);

  // The gradient node's outputs have the same types as the node 'n's
  // inputs.
  AddNodeAttr("Tout", n->input_types(), &ndef);
  NameAttrList func;
  func.set_name(n->type_string());
  for (const auto& attr : n->attrs()) {
    (*func.mutable_attr())[attr.first] = attr.second;
  }
  AddNodeAttr("f", func, &ndef);
  Status s;
  Node* ret = g->AddNode(ndef, &s);
  TF_CHECK_OK(s);
  return ret;
}

class SymbolicGradientBuilder {
 public:
  SymbolicGradientBuilder(gtl::ArraySlice<NodeOut> y_node_outputs,
                          gtl::ArraySlice<NodeOut> x_node_outputs,
                          gtl::ArraySlice<NodeOut> y_grad_node_outputs,
                          std::vector<NodeOut>* x_grad_node_outputs,
                          Graph* graph);

  Status Compute();

 private:
  gtl::ArraySlice<NodeOut> y_node_outputs_;
  gtl::ArraySlice<NodeOut> x_node_outputs_;
  gtl::ArraySlice<NodeOut> y_grad_node_outputs_;
  std::vector<NodeOut>* x_grad_node_outputs_;
  Graph* graph_;  // Not owned.

  // A vector of output endpoints which represents backpropagated
  // gradients
  typedef std::vector<NodeOut> BackpropedGradients;

  // backprops_ is a map from a node output to its accumulated
  // gradients.  When a node output has accumulated all its
  // gradients, we add a node which sums them up.
  std::unordered_map<NodeOut, BackpropedGradients, NodeOutHash, NodeOutEq>
      backprops_;

  // pending[i] is count-down counter for i-th node's expected
  // backprops.  When pending[i] becomes zero, we collected all
  // backprop gradients for all outputs of the ith-node.
  std::vector<int> pending_;

  // 'ready' keeps track of nodes that have been completely
  // backpropped. Initially, for every output y of the function f, we
  // add dy as an input of the gradient function.
  std::deque<Node*> ready_;

  // The set of node ids at which to stop backprop.
  std::unordered_set<int> stop_nodes_;

  // Initialize pending_ and ready_.
  void InitBackprop();

  // In the original function body, there is a forward edge from 'src'
  // to 'dst', when the backprop algorithm constructs the node
  // 'dst_grad' which computes the gradient, we need to propagate it
  // to 'src'.
  void BackpropAlongEdge(const NodeOut& dst_grad, const NodeOut& src);
  void BackpropZerosAlongEdge(const NodeOut& src);

  NodeOut SumGradients(const NodeOut& src);

  TF_DISALLOW_COPY_AND_ASSIGN(SymbolicGradientBuilder);
};

SymbolicGradientBuilder::SymbolicGradientBuilder(
    gtl::ArraySlice<NodeOut> y_node_outputs,
    gtl::ArraySlice<NodeOut> x_node_outputs,
    gtl::ArraySlice<NodeOut> y_grad_node_outputs,
    std::vector<NodeOut>* x_grad_node_outputs, Graph* graph)
    : y_node_outputs_(y_node_outputs),
      x_node_outputs_(x_node_outputs),
      y_grad_node_outputs_(y_grad_node_outputs),
      x_grad_node_outputs_(x_grad_node_outputs),
      graph_(graph) {
  CHECK_EQ(y_node_outputs_.size(), y_grad_node_outputs.size());
  x_grad_node_outputs_->clear();
  x_grad_node_outputs_->resize(x_node_outputs_.size());
  stop_nodes_.reserve(x_node_outputs_.size());
  for (int i = 0; i < x_node_outputs_.size(); ++i) {
    stop_nodes_.insert(x_node_outputs_[i].node->id());
  }
}

void SymbolicGradientBuilder::BackpropAlongEdge(const NodeOut& dst_grad,
                                                const NodeOut& src) {
  CHECK_NOTNULL(src.node);
  auto iter = backprops_.find(src);
  if (iter != backprops_.end()) {
    auto* grads = &iter->second;
    grads->push_back(dst_grad);
    if (--pending_[src.node->id()] == 0) {
      ready_.push_back(src.node);
    }
  }
}

void SymbolicGradientBuilder::BackpropZerosAlongEdge(const NodeOut& src) {
  CHECK_NOTNULL(src.node);
  auto iter = backprops_.find(src);
  if (iter != backprops_.end()) {
    if (--pending_[src.node->id()] == 0) {
      ready_.push_back(src.node);
    }
  }
}

void SymbolicGradientBuilder::InitBackprop() {
  pending_.resize(graph_->num_node_ids(), 0);
  {
    backprops_.clear();
    std::unordered_set<Node*> visited;
    std::deque<Node*> queue;
    for (const NodeOut& nout : x_node_outputs_) {
      queue.push_back(nout.node);
      visited.insert(nout.node);
    }

    // Going forward to figure out which endpoints need backprop-ed.
    // A node's endpoints need to be backprop-ed only if one of the
    // arg node can reach the node via data edges.
    while (!queue.empty()) {
      Node* n = queue.front();
      queue.pop_front();
      for (int i = 0; i < n->num_outputs(); ++i) {
        backprops_[{n, i}].clear();
      }
      int num_expected_backprops = 0;
      for (const Edge* e : n->out_edges()) {
        if (e->IsControlEdge()) continue;
        ++num_expected_backprops;
        if (visited.find(e->dst()) == visited.end()) {
          queue.push_back(e->dst());
          visited.insert(e->dst());
        }
      }
      pending_[n->id()] = num_expected_backprops;
    }
  }

  {
    const int num_y = y_grad_node_outputs_.size();
    for (int i = 0; i < num_y; ++i) {
      Node* y = y_node_outputs_[i].node;
      for (const Edge* e : y->in_edges()) {
        if (e->IsControlEdge()) continue;
        BackpropAlongEdge(y_grad_node_outputs_[i], {e->src(), e->src_output()});
      }
    }
  }
  CHECK(!ready_.empty());
}

NodeOut SymbolicGradientBuilder::SumGradients(const NodeOut& src) {
  const DataType dtype = src.dtype();
  auto iter = backprops_.find(src);
  CHECK(iter != backprops_.end());
  const auto& grads = iter->second;
  if (grads.empty()) {
    // Nothing propagated back. The best we can come up is zeros.
    Node* zero_like = AddZerosLike(graph_, src);
    return {zero_like, 0};
  }
  if (grads.size() == 1) {
    // Just one backprop edge.
    return grads[0];
  }
  // Otherwise, adds backprop-ed gradients.
  NodeDef ndef;
  ndef.set_name(graph_->NewName(kNodeLabel));
  ndef.set_op("AddN");  // N-way Add
  for (const NodeOut& nout : grads) {
    ndef.add_input(nout.name());
  }
  AddNodeAttr("N", static_cast<int64>(grads.size()), &ndef);
  AddNodeAttr("T", dtype, &ndef);
  Status s;
  Node* add = graph_->AddNode(ndef, &s);
  TF_CHECK_OK(s);
  for (size_t i = 0; i < grads.size(); ++i) {
    const NodeOut& nout = grads[i];
    graph_->AddEdge(nout.node, nout.index, add, i);
  }
  return {add, 0};
}

static bool IsPrimitiveOpWithNoGrad(const string& func) {
  gradient::Creator creator;
  Status s = gradient::GetOpGradientCreator(func, &creator);
  return s.ok() && (creator == nullptr);
}

Status SymbolicGradientBuilder::Compute() {
  // Initialize backprops.
  InitBackprop();

  // Backward propagation.
  gtl::InlinedVector<NodeOut, 8> dy;
  while (!ready_.empty()) {
    // n has collected all gradients.
    Node* n = ready_.front();
    ready_.pop_front();

    // "n" has num_x inputs and num_y outputs.
    const int num_x = n->num_inputs();
    const int num_y = n->num_outputs();

    auto iter = stop_nodes_.find(n->id());
    if (iter != stop_nodes_.end()) {
      // Stop backprop.
      // TODO(andydavis) Support stop nodes with more than one output.
      CHECK_EQ(1, num_y);
      continue;
    }

    // dy[i] is the sum of i-th output's backpropped gradients.
    dy.clear();
    dy.resize(num_y, {nullptr, 0});
    for (int i = 0; i < num_y; ++i) {
      dy[i] = SumGradients({n, i});
    }

    if (IsPrimitiveOpWithNoGrad(n->type_string())) {
      // No grad defined for this op: Backprop zeros along the in edges.
      for (const Edge* e : n->in_edges()) {
        if (e->IsControlEdge()) continue;
        BackpropZerosAlongEdge({e->src(), e->src_output()});
      }
      continue;
    }

    // Adds a gradient node with num_x + num_y inputs and num_x
    // outputs.
    // TODO(andydavis) Support primitive gradient ops.
    Node* grad = AddSymGrad(graph_, n, dy);
    for (const Edge* e : n->in_edges()) {
      if (e->IsControlEdge()) continue;
      graph_->AddEdge(e->src(), e->src_output(), grad, e->dst_input());
    }
    for (int i = 0; i < num_y; ++i) {
      graph_->AddEdge(dy[i].node, dy[i].index, grad, num_x + i);
    }

    // Backprops along the in edges.
    for (const Edge* e : n->in_edges()) {
      if (e->IsControlEdge()) continue;
      BackpropAlongEdge({grad, e->dst_input()}, {e->src(), e->src_output()});
    }
  }

  for (int i = 0; i < x_node_outputs_.size(); ++i) {
    (*x_grad_node_outputs_)[i] = SumGradients(x_node_outputs_[i]);
  }

  return Status::OK();
}

Status AddSymbolicGradients(gtl::ArraySlice<NodeOut> y_node_outputs,
                            gtl::ArraySlice<NodeOut> x_node_outputs,
                            gtl::ArraySlice<NodeOut> y_grad_node_outputs,
                            std::vector<NodeOut>* x_grad_node_outputs,
                            Graph* graph) {
  SymbolicGradientBuilder builder(y_node_outputs, x_node_outputs,
                                  y_grad_node_outputs, x_grad_node_outputs,
                                  graph);
  return builder.Compute();
}

}  // end namespace tensorflow