graph_pattern_detector.cc 79.1 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14
// Copyright (c) 2018 PaddlePaddle 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.

15
#include <algorithm>
Q
Qiao Longfei 已提交
16
#include <array>
17
#include <memory>
18
#include <string>
19 20
#include <unordered_map>
#include <unordered_set>
21 22 23
#include <vector>

#include "paddle/fluid/framework/ir/graph_helper.h"
24
#include "paddle/fluid/framework/ir/graph_pattern_detector.h"
25
#include "paddle/fluid/framework/ir/graph_traits.h"
26
#include "paddle/fluid/framework/ir/graph_viz_pass.h"
C
chengduo 已提交
27
#include "paddle/fluid/framework/operator.h"
28
#include "paddle/fluid/platform/enforce.h"
Y
Yan Chunwei 已提交
29
#include "paddle/fluid/string/pretty_log.h"
Y
Yan Chunwei 已提交
30
#include "paddle/fluid/string/printf.h"
31

32 33 34 35
namespace paddle {
namespace framework {
namespace ir {

Y
Yan Chunwei 已提交
36 37 38
using string::PrettyLog;
using string::Style;

39 40
size_t PDPattern::id_ = 0UL;

C
chengduo 已提交
41
PDNode *PDPattern::NewNode(const std::string &name) {
Y
Yan Chunwei 已提交
42
  if (!name.empty()) {
43 44 45 46
    PADDLE_ENFORCE_EQ(
        node_map_.count(name), 0UL,
        platform::errors::PreconditionNotMet(
            "PDNode's name should be unique, get duplicate [%s]", name));
Y
Yan Chunwei 已提交
47 48 49
  }

  nodes_.emplace_back(new PDNode(this, name));
C
chengduo 已提交
50
  auto *cur = nodes_.back().get();
Y
Yan Chunwei 已提交
51 52 53 54
  node_map_[name] = cur;
  return cur;
}

C
chengduo 已提交
55
PDNode *PDPattern::NewNode(PDNode::teller_t &&teller, const std::string &name) {
56
  if (!name.empty()) {
57 58 59 60
    PADDLE_ENFORCE_EQ(
        node_map_.count(name), 0UL,
        platform::errors::PreconditionNotMet(
            "PDNode's name should be unique, get duplicate [%s]", name));
61 62
  }

63
  nodes_.emplace_back(new PDNode(std::move(teller), this, name));
C
chengduo 已提交
64
  auto *cur = nodes_.back().get();
65
  node_map_[name] = cur;
66 67 68
  return cur;
}

C
chengduo 已提交
69
PDNode *PDPattern::RetrieveNode(const std::string &id) const {
70 71 72 73 74 75 76 77
  auto it = node_map_.find(id);
  if (it == node_map_.end()) {
    return nullptr;
  }

  return it->second;
}

C
chengduo 已提交
78
void PDPattern::AddEdge(PDNode *a, PDNode *b) {
79 80 81 82
  PADDLE_ENFORCE_NOT_NULL(
      a, platform::errors::NotFound("PDNode %s is not found.", a->name()));
  PADDLE_ENFORCE_NOT_NULL(
      b, platform::errors::NotFound("PDNode %s is not found.", b->name()));
83 84
  PADDLE_ENFORCE_NE(a, b, platform::errors::PermissionDenied(
                              "Cannot connect the same node in the graph."));
85 86 87
  edges_.emplace_back(a, b);
}

C
chengduo 已提交
88
void GraphPatternDetector::operator()(Graph *graph,
89
                                      GraphPatternDetector::handle_t handler) {
90 91 92 93
  if (!MarkPDNodesInGraph(*graph)) {
    return;
  }

94 95 96
  auto subgraphs = DetectPatterns();
  UniquePatterns(&subgraphs);
  RemoveOverlappedMatch(&subgraphs);
Y
Yan Chunwei 已提交
97
  ValidateByNodeRole(&subgraphs);
98

Y
Yan Chunwei 已提交
99
  if (subgraphs.empty()) return;
100
  LOG(INFO) << "---  detected " << subgraphs.size() << " subgraphs";
101
  int id = 0;
C
chengduo 已提交
102
  for (auto &g : subgraphs) {
M
minqiyang 已提交
103
    VLOG(3) << "optimizing #" << id++ << " subgraph";
104 105 106 107
    handler(g, graph);
  }
}

C
chengduo 已提交
108
bool GraphPatternDetector::MarkPDNodesInGraph(const ir::Graph &graph) {
M
minqiyang 已提交
109
  VLOG(3) << "mark pdnodes in graph";
110 111
  if (graph.Nodes().empty()) return false;

C
chengduo 已提交
112 113
  for (auto &node : GraphTraits::DFS(graph)) {
    for (const auto &pdnode : pattern_.nodes()) {
114
      if (pdnode->Tell(&node)) {
115
        VLOG(4) << "Node " << node.Name() << " marked as " << pdnode->name();
116 117 118 119
        pdnodes2nodes_[pdnode.get()].insert(&node);
      }
    }
  }
Y
Yan Chunwei 已提交
120
  // Check to early stop if some PDNode can't find matched Node.
C
chengduo 已提交
121
  for (auto &pdnode : pattern_.nodes()) {
Y
Yan Chunwei 已提交
122
    if (!pdnodes2nodes_.count(pdnode.get())) {
M
minqiyang 已提交
123
      VLOG(4) << pdnode->name() << " can't find matched Node, early stop";
Y
Yan Chunwei 已提交
124
      // return false;
Y
Yan Chunwei 已提交
125 126
    }
  }
M
minqiyang 已提交
127
  VLOG(3) << pdnodes2nodes_.size() << " nodes marked";
128

129 130 131
  return !pdnodes2nodes_.empty();
}

Y
Yan Chunwei 已提交
132
// The intermediate Nodes can only link to the nodes inside the pattern, or this
T
tianshuo78520a 已提交
133
// subgraph will be dropped.
Y
Yan Chunwei 已提交
134
void GraphPatternDetector::ValidateByNodeRole(
C
chengduo 已提交
135
    std::vector<GraphPatternDetector::subgraph_t> *subgraphs) {
Y
Yan Chunwei 已提交
136 137 138 139 140
  std::vector<GraphPatternDetector::subgraph_t> result;

  subgraphs->erase(
      std::remove_if(
          subgraphs->begin(), subgraphs->end(),
C
chengduo 已提交
141
          [](const GraphPatternDetector::subgraph_t &subgraph) -> bool {
Y
Yan Chunwei 已提交
142
            // Collect the inputs and outputs.
C
chengduo 已提交
143 144
            std::unordered_set<Node *> ios;
            for (auto &item : subgraph) {
Y
Yan Chunwei 已提交
145 146 147 148
              if (!item.first->IsIntermediate()) {
                ios.insert(item.second);
              }
            }
C
chengduo 已提交
149
            for (auto &item : subgraph) {
Y
Yan Chunwei 已提交
150
              if (item.first->IsIntermediate()) {
C
chengduo 已提交
151
                for (auto *x : item.second->inputs) {
Y
Yan Chunwei 已提交
152 153 154 155
                  if (!ios.count(x)) {
                    return true;
                  }
                }
C
chengduo 已提交
156
                for (auto *x : item.second->outputs) {
Y
Yan Chunwei 已提交
157 158 159 160 161 162 163 164 165 166 167
                  if (!ios.count(x)) {
                    return true;
                  }
                }
              }
            }
            return false;
          }),
      subgraphs->end());
}

168
struct HitGroup {
C
chengduo 已提交
169
  std::unordered_map<PDNode *, Node *> roles;
170

C
chengduo 已提交
171
  bool Match(Node *node, PDNode *pat) {
172
    if (nodes_.count(node)) {
T
Tao Luo 已提交
173 174 175 176 177
      if (roles.count(pat) && roles[pat] == node) return true;
      return false;
    } else {
      if (roles.count(pat) && roles[pat] != node) return false;
      return true;
178
    }
179 180
  }

C
chengduo 已提交
181
  void Register(Node *node, PDNode *pat) {
182 183 184 185 186
    roles[pat] = node;
    nodes_.insert(node);
  }

 private:
C
chengduo 已提交
187
  std::unordered_set<Node *> nodes_;
188 189 190
};

// Tell whether Node a links to b.
C
chengduo 已提交
191 192
bool IsNodesLink(Node *a, Node *b) {
  for (auto *node : a->outputs) {
193 194 195 196 197 198 199
    if (b == node) {
      return true;
    }
  }
  return false;
}

200 201
std::vector<GraphPatternDetector::subgraph_t>
GraphPatternDetector::DetectPatterns() {
202
  // Init empty subgraphs.
203
  std::vector<GraphPatternDetector::subgraph_t> result;
204
  std::vector<HitGroup> init_groups;
205
  std::array<std::vector<HitGroup>, 2> bi_records;
C
chengduo 已提交
206
  auto *first_pnode = pattern_.edges().empty() ? pattern().nodes().front().get()
207
                                               : pattern_.edges().front().first;
208
  if (!pdnodes2nodes_.count(first_pnode)) return result;
C
chengduo 已提交
209
  for (auto *node : pdnodes2nodes_[first_pnode]) {
210 211 212 213 214 215 216 217 218 219
    HitGroup group;
    group.roles[first_pnode] = node;
    init_groups.emplace_back(group);
  }

  int step = 0;
  bi_records[0] = std::move(init_groups);

  // Extend a PDNode to subgraphs by deducing the connection relations defined
  // in edges of PDNodes.
C
chengduo 已提交
220
  for (const auto &edge : pattern_.edges()) {
M
minqiyang 已提交
221
    VLOG(4) << "check " << edge.first->name() << " -> " << edge.second->name();
Y
Yan Chunwei 已提交
222
    // TODO(Superjomn) Fix bug here, the groups might be duplicate here.
223 224
    // Each role has two PDNodes, which indicates two roles.
    // Detect two Nodes that can match these two roles and they are connected.
C
chengduo 已提交
225 226
    auto &pre_groups = bi_records[step % 2];
    auto &cur_groups = bi_records[1 - (step++ % 2)];
227
    cur_groups.clear();
228
    if (pre_groups.empty()) break;
229
    // source -> target
C
chengduo 已提交
230 231
    for (Node *source : pdnodes2nodes_[edge.first]) {
      for (Node *target : pdnodes2nodes_[edge.second]) {
M
minqiyang 已提交
232
        VLOG(8) << "check " << source->id() << " -- " << target->id();
233
        // TODO(Superjomn) add some prune strategies.
C
chengduo 已提交
234
        for (const auto &group : pre_groups) {
T
Tao Luo 已提交
235 236 237 238 239 240
          if (IsNodesLink(source, target)) {
            HitGroup new_group = group;
            bool flag = new_group.Match(source, edge.first) &&
                        new_group.Match(target, edge.second);
            if (flag) {
              new_group.Register(source, edge.first);
241 242 243 244 245 246 247 248
              new_group.Register(target, edge.second);
              cur_groups.push_back(new_group);
              // TODO(Superjomn) need to unique
            }
          }
        }
      }
    }
M
minqiyang 已提交
249
    VLOG(3) << "step " << step << " get records: " << cur_groups.size();
C
chengduo 已提交
250 251
    for (auto &group : cur_groups) {
      for (auto &item : group.roles) {
M
minqiyang 已提交
252
        VLOG(4) << "node " << item.second->id() << " as " << item.first->name();
Y
Yan Chunwei 已提交
253
      }
M
minqiyang 已提交
254
      VLOG(4) << "=========================================================";
Y
Yan Chunwei 已提交
255
    }
256 257
  }

C
chengduo 已提交
258
  for (auto &group : bi_records[step % 2]) {
259
    GraphPatternDetector::subgraph_t subgraph;
C
chengduo 已提交
260
    for (auto &role : group.roles) {
261 262 263 264 265 266 267
      subgraph.emplace(role.first, role.second);
    }
    result.emplace_back(subgraph);
  }
  return result;
}

Y
Yan Chunwei 已提交
268 269
struct GraphItemLessThan {
  bool operator()(const std::pair<PDNode *, Node *> &a,
Y
Yan Chunwei 已提交
270
                  const std::pair<PDNode *, Node *> &b) {
Y
Yan Chunwei 已提交
271 272 273 274 275
    if (a.first != b.first) {
      return a.first < b.first;
    } else {
      return a.second < b.second;
    }
Y
Yan Chunwei 已提交
276
  }
Y
Yan Chunwei 已提交
277
};
Y
Yan Chunwei 已提交
278

279 280
// TODO(Superjomn) enhance the function as it marks unique unique as duplicates
// see https://github.com/PaddlePaddle/Paddle/issues/13550
281
void GraphPatternDetector::UniquePatterns(
C
chengduo 已提交
282
    std::vector<GraphPatternDetector::subgraph_t> *subgraphs) {
283
  if (subgraphs->empty()) return;
284
  std::vector<GraphPatternDetector::subgraph_t> result;
285 286

  std::unordered_set<size_t> set;
Y
Yan Chunwei 已提交
287
  std::hash<std::string> hasher;
C
chengduo 已提交
288
  for (auto &g : *subgraphs) {
Y
Yan Chunwei 已提交
289 290
    // Sort the items in the sub-graph, and transform to a string key.
    std::vector<std::pair<PDNode *, Node *>> sorted_keys(g.begin(), g.end());
Y
Yan Chunwei 已提交
291
    std::sort(sorted_keys.begin(), sorted_keys.end(), GraphItemLessThan());
Y
Yan Chunwei 已提交
292 293 294
    std::stringstream ss;
    for (auto &item : sorted_keys) {
      ss << item.first << ":" << item.second;
295
    }
Y
Yan Chunwei 已提交
296
    auto key = hasher(ss.str());
297 298 299 300 301 302 303 304
    if (!set.count(key)) {
      result.emplace_back(g);
      set.insert(key);
    }
  }
  *subgraphs = result;
}

305
void GraphPatternDetector::RemoveOverlappedMatch(
C
chengduo 已提交
306
    std::vector<subgraph_t> *subgraphs) {
307
  std::vector<subgraph_t> result;
C
chengduo 已提交
308
  std::unordered_set<Node *> node_set;
309

C
chengduo 已提交
310
  for (const auto &subgraph : *subgraphs) {
311
    bool valid = true;
C
chengduo 已提交
312
    for (auto &item : subgraph) {
Y
Yan Chunwei 已提交
313
      if (item.first->IsIntermediate() && node_set.count(item.second)) {
314 315 316 317 318
        valid = false;
        break;
      }
    }
    if (valid) {
C
chengduo 已提交
319
      for (auto &item : subgraph) {
320 321 322 323 324 325 326 327
        node_set.insert(item.second);
      }
      result.push_back(subgraph);
    }
  }
  *subgraphs = result;
}

328 329 330 331 332
std::string PDPattern::DotString() const {
  using inference::analysis::Dot;
  Dot dot;
  int id = 0;
  // Create Nodes
C
chengduo 已提交
333 334
  std::unordered_map<PDNode *, std::string> node2dot;
  for (const auto &node : nodes()) {
335 336 337 338 339
    std::string node_id = "Node" + std::to_string(id++);
    dot.AddNode(node_id, {}, node->name());
    node2dot[node.get()] = node_id;
  }
  // Create Edges
C
chengduo 已提交
340
  for (const auto &edge : edges()) {
341 342 343 344
    if (!node2dot.count(edge.first) || !node2dot.count(edge.second)) {
      LOG(ERROR) << "no node " << edge.first << " " << edge.second;
      continue;
    }
C
chengduo 已提交
345 346
    auto &src = node2dot.at(edge.first);
    auto &trg = node2dot.at(edge.second);
347 348 349 350 351
    dot.AddEdge(src, trg, {});
  }
  return dot.Build();
}

C
chengduo 已提交
352
PDNode &PDNode::LinksTo(const std::vector<PDNode *> &others) {
353
  // extend outlinks.
C
chengduo 已提交
354
  for (PDNode *x : others) {
355 356 357 358 359
    pattern_->AddEdge(this, x);
  }
  return *this;
}

C
chengduo 已提交
360
PDNode &PDNode::LinksFrom(const std::vector<PDNode *> &others) {
361
  // extend outlinks.
C
chengduo 已提交
362
  for (PDNode *x : others) {
363 364 365 366 367
    pattern_->AddEdge(x, this);
  }
  return *this;
}

C
chengduo 已提交
368 369
PDNode *PDNode::assert_is_op() {
  asserts_.emplace_back([](Node *x) { return x && x->IsOp(); });
Y
Yan Chunwei 已提交
370 371
  return this;
}
C
chengduo 已提交
372 373 374

PDNode *PDNode::assert_is_op(const std::string &op_type) {
  asserts_.emplace_back([op_type](Node *x) {
Y
Yan Chunwei 已提交
375 376 377 378
    return x && x->IsOp() && x->Op()->Type() == op_type;
  });
  return this;
}
C
chengduo 已提交
379 380 381 382 383 384

PDNode *PDNode::assert_is_var() {
  asserts_.emplace_back([](Node *x) { return x && x->IsVar(); });
  return this;
}

Z
Zhen Wang 已提交
385 386 387 388 389 390 391
PDNode *PDNode::assert_var_dtype(proto::VarType::Type dtype) {
  assert_is_var();
  asserts_.emplace_back(
      [dtype](Node *x) { return x->Var()->GetDataType() == dtype; });
  return this;
}

C
chengduo 已提交
392 393
PDNode *PDNode::assert_is_not_ctrl_var() {
  asserts_.emplace_back([](Node *x) { return x && !x->IsCtrlVar(); });
Y
Yan Chunwei 已提交
394 395
  return this;
}
C
chengduo 已提交
396 397

PDNode *PDNode::assert_var_not_persistable() {
Y
Yan Chunwei 已提交
398
  assert_is_var();
C
chengduo 已提交
399
  asserts_.emplace_back([](Node *x) { return !x->Var()->Persistable(); });
Y
Yan Chunwei 已提交
400 401
  return this;
}
C
chengduo 已提交
402 403

PDNode *PDNode::assert_is_persistable_var() {
Y
Yan Chunwei 已提交
404
  assert_is_var();
C
chengduo 已提交
405
  asserts_.emplace_back([=](Node *x) { return x->Var()->Persistable(); });
Y
Yan Chunwei 已提交
406 407
  return this;
}
C
chengduo 已提交
408 409 410

PDNode *PDNode::assert_is_op_nth_input(const std::string &op_type,
                                       const std::string &argument, int nth) {
Y
Yan Chunwei 已提交
411 412
  assert_is_var();
  assert_is_op_input(op_type);
C
chengduo 已提交
413 414
  asserts_.emplace_back([=](Node *x) {
    for (auto *op : x->outputs) {
415 416 417
      if (op->IsOp() && op->Op()->Type() == op_type &&
          IsNthInput(x, op, argument, nth))
        return true;
Y
Yan Chunwei 已提交
418 419 420 421 422
    }
    return false;
  });
  return this;
}
C
chengduo 已提交
423 424 425

PDNode *PDNode::assert_is_op_nth_output(const std::string &op_type,
                                        const std::string &argument, int nth) {
Y
Yan Chunwei 已提交
426
  assert_is_var();
C
chengduo 已提交
427 428
  asserts_.emplace_back([=](Node *x) {
    for (auto *op : x->inputs) {
429 430 431
      if (op->IsOp() && op->Op()->Type() == op_type &&
          IsNthOutput(x, op, argument, nth))
        return true;
Y
Yan Chunwei 已提交
432 433 434 435 436
    }
    return false;
  });
  return this;
}
C
chengduo 已提交
437 438

PDNode *PDNode::assert_is_only_input_of_op(const std::string &op_type) {
Y
Yan Chunwei 已提交
439
  assert_is_var();
C
chengduo 已提交
440 441
  asserts_.emplace_back([=](Node *x) {
    for (auto *op : x->outputs) {
Y
Yan Chunwei 已提交
442 443 444 445 446 447 448 449 450
      if (op && op->IsOp() && op->Op() && op->Op()->Type() == op_type &&
          op->inputs.size() == 1) {
        return true;
      }
    }
    return false;
  });
  return this;
}
C
chengduo 已提交
451 452

PDNode *PDNode::assert_is_only_output_of_op(const std::string &op_type) {
Y
Yan Chunwei 已提交
453
  assert_is_var();
C
chengduo 已提交
454 455
  asserts_.emplace_back([=](Node *x) {
    for (auto *op : x->inputs) {
Y
Yan Chunwei 已提交
456 457 458 459 460 461 462 463 464
      if (op && op->IsOp() && op->Op() && op->Op()->Type() == op_type &&
          op->outputs.size() == 1) {
        return true;
      }
    }
    return false;
  });
  return this;
}
C
chengduo 已提交
465 466

PDNode *PDNode::assert_is_op_output(const std::string &op_type) {
Y
Yan Chunwei 已提交
467
  assert_is_var();
C
chengduo 已提交
468 469
  asserts_.emplace_back([=](Node *x) {
    for (auto *op : x->inputs) {
Y
Yan Chunwei 已提交
470 471 472 473 474 475 476 477
      if (op && op->IsOp() && op->Op() && op->Op()->Type() == op_type) {
        return true;
      }
    }
    return false;
  });
  return this;
}
C
chengduo 已提交
478 479 480

PDNode *PDNode::assert_is_op_output(const std::string &op_type,
                                    const std::string &argument) {
481 482 483 484
  assert_is_var();
  assert_is_op_nth_output(op_type, argument, 0);
  return this;
}
Z
Zhen Wang 已提交
485

C
chengduo 已提交
486
PDNode *PDNode::assert_is_op_input(const std::string &op_type) {
Y
Yan Chunwei 已提交
487
  assert_is_var();
C
chengduo 已提交
488 489
  asserts_.emplace_back([=](Node *x) {
    for (auto *op : x->outputs) {
Y
Yan Chunwei 已提交
490 491 492 493 494 495 496 497
      if (op && op->IsOp() && op->Op() && op->Op()->Type() == op_type) {
        return true;
      }
    }
    return false;
  });
  return this;
}
C
chengduo 已提交
498

Z
Zhen Wang 已提交
499 500 501 502 503 504 505 506 507 508
PDNode *PDNode::assert_is_not_op_input(const std::string &argument) {
  assert_is_op();
  asserts_.emplace_back([=](Node *x) {
    auto &ins = x->Op()->Inputs();
    auto iter = ins.find(argument);
    return iter == ins.end() || iter->second.empty();
  });
  return this;
}

C
chengduo 已提交
509 510
PDNode *PDNode::assert_is_op_input(const std::string &op_type,
                                   const std::string &argument) {
511 512 513 514
  assert_is_var();
  assert_is_op_nth_input(op_type, argument, 0);
  return this;
}
C
chengduo 已提交
515 516

PDNode *PDNode::assert_op_has_n_inputs(const std::string &op_type, size_t n) {
Y
Yan Chunwei 已提交
517
  assert_is_op(op_type);
C
chengduo 已提交
518
  asserts_.emplace_back([=](Node *x) { return x->inputs.size() == n; });
Y
Yan Chunwei 已提交
519 520
  return this;
}
C
chengduo 已提交
521 522

PDNode *PDNode::assert_op_has_n_outputs(const std::string &op_type, size_t n) {
Y
Yan Chunwei 已提交
523
  assert_is_op(op_type);
C
chengduo 已提交
524
  asserts_.emplace_back([=](Node *x) { return x->outputs.size() == n; });
Y
Yan Chunwei 已提交
525 526
  return this;
}
C
chengduo 已提交
527

528 529 530 531 532 533 534 535 536 537
PDNode *PDNode::assert_has_n_inputs(size_t n) {
  asserts_.emplace_back([=](Node *x) { return x->inputs.size() == n; });
  return this;
}

PDNode *PDNode::assert_has_n_outputs(size_t n) {
  asserts_.emplace_back([=](Node *x) { return x->outputs.size() == n; });
  return this;
}

C
chengduo 已提交
538
PDNode *PDNode::assert_more(PDNode::teller_t &&teller) {
Y
Yan Chunwei 已提交
539 540 541 542
  asserts_.emplace_back(std::move(teller));
  return this;
}

C
chengduo 已提交
543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625
PDNode *PDNode::assert_is_ops(const std::unordered_set<std::string> &op_types) {
  asserts_.emplace_back([op_types](Node *x) {
    return x && x->IsOp() && op_types.count(x->Op()->Type());
  });
  return this;
}

PDNode *PDNode::assert_is_ops_nth_input(
    const std::unordered_set<std::string> &op_types,
    const std::string &argument, int nth) {
  assert_is_var();
  assert_is_ops_input(op_types);
  asserts_.emplace_back([=](Node *x) {
    for (auto *op : x->outputs) {
      if (op->IsOp() && op_types.count(op->Op()->Type()) &&
          IsNthInput(x, op, argument, nth))
        return true;
    }
    return false;
  });
  return this;
}

PDNode *PDNode::assert_is_ops_nth_output(
    const std::unordered_set<std::string> &op_types,
    const std::string &argument, int nth) {
  assert_is_var();
  asserts_.emplace_back([=](Node *x) {
    for (auto *op : x->inputs) {
      if (op->IsOp() && op_types.count(op->Op()->Type()) &&
          IsNthOutput(x, op, argument, nth))
        return true;
    }
    return false;
  });
  return this;
}
PDNode *PDNode::assert_is_ops_output(
    const std::unordered_set<std::string> &op_types) {
  assert_is_var();
  asserts_.emplace_back([=](Node *x) {
    for (auto *op : x->inputs) {
      if (op && op->IsOp() && op->Op() && op_types.count(op->Op()->Type())) {
        return true;
      }
    }
    return false;
  });
  return this;
}

PDNode *PDNode::assert_is_ops_output(
    const std::unordered_set<std::string> &op_types,
    const std::string &argument) {
  assert_is_var();
  assert_is_ops_nth_output(op_types, argument, 0);
  return this;
}

PDNode *PDNode::assert_is_ops_input(
    const std::unordered_set<std::string> &op_types) {
  assert_is_var();
  asserts_.emplace_back([=](Node *x) {
    for (auto *op : x->outputs) {
      if (op && op->IsOp() && op->Op() && op_types.count(op->Op()->Type())) {
        return true;
      }
    }
    return false;
  });
  return this;
}

PDNode *PDNode::assert_is_ops_input(
    const std::unordered_set<std::string> &op_types,
    const std::string &argument) {
  assert_is_var();
  assert_is_ops_nth_input(op_types, argument, 0);
  return this;
}

bool VarLinksToOp(Node *node, const std::string &op_type) {
  for (auto *out : node->outputs) {
626 627 628 629 630 631
    if (out->IsOp() && out->Op()->Type() == op_type) {
      return true;
    }
  }
  return false;
}
C
chengduo 已提交
632 633

bool IsNthInput(Node *var, Node *op, const std::string &argument, size_t nth) {
634 635 636 637 638 639 640 641
  PADDLE_ENFORCE_EQ(
      var->IsVar(), true,
      platform::errors::InvalidArgument(
          "First parameter of function IsNthInput must be Node::Var"));
  PADDLE_ENFORCE_EQ(
      op->IsOp(), true,
      platform::errors::InvalidArgument(
          "Second parameter of function IsNthInput must be Node::Op"));
642 643
  if (!HasInput(op, argument) || op->Op()->Input(argument).size() <= nth)
    return false;
644 645
  return var->Name() == op->Op()->Input(argument)[nth];
}
C
chengduo 已提交
646

647
bool HasInput(Node *op, const std::string &argument) {
648 649 650 651
  PADDLE_ENFORCE_EQ(
      op->IsOp(), true,
      platform::errors::InvalidArgument(
          "First parameter of function HasInput must be Node::Op"));
652 653 654 655 656 657
  auto const &names = op->Op()->InputNames();
  if (std::find(names.begin(), names.end(), argument) == names.end())
    return false;
  return true;
}

658 659 660 661 662 663 664 665 666 667 668
bool HasOutput(Node *op, const std::string &argument) {
  PADDLE_ENFORCE_EQ(
      op->IsOp(), true,
      platform::errors::InvalidArgument(
          "First parameter of function HasOuput must be Node::Op"));
  auto const &names = op->Op()->OutputNames();
  if (std::find(names.begin(), names.end(), argument) == names.end())
    return false;
  return true;
}

C
chengduo 已提交
669
bool IsNthOutput(Node *var, Node *op, const std::string &argument, size_t nth) {
670 671 672 673 674 675 676 677
  PADDLE_ENFORCE_EQ(
      var->IsVar(), true,
      platform::errors::InvalidArgument(
          "First parameter of function IsNthOutput must be Node::Var"));
  PADDLE_ENFORCE_EQ(
      op->IsOp(), true,
      platform::errors::InvalidArgument(
          "Second parameter of function IsNthOutput must be Node::Op"));
678 679
  if (!HasOutput(op, argument) || op->Op()->Output(argument).size() <= nth)
    return false;
680 681
  return var->Name() == op->Op()->Output(argument)[nth];
}
C
chengduo 已提交
682 683 684 685 686

void GraphSafeRemoveNodes(Graph *graph,
                          const std::unordered_set<const Node *> &nodes) {
  for (auto *node : nodes) {
    graph->RemoveNode(const_cast<Node *>(node));
687 688
  }

C
chengduo 已提交
689
  for (auto *node : graph->Nodes()) {
690 691
    for (auto it = node->inputs.begin(); it != node->inputs.end();) {
      if (nodes.count(*it)) {
C
chengduo 已提交
692
        it = const_cast<Node *>(node)->inputs.erase(it);
693
      } else {
694
        it++;
695
      }
696 697 698
    }
    for (auto it = node->outputs.begin(); it != node->outputs.end();) {
      if (nodes.count(*it)) {
C
chengduo 已提交
699
        it = const_cast<Node *>(node)->outputs.erase(it);
700
      } else {
701
        it++;
702
      }
703 704 705
    }
  }
}
C
chengduo 已提交
706 707 708

bool VarLinksFromOp(Node *node, const std::string &op_type) {
  for (auto *out : node->inputs) {
709 710 711 712 713 714 715
    if (out->IsOp() && out->Op()->Type() == op_type) {
      return true;
    }
  }
  return false;
}

S
Sylwester Fraczek 已提交
716
PDNode *patterns::ConvBN::operator()(paddle::framework::ir::PDNode *conv_input,
717
                                     const std::string &conv_type,
S
Sylwester Fraczek 已提交
718 719
                                     bool with_eltwise_add) {
  // Create Operators
720 721
  conv_input->assert_is_op_input(conv_type, "Input");
  auto *conv_op = pattern->NewNode(conv_repr())->assert_is_op(conv_type);
S
Sylwester Fraczek 已提交
722 723 724 725 726 727 728 729 730 731 732 733 734

  PDNode *eltwise_op = nullptr;
  if (with_eltwise_add) {
    eltwise_op =
        pattern->NewNode(eltwise_repr())->assert_is_op("elementwise_add");
  }
  auto *batch_norm_op =
      pattern->NewNode(batch_norm_repr())->assert_is_op("batch_norm");
  // Create variables
  // Conv Filter
  auto *conv_weight_var = pattern->NewNode(conv_weight_repr())
                              ->AsInput()
                              ->assert_is_persistable_var()
735
                              ->assert_is_op_input(conv_type, "Filter");
S
Sylwester Fraczek 已提交
736 737 738

  auto *conv_out_var = pattern->NewNode(conv_out_repr())
                           ->AsIntermediate()
739
                           ->assert_is_only_output_of_op(conv_type);
S
Sylwester Fraczek 已提交
740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822

  PDNode *eltwise_y_in_var = nullptr;
  PDNode *eltwise_out_var = nullptr;
  if (with_eltwise_add) {
    // Conv output as Bias input
    conv_out_var->assert_is_op_input("elementwise_add", "X");
    // Bias
    eltwise_y_in_var = pattern->NewNode(eltwise_y_in_repr())
                           ->assert_is_op_input("elementwise_add", "Y")
                           ->AsInput();
    eltwise_out_var = pattern->NewNode(eltwise_out_repr())
                          ->AsIntermediate()
                          ->assert_is_only_output_of_op("elementwise_add");
  } else {
    // Conv output as BN input
    conv_out_var->assert_is_op_input("batch_norm", "X");
  }

  // BN Scale
  auto *bn_scale_var = pattern->NewNode(bn_scale_repr())
                           ->AsInput()
                           ->assert_is_persistable_var()
                           ->assert_is_op_input("batch_norm", "Scale");
  // BN Bias
  auto *bn_bias_var = pattern->NewNode(bn_bias_repr())
                          ->AsInput()
                          ->assert_is_persistable_var()
                          ->assert_is_op_input("batch_norm", "Bias");
  // BN Mean
  auto *bn_mean_var = pattern->NewNode(bn_mean_repr())
                          ->AsInput()
                          ->assert_is_persistable_var()
                          ->assert_is_op_input("batch_norm", "Mean");
  // BN Variance
  auto *bn_variance_var = pattern->NewNode(bn_variance_repr())
                              ->AsInput()
                              ->assert_is_persistable_var()
                              ->assert_is_op_input("batch_norm", "Variance");

  // BN output
  auto *bn_out_var = pattern->NewNode(bn_out_repr())
                         ->AsOutput()
                         ->assert_is_op_output("batch_norm");

  auto *bn_mean_out_var = pattern->NewNode(bn_mean_out_repr())
                              ->AsOutput()
                              ->assert_is_op_output("batch_norm", "MeanOut");

  auto *bn_variance_out_var =
      pattern->NewNode(bn_variance_out_repr())
          ->AsOutput()
          ->assert_is_op_output("batch_norm", "VarianceOut");

  auto *bn_saved_mean_var =
      pattern->NewNode(bn_saved_mean_repr())
          ->AsOutput()
          ->assert_is_op_output("batch_norm", "SavedMean");

  auto *bn_saved_variance_var =
      pattern->NewNode(bn_saved_variance_repr())
          ->AsOutput()
          ->assert_is_op_output("batch_norm", "SavedVariance");

  conv_op->LinksFrom({conv_input, conv_weight_var}).LinksTo({conv_out_var});

  if (with_eltwise_add) {
    eltwise_op->LinksFrom({conv_out_var, eltwise_y_in_var})
        .LinksTo({eltwise_out_var});
    batch_norm_op
        ->LinksFrom({eltwise_out_var, bn_scale_var, bn_bias_var, bn_mean_var,
                     bn_variance_var})
        .LinksTo({bn_out_var, bn_mean_out_var, bn_variance_out_var,
                  bn_saved_mean_var, bn_saved_variance_var});
  } else {
    batch_norm_op
        ->LinksFrom({conv_out_var, bn_scale_var, bn_bias_var, bn_mean_var,
                     bn_variance_var})
        .LinksTo({bn_out_var, bn_mean_out_var, bn_variance_out_var,
                  bn_saved_mean_var, bn_saved_variance_var});
  }
  return bn_out_var;
}

823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851
PDNode *patterns::ConvActivation::operator()(
    paddle::framework::ir::PDNode *conv_input, std::string conv_type,
    std::string activation_type) {
  // Create Operators
  conv_input->assert_is_op_input(conv_type, "Input");
  auto *conv_op = pattern->NewNode(conv_repr())->assert_is_op(conv_type);
  auto *activation_op =
      pattern->NewNode(activation_repr())->assert_is_op(activation_type);
  // Create variables
  // Filter
  auto *conv_weight_var = pattern->NewNode(conv_weight_repr())
                              ->AsInput()
                              ->assert_is_persistable_var()
                              ->assert_is_op_input(conv_type, "Filter");
  // intermediate variable, will be removed in the IR after fuse.
  auto *conv_out_var = pattern->NewNode(conv_out_repr())
                           ->AsIntermediate()
                           ->assert_is_only_output_of_op(conv_type)
                           ->assert_is_op_input(activation_type);
  // output
  auto *activation_out_var = pattern->NewNode(activation_out_repr())
                                 ->AsOutput()
                                 ->assert_is_op_output(activation_type);

  conv_op->LinksFrom({conv_input, conv_weight_var}).LinksTo({conv_out_var});
  activation_op->LinksFrom({conv_out_var}).LinksTo({activation_out_var});
  return activation_out_var;
}

T
tensor-tang 已提交
852 853 854 855
PDNode *patterns::SeqConvEltAddRelu::operator()(
    paddle::framework::ir::PDNode *seqconv_input) {
  // Create Operators
  seqconv_input->assert_is_op_input("sequence_conv", "X");
T
tensor-tang 已提交
856 857 858 859
  auto *seqconv_op = pattern->NewNode(seqconv_repr())
                         ->assert_is_op("sequence_conv")
                         ->assert_op_attr<bool>("paddingTrainable", false)
                         ->assert_op_attr<int>("contextStride", 1);
T
tensor-tang 已提交
860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896

  auto *eltadd_op =
      pattern->NewNode(eltadd_repr())->assert_is_op("elementwise_add");
  auto *relu_op = pattern->NewNode(relu_repr())->assert_is_op("relu");
  // Create variables
  // Filter
  auto *seqconv_weight_var =
      pattern->NewNode(seqconv_weight_repr())
          ->AsInput()
          ->assert_is_persistable_var()
          ->assert_is_op_input("sequence_conv", "Filter");
  // Bias
  auto *eltadd_bias_var = pattern->NewNode(eltadd_bias_repr())
                              ->AsInput()
                              ->assert_is_op_input("elementwise_add");
  // intermediate variable, will be removed in the IR after fuse.
  auto *seqconv_out_var = pattern->NewNode(seqconv_out_repr())
                              ->AsIntermediate()
                              ->assert_is_only_output_of_op("sequence_conv")
                              ->assert_is_op_input("elementwise_add");
  auto *eltadd_out_var = pattern->NewNode(eltadd_out_repr())
                             ->AsIntermediate()
                             ->assert_is_only_output_of_op("elementwise_add")
                             ->assert_is_only_input_of_op("relu");
  // output
  auto *relu_out_var = pattern->NewNode(relu_out_repr())
                           ->AsOutput()
                           ->assert_is_op_output("relu");

  seqconv_op->LinksFrom({seqconv_input, seqconv_weight_var})
      .LinksTo({seqconv_out_var});
  eltadd_op->LinksFrom({seqconv_out_var, eltadd_bias_var})
      .LinksTo({eltadd_out_var});
  relu_op->LinksFrom({eltadd_out_var}).LinksTo({relu_out_var});
  return relu_out_var;
}

C
chengduo 已提交
897
PDNode *patterns::FC::operator()(paddle::framework::ir::PDNode *x,
898
                                 bool with_bias, bool with_relu) {
Y
Yan Chunwei 已提交
899 900
  // Create shared nodes.
  x->assert_is_op_input("mul", "X");
C
chengduo 已提交
901
  auto *mul = pattern->NewNode(mul_repr())->assert_is_op("mul");
Y
Yan Chunwei 已提交
902

C
chengduo 已提交
903
  auto *mul_w_var = pattern->NewNode(w_repr())
Y
Yan Chunwei 已提交
904 905 906 907
                        ->AsInput()
                        ->assert_is_persistable_var()
                        ->assert_is_op_input("mul", "Y");

C
chengduo 已提交
908
  auto *mul_out_var =
Y
Yan Chunwei 已提交
909 910
      pattern->NewNode(mul_out_repr())->assert_is_op_output("mul");

911 912
  // Add links.
  mul->LinksFrom({x, mul_w_var}).LinksTo({mul_out_var});
Y
Yan Chunwei 已提交
913 914 915 916 917
  if (!with_bias) {  // not with bias
    return mul_out_var;
  } else {  // with bias
    mul_out_var->AsIntermediate()->assert_is_op_input("elementwise_add");
    // Create operators.
C
chengduo 已提交
918
    auto *elementwise_add = pattern->NewNode(elementwise_add_repr())
Y
Yan Chunwei 已提交
919 920
                                ->assert_is_op("elementwise_add");
    // Create variables.
C
chengduo 已提交
921
    auto *bias = pattern->NewNode(bias_repr())
Y
Yan Chunwei 已提交
922
                     ->assert_is_op_input("elementwise_add")
923
                     ->assert_is_persistable_var()
Y
Yan Chunwei 已提交
924 925
                     ->AsInput();

926 927 928 929
    auto *elementwise_add_out_var =
        pattern->NewNode(elementwise_add_out_repr())
            ->AsOutput()
            ->assert_is_op_output("elementwise_add");
Y
Yan Chunwei 已提交
930

931 932 933 934 935 936 937 938 939 940 941 942 943 944 945
    elementwise_add->LinksFrom({mul_out_var, bias})
        .LinksTo({elementwise_add_out_var});
    if (!with_relu) {
      return elementwise_add_out_var;
    } else {
      elementwise_add_out_var->AsIntermediate()->assert_is_op_input("relu");
      // Create operators.
      auto *relu = pattern->NewNode(relu_repr())->assert_is_op("relu");
      auto *relu_out_var = pattern->NewNode(relu_out_repr())
                               ->AsOutput()
                               ->assert_is_op_output("relu");

      relu->LinksFrom({elementwise_add_out_var}).LinksTo({relu_out_var});
      return relu_out_var;
    }
946 947
  }
}
T
tensor-tang 已提交
948

949 950 951 952 953 954 955
PDNode *patterns::FCMKLDNN::operator()(paddle::framework::ir::PDNode *x,
                                       bool with_bias) {
  // Create shared nodes.
  x->assert_is_op_input("fc", "Input");

  auto *fc_op = pattern->NewNode(fc_repr())->assert_is_op("fc");
  // Create variables
M
Michał Gallus 已提交
956 957 958 959
  // Input
  auto *input_var = pattern->NewNode(input_repr())
                        ->AsInput()
                        ->assert_is_op_input("fc", "Input");
960 961 962 963 964 965 966 967 968 969 970 971 972 973
  // Filter
  auto *fc_weight_var = pattern->NewNode(weights_repr())
                            ->AsInput()
                            ->assert_is_op_input("fc", "W");
  // Bias
  auto *fc_bias_var = pattern->NewNode(bias_repr())
                          ->AsInput()
                          ->assert_is_op_input("fc", "Bias");
  // Output
  auto *fc_out_var = pattern->NewNode(output_repr())
                         ->AsOutput()
                         ->assert_is_op_output("fc", "Out")
                         ->assert_is_only_output_of_op("fc");

M
Michał Gallus 已提交
974 975
  fc_op->LinksFrom({input_var, fc_weight_var, fc_bias_var})
      .LinksTo({fc_out_var});
976 977 978
  return fc_out_var;
}

979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996
PDNode *patterns::Embedding::operator()(PDNode *x) {
  x->assert_is_op_input("lookup_table", "Ids");
  auto *lookup_table_op =
      pattern->NewNode(lookup_table_repr())->assert_is_op("lookup_table");
#define NEW_NODE(arg__, io__)                    \
  auto *arg__ = pattern->NewNode(arg__##_repr()) \
                    ->assert_is_op_##io__("lookup_table", #arg__);

  NEW_NODE(W, input);

  NEW_NODE(Out, output);
#undef NEW_NODE

  lookup_table_op->LinksFrom({x, W});
  lookup_table_op->LinksTo({Out});
  return Out;
}

C
chengduo 已提交
997
PDNode *patterns::LSTM::operator()(PDNode *x) {
998
  x->assert_is_op_input("lstm", "Input");
C
chengduo 已提交
999
  auto *lstm_op = pattern->NewNode(lstm_repr())->assert_is_op("lstm");
Y
Yan Chunwei 已提交
1000
#define NEW_NODE(arg__, io__) \
C
chengduo 已提交
1001
  auto *arg__ =               \
Y
Yan Chunwei 已提交
1002
      pattern->NewNode(arg__##_repr())->assert_is_op_##io__("lstm", #arg__);
1003 1004 1005 1006 1007

  // Currently, the H0 and C0 are optional
  // TODO(Superjomn) upgrade the fuse framework to support optional.
  // NEW_NODE(H0, input);
  // NEW_NODE(C0, input);
Y
Yan Chunwei 已提交
1008 1009
  NEW_NODE(Weight, input);
  NEW_NODE(Bias, input);
1010

Y
Yan Chunwei 已提交
1011 1012 1013 1014 1015
  NEW_NODE(Hidden, output);
  NEW_NODE(Cell, output);
  NEW_NODE(BatchGate, output);
  NEW_NODE(BatchCellPreAct, output);
#undef NEW_NODE
1016 1017 1018 1019 1020

  lstm_op->LinksFrom({x, Weight, Bias});
  lstm_op->LinksTo({Hidden, Cell, BatchGate, BatchCellPreAct});
  return Hidden;
}
T
tensor-tang 已提交
1021

C
chengduo 已提交
1022
PDNode *patterns::GRU::operator()(PDNode *x) {
T
tensor-tang 已提交
1023
  x->assert_is_op_input("gru", "Input");
C
chengduo 已提交
1024
  auto *gru_op = pattern->NewNode(gru_repr())->assert_is_op("gru");
Y
Yan Chunwei 已提交
1025
#define NEW_NODE(arg__, io__) \
C
chengduo 已提交
1026
  auto *arg__ =               \
Y
Yan Chunwei 已提交
1027
      pattern->NewNode(arg__##_repr())->assert_is_op_##io__("gru", #arg__);
T
tensor-tang 已提交
1028

Y
Yan Chunwei 已提交
1029
  NEW_NODE(Weight, input);
T
tensor-tang 已提交
1030 1031
  // TODO(Superjomn): upgrade the fuse framework to support optional.
  // H0 and bias are optional
Y
Yan Chunwei 已提交
1032
  NEW_NODE(Bias, input);  // also optional
T
tensor-tang 已提交
1033 1034
  // NEW_NODE(H0, input);

Y
Yan Chunwei 已提交
1035
  NEW_NODE(Hidden, output);
T
tensor-tang 已提交
1036
  // below are intermediate
Y
Yan Chunwei 已提交
1037 1038 1039 1040
  NEW_NODE(BatchGate, output);
  NEW_NODE(BatchResetHiddenPrev, output);
  NEW_NODE(BatchHidden, output);
#undef NEW_NODE
T
tensor-tang 已提交
1041

T
tensor-tang 已提交
1042 1043 1044 1045
  BatchGate->AsIntermediate();
  BatchResetHiddenPrev->AsIntermediate();
  BatchHidden->AsIntermediate();

T
tensor-tang 已提交
1046 1047 1048 1049 1050
  gru_op->LinksFrom({x, Weight, Bias});
  gru_op->LinksTo({Hidden, BatchGate, BatchResetHiddenPrev, BatchHidden});
  return Hidden;
}

C
chengduo 已提交
1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079
PDNode *patterns::ActElewiseAdd::operator()(
    paddle::framework::ir::PDNode *in_var,
    std::unordered_set<std::string> act_types) {
  in_var->assert_is_ops_input(act_types, "X");

  auto *act = pattern->NewNode(act_repr())->assert_is_ops(act_types);
  auto *act_out_var = pattern->NewNode(act_out_repr())
                          ->assert_is_not_ctrl_var()
                          ->assert_is_ops_output(act_types);
  act_out_var->AsIntermediate()->assert_is_op_input("elementwise_add");

  auto *ele_x_var = pattern->NewNode(ele_x_repr())
                        ->assert_is_not_ctrl_var()
                        ->assert_is_op_input("elementwise_add")
                        ->AsInput();
  auto *elementwise_add =
      pattern->NewNode(ele_add_repr())->assert_is_op("elementwise_add");

  auto *elewise_add_out = pattern->NewNode(elewise_add_out_repr())
                              ->AsOutput()
                              ->assert_is_op_output("elementwise_add", "Out");

  act->LinksFrom({in_var}).LinksTo({act_out_var});
  elementwise_add->LinksFrom({act_out_var, ele_x_var})
      .LinksTo({elewise_add_out});

  return elewise_add_out;
}

Z
Zhen Wang 已提交
1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190
PDNode *patterns::BatchNormAct::operator()(
    paddle::framework::ir::PDNode *bn_x_var,
    std::unordered_set<std::string> act_types) {
  auto *bn_scale_var = pattern->NewNode(bn_scale_repr())
                           ->assert_is_op_input("batch_norm", "Scale");
  auto *bn_bias_var = pattern->NewNode(bn_bias_repr())
                          ->assert_is_op_input("batch_norm", "Bias");
  auto *bn_variance_var = pattern->NewNode(bn_variance_repr())
                              ->assert_is_op_input("batch_norm", "Variance");
  auto *bn_mean_var = pattern->NewNode(bn_mean_repr())
                          ->assert_is_op_input("batch_norm", "Mean");

  auto *bn = pattern->NewNode(batch_norm_repr())
                 ->assert_is_op("batch_norm")
                 ->assert_is_not_op_input("MomentumTensor")
                 ->assert_op_attr<bool>("is_test", false)
                 ->assert_op_attr<bool>("use_global_stats", false)
                 ->assert_op_attr<std::string>("data_layout", "NHWC");

  auto *bn_mean_out_var = pattern->NewNode(bn_mean_out_repr())
                              ->assert_is_op_output("batch_norm", "MeanOut");
  auto *bn_variance_out_var =
      pattern->NewNode(bn_variance_out_repr())
          ->assert_is_op_output("batch_norm", "VarianceOut");
  auto *bn_saved_variance_var =
      pattern->NewNode(bn_saved_variance_repr())
          ->assert_is_op_output("batch_norm", "SavedVariance");
  auto *bn_saved_mean_var =
      pattern->NewNode(bn_saved_mean_repr())
          ->assert_is_op_output("batch_norm", "SavedMean");
  auto *bn_reserve_space =
      pattern->NewNode(bn_reserve_space_repr())
          ->assert_is_op_output("batch_norm", "ReserveSpace");
  auto *bn_out_var = pattern->NewNode(bn_out_repr())
                         ->assert_is_op_output("batch_norm", "Y")
                         ->assert_has_n_outputs(1);

  bn_out_var->AsIntermediate()->assert_is_ops_input(act_types);

  auto *act = pattern->NewNode(act_repr())->assert_is_ops(act_types);

  auto *act_out_var =
      pattern->NewNode(act_out_repr())->assert_is_ops_output(act_types, "Out");

  bn->LinksFrom(
        {bn_x_var, bn_scale_var, bn_bias_var, bn_variance_var, bn_mean_var})
      .LinksTo({bn_mean_out_var, bn_variance_out_var, bn_saved_variance_var,
                bn_saved_mean_var, bn_reserve_space, bn_out_var});
  act->LinksFrom({bn_out_var}).LinksTo({act_out_var});

  return act_out_var;
}

PDNode *patterns::BatchNormActGrad::operator()(
    paddle::framework::ir::PDNode *d_act_out_var,
    std::unordered_set<std::string> act_grad_types) {
  auto *act_grad =
      pattern->NewNode(act_grad_repr())->assert_is_ops(act_grad_types);
  auto *bn_grad = pattern->NewNode(batch_norm_grad_repr())
                      ->assert_is_op("batch_norm_grad")
                      ->assert_op_attr<bool>("use_global_stats", false)
                      ->assert_op_attr<std::string>("data_layout", "NHWC");

  auto *act_out_var = pattern->NewNode(act_out_repr())
                          ->assert_is_ops_input(act_grad_types, "Out");
  auto *d_intermediate_var =
      pattern->NewNode(d_itermediate_out_repr())
          ->assert_is_ops_output(act_grad_types, GradVarName("X"))
          ->assert_has_n_outputs(1);
  auto *bn_x_var = pattern->NewNode(bn_x_repr())
                       ->assert_is_op_input("batch_norm_grad", "X")
                       ->assert_var_dtype(proto::VarType::FP16);
  auto *bn_scale_var = pattern->NewNode(bn_scale_repr())
                           ->assert_is_op_input("batch_norm_grad", "Scale");
  auto *bn_bias_var = pattern->NewNode(bn_bias_repr())
                          ->assert_is_op_input("batch_norm_grad", "Bias");
  auto *bn_saved_mean_var =
      pattern->NewNode(bn_saved_mean_repr())
          ->assert_is_op_input("batch_norm_grad", "SavedMean");
  auto *bn_saved_variance_var =
      pattern->NewNode(bn_saved_variance_repr())
          ->assert_is_op_input("batch_norm_grad", "SavedVariance");
  // ReserveSpace as the output is equal to:
  // data_layout == 'NHWC' && FLAGS_cudnn_batchnorm_spatial_persistent == true
  auto *bn_reserve_space =
      pattern->NewNode(bn_reserve_space_repr())
          ->assert_is_op_input("batch_norm_grad", "ReserveSpace");
  auto *d_bn_x_var =
      pattern->NewNode(d_bn_x_repr())
          ->assert_is_not_ctrl_var()
          ->assert_is_op_output("batch_norm_grad", GradVarName("X"));
  auto *d_bn_scale_var =
      pattern->NewNode(d_bn_scale_repr())
          ->assert_is_not_ctrl_var()
          ->assert_is_op_output("batch_norm_grad", GradVarName("Scale"));
  auto *d_bn_bias_var =
      pattern->NewNode(d_bn_bias_repr())
          ->assert_is_not_ctrl_var()
          ->assert_is_op_output("batch_norm_grad", GradVarName("Bias"));

  act_grad->LinksFrom({d_act_out_var, act_out_var})
      .LinksTo({d_intermediate_var});

  bn_grad
      ->LinksFrom({bn_x_var, d_intermediate_var, bn_scale_var, bn_bias_var,
                   bn_saved_mean_var, bn_saved_variance_var, bn_reserve_space})
      .LinksTo({d_bn_x_var, d_bn_scale_var, d_bn_bias_var});

  return bn_grad;
}

C
chengduo 已提交
1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255
PDNode *patterns::ElewiseAddAct::operator()(
    paddle::framework::ir::PDNode *ele_x_var,
    std::unordered_set<std::string> act_types) {
  auto *ele_y_var = pattern->NewNode(ele_y_repr())
                        ->assert_is_op_input("elementwise_add", "Y");

  auto *ele_add =
      pattern->NewNode(ele_add_repr())->assert_is_op("elementwise_add");

  auto *ele_out_var = pattern->NewNode(elewise_add_out_repr())
                          ->assert_is_op_output("elementwise_add", "Out");

  ele_out_var->AsIntermediate()->assert_is_ops_input(act_types);

  auto *act = pattern->NewNode(act_repr())->assert_is_ops(act_types);

  auto *act_out_var =
      pattern->NewNode(act_out_repr())->assert_is_ops_output(act_types, "Out");

  ele_add->LinksFrom({ele_x_var, ele_y_var}).LinksTo({ele_out_var});
  act->LinksFrom({ele_out_var}).LinksTo({act_out_var});

  return act_out_var;
}

PDNode *patterns::ElewiseAddActInplaceGrad::operator()(
    paddle::framework::ir::PDNode *d_act_out_var,
    std::unordered_set<std::string> act_types) {
  // act_grad: in["Out", "Out@GRAD"], out["X@GRAD"]
  // ele_add_grad: in["Y", "Out@GRAD"], out["X@GRAD", "Y@GRAD"]
  auto *act_grad = pattern->NewNode(act_grad_repr())->assert_is_ops(act_types);

  auto *act_out_var =
      pattern->NewNode(act_out_repr())->assert_is_ops_input(act_types, "Out");

  auto *d_intermediate_var =
      pattern->NewNode(d_itermediate_out_repr())
          ->assert_is_ops_output(act_types, GradVarName("X"));

  act_grad->LinksFrom({d_act_out_var, act_out_var})
      .LinksTo({d_intermediate_var});

  auto *ele_y_var = pattern->NewNode(ele_y_repr())
                        ->assert_is_not_ctrl_var()
                        ->assert_is_op_input("elementwise_add_grad", "Y");

  auto *ele_add_grad = pattern->NewNode(ele_add_grad_repr())
                           ->assert_is_op("elementwise_add_grad");

  auto *d_ele_x_var =
      pattern->NewNode(d_ele_x_repr())
          ->assert_is_not_ctrl_var()
          ->assert_is_op_output("elementwise_add_grad", GradVarName("X"));

  auto *d_ele_y_var =
      pattern->NewNode(d_ele_y_repr())
          ->assert_is_not_ctrl_var()
          ->assert_is_op_output("elementwise_add_grad", GradVarName("Y"));

  ele_add_grad->LinksFrom({d_intermediate_var, ele_y_var})
      .LinksTo({d_ele_x_var, d_ele_y_var});

  return ele_add_grad;
}

1256
// conv_type: conv2d, conv3d, conv2d_transpose
M
Michal Gallus 已提交
1257
PDNode *patterns::ConvBias::operator()(
1258
    paddle::framework::ir::PDNode *conv_input, std::string conv_type) {
M
Michal Gallus 已提交
1259
  // Create Operators
1260 1261
  conv_input->assert_is_op_input(conv_type, "Input");
  auto *conv_op = pattern->NewNode(conv_repr())->assert_is_op(conv_type);
M
Michal Gallus 已提交
1262 1263 1264 1265
  auto *eltiwse_op =
      pattern->NewNode(eltwise_repr())->assert_is_op("elementwise_add");
  // Create variables
  // Filter
Y
Yihua Xu 已提交
1266 1267 1268
  auto *conv_weight_var = pattern->NewNode(conv_weight_repr())
                              ->AsInput()
                              ->assert_is_persistable_var()
1269
                              ->assert_is_op_input(conv_type, "Filter");
M
Michal Gallus 已提交
1270
  // intermediate variable, will be removed in the IR after fuse.
Y
Yihua Xu 已提交
1271 1272
  auto *conv_out_var = pattern->NewNode(conv_out_repr())
                           ->AsIntermediate()
1273
                           ->assert_is_only_output_of_op(conv_type)
Y
Yihua Xu 已提交
1274
                           ->assert_is_op_input("elementwise_add");
M
Michal Gallus 已提交
1275 1276 1277
  // Bias stored in elementwise_add
  auto *eltwise_bias_var = pattern->NewNode(eltwise_bias_repr())
                               ->AsInput()
M
Michal Gallus 已提交
1278
                               ->assert_is_persistable_var()
M
Michal Gallus 已提交
1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289
                               ->assert_is_op_input("elementwise_add", "Y");
  // output
  auto *eltwise_out_var = pattern->NewNode(eltwise_out_repr())
                              ->AsOutput()
                              ->assert_is_op_output("elementwise_add");
  conv_op->LinksFrom({conv_input, conv_weight_var}).LinksTo({conv_out_var});
  eltiwse_op->LinksFrom({conv_out_var, eltwise_bias_var})
      .LinksTo({eltwise_out_var});
  return eltwise_out_var;
}

1290 1291 1292 1293
PDNode *patterns::Conv::operator()() {
  auto conv_op = pattern->NewNode(conv_op_repr())->assert_is_op("conv2d");

  auto input_var = pattern->NewNode(conv_input_repr())
1294
                       ->AsInput()
1295 1296 1297
                       ->assert_is_op_input("conv2d", "Input");

  auto filter_var = pattern->NewNode(conv_filter_repr())
1298
                        ->AsInput()
1299 1300 1301
                        ->assert_is_op_input("conv2d", "Filter");

  auto output_var = pattern->NewNode(conv_output_repr())
1302
                        ->AsOutput()
1303 1304
                        ->assert_is_op_output("conv2d", "Output");

1305 1306 1307 1308
  conv_op->LinksFrom({input_var, filter_var}).LinksTo({output_var});
  return output_var;
}

1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329
PDNode *patterns::Transpose::operator()() {
  auto prev_op = pattern->NewNode(prev_op_repr())->assert_is_op();

  auto transpose_op =
      pattern->NewNode(transpose_op_repr())->assert_is_op("transpose2");

  auto transpose_in = pattern->NewNode(transpose_in_repr())
                          ->AsInput()
                          ->assert_is_op_input("transpose2");
  auto transpose_out = pattern->NewNode(transpose_out_repr())
                           ->AsOutput()
                           ->assert_is_op_output("transpose2", "Out");

  auto next_op = pattern->NewNode(next_op_repr())->assert_is_op();

  prev_op->LinksTo({transpose_in});
  transpose_op->LinksFrom({transpose_in}).LinksTo({transpose_out});
  next_op->LinksFrom({transpose_out});
  return transpose_out;
}

1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350
PDNode *patterns::Reshape::operator()() {
  auto prev_op = pattern->NewNode(prev_op_repr())->assert_is_op();

  auto reshape_op =
      pattern->NewNode(reshape_op_repr())->assert_is_op("reshape2");

  auto reshape_in = pattern->NewNode(reshape_in_repr())
                        ->AsInput()
                        ->assert_is_op_input("reshape2", "X");
  auto reshape_out = pattern->NewNode(reshape_out_repr())
                         ->AsOutput()
                         ->assert_is_op_output("reshape2", "Out");

  auto next_op = pattern->NewNode(next_op_repr())->assert_is_op();

  prev_op->LinksTo({reshape_in});
  reshape_op->LinksFrom({reshape_in}).LinksTo({reshape_out});
  next_op->LinksFrom({reshape_out});
  return reshape_out;
}

1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371
PDNode *patterns::Matmul::operator()() {
  auto prev_op_x = pattern->NewNode(prev_op_x_repr())->assert_is_op();
  auto prev_op_y = pattern->NewNode(prev_op_y_repr())->assert_is_op();

  auto matmul_op = pattern->NewNode(matmul_op_repr())->assert_is_op("matmul");
  auto matmul_in_x = pattern->NewNode(matmul_in_x_repr())
                         ->AsInput()
                         ->assert_is_op_input("matmul", "X");
  auto matmul_in_y = pattern->NewNode(matmul_in_y_repr())
                         ->AsInput()
                         ->assert_is_op_input("matmul", "Y");
  auto matmul_out = pattern->NewNode(matmul_out_repr())
                        ->AsOutput()
                        ->assert_is_op_output("matmul", "Out");

  prev_op_x->LinksTo({matmul_in_x});
  prev_op_y->LinksTo({matmul_in_y});
  matmul_op->LinksFrom({matmul_in_x, matmul_in_y}).LinksTo({matmul_out});
  return matmul_out;
}

1372 1373 1374
PDNode *patterns::ConvResidual::operator()(bool with_residual_data) {
  auto conv_op = pattern->NewNode(conv_op_repr())->assert_is_op("conv2d");

1375 1376 1377 1378 1379 1380 1381 1382 1383
  if (!with_residual_data) {
    conv_op->assert_more([&](Node *x) {
      auto node_names = x->Op()->InputNames();
      if (!HasInput(x, "ResidualData") ||
          x->Op()->Input("ResidualData").size() == 0)
        return true;
      return false;
    });
  }
1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419

  auto input_var = pattern->NewNode(conv_input_repr())
                       ->AsInput()
                       ->assert_is_op_input("conv2d", "Input");

  auto filter_var = pattern->NewNode(conv_filter_repr())
                        ->AsInput()
                        ->assert_is_op_input("conv2d", "Filter");

  auto output_var = pattern->NewNode(conv_output_repr())
                        ->AsOutput()
                        ->assert_is_op_output("conv2d", "Output");

  std::vector<PDNode *> links_from{input_var, filter_var};

  if (with_residual_data) {
    auto res_conn_var = pattern->NewNode(conv_residual_data_repr())
                            ->AsInput()
                            ->assert_is_op_input("conv2d", "ResidualData");
    links_from.push_back(res_conn_var);
  }

  conv_op->LinksFrom(links_from).LinksTo({output_var});
  return output_var;
}

PDNode *patterns::Pool::operator()() {
  auto pool_op = pattern->NewNode(pool_op_repr())->assert_is_op("pool2d");

  auto input_var = pattern->NewNode(pool_input_repr())
                       ->AsInput()
                       ->assert_is_op_input("pool2d", "X");

  auto output_var = pattern->NewNode(pool_output_repr())
                        ->AsOutput()
                        ->assert_is_op_output("pool2d", "Out");
1420

1421
  pool_op->LinksFrom({input_var}).LinksTo({output_var});
1422 1423 1424
  return output_var;
}

1425
PDNode *patterns::ElementwiseAdd::operator()(PDNode *x_var, PDNode *y_var) {
1426 1427 1428
  auto elementwise_add_op = pattern->NewNode(elementwise_add_op_repr())
                                ->assert_is_op("elementwise_add");

1429 1430
  x_var->AsInput()->assert_is_op_input("elementwise_add", "X");
  y_var->AsInput()->assert_is_op_input("elementwise_add", "Y");
1431 1432 1433 1434
  auto out_var = pattern->NewNode(elementwise_add_out_repr())
                     ->AsOutput()
                     ->assert_is_op_output("elementwise_add", "Out");

1435
  elementwise_add_op->LinksFrom({x_var, y_var});
1436 1437 1438 1439
  elementwise_add_op->LinksTo({out_var});

  return out_var;
}
1440

1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451
PDNode *patterns::Concat::operator()() {
  auto concat_op = pattern->NewNode(concat_op_repr())->assert_is_op("concat");

  auto output_var = pattern->NewNode(concat_out_repr())
                        ->AsOutput()
                        ->assert_is_op_output("concat", "Out");

  concat_op->LinksTo({output_var});
  return output_var;
}

1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491
PDNode *patterns::ConcatReLU::operator()() {
  auto concat_op = pattern->NewNode(concat_op_repr())->assert_is_op("concat");
  auto relu_op = pattern->NewNode(relu_op_repr())->assert_is_op("relu");

  auto concat_out =
      pattern->NewNode(concat_out_repr())->assert_is_op_output("concat", "Out");

  auto relu_out = pattern->NewNode(relu_out_repr())
                      ->AsOutput()
                      ->assert_is_op_output("relu", "Out");

  concat_op->LinksTo({concat_out});
  relu_op->LinksFrom({concat_out}).LinksTo({relu_out});

  return relu_out;
}

PDNode *patterns::ConvConcatReLU::operator()() {
  auto conv_op = pattern->NewNode(conv_op_repr())->assert_is_op("conv2d");
  auto concat_op = pattern->NewNode(concat_op_repr())->assert_is_op("concat");
  auto relu_op = pattern->NewNode(relu_op_repr())->assert_is_op("relu");

  auto conv_out = pattern->NewNode(conv_out_repr())
                      ->assert_is_op_output("conv2d", "Output");

  auto concat_out = pattern->NewNode(concat_out_repr())
                        ->assert_is_op_output("concat", "Out")
                        ->assert_is_op_input("relu", "X");

  auto relu_out = pattern->NewNode(relu_out_repr())
                      ->AsOutput()
                      ->assert_is_op_output("relu", "Out");

  conv_op->LinksTo({conv_out});
  concat_op->LinksFrom({conv_out}).LinksTo({concat_out});
  relu_op->LinksFrom({concat_out}).LinksTo({relu_out});

  return relu_out;
}

J
joanna.wozna.intel 已提交
1492 1493 1494 1495 1496 1497 1498 1499
PDNode *patterns::OpRequant::operator()() {
  auto any_op = pattern->NewNode(any_op_repr())
                    ->assert_is_op()
                    ->assert_more([&](Node *node) {
                      return node->Op()->HasAttr("Scale_out") ? true : false;
                    });
  auto requant_in = pattern->NewNode(requant_in_repr())
                        ->assert_is_op_input("requantize", "Input");
1500 1501 1502 1503 1504 1505
  auto requant_op =
      pattern->NewNode(requant_op_repr())->assert_is_op("requantize");
  auto requant_out = pattern->NewNode(requant_out_repr())
                         ->AsOutput()
                         ->assert_is_op_output("requantize", "Output");

J
joanna.wozna.intel 已提交
1506 1507
  any_op->LinksTo({requant_in});
  requant_op->LinksFrom({requant_in}).LinksTo({requant_out});
1508 1509 1510
  return requant_out;
}

1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531
PDNode *patterns::RequantOp::operator()() {
  auto requant_in = pattern->NewNode(requant_in_repr())
                        ->assert_is_op_input("requantize", "Input");
  auto requant_op =
      pattern->NewNode(requant_op_repr())->assert_is_op("requantize");
  auto requant_out = pattern->NewNode(requant_out_repr())
                         ->AsOutput()
                         ->assert_is_op_output("requantize", "Output");
  auto any_op = pattern->NewNode(any_op_repr())
                    ->assert_is_op()
                    ->assert_more([&](Node *node) {
                      return (node->Op()->HasAttr("Scale_in") ||
                              node->Op()->HasAttr("Scale_x") ||
                              node->Op()->HasAttr("Scale_y"));
                    });

  requant_op->LinksFrom({requant_in}).LinksTo({requant_out});
  any_op->LinksFrom({requant_out});
  return any_op;
}

1532 1533 1534 1535 1536 1537 1538 1539 1540 1541
PDNode *patterns::OpDequant::operator()() {
  auto any_op = pattern->NewNode(any_op_repr())
                    ->assert_is_op()
                    ->assert_more([&](Node *node) {
                      return (node->Op()->Type() == "matmul" ||
                              node->Op()->Type() == "conv2d" ||
                              node->Op()->Type() == "fc");
                    });
  auto dequant_in = pattern->NewNode(dequant_in_repr())
                        ->assert_is_op_input("dequantize", "Input");
1542 1543 1544 1545 1546 1547
  auto dequant_op =
      pattern->NewNode(dequant_op_repr())->assert_is_op("dequantize");
  auto dequant_out = pattern->NewNode(dequant_out_repr())
                         ->AsOutput()
                         ->assert_is_op_output("dequantize", "Output");

1548 1549
  any_op->LinksTo({dequant_in});
  dequant_op->LinksFrom({dequant_in}).LinksTo({dequant_out});
1550 1551 1552
  return dequant_out;
}

1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571
PDNode *patterns::DequantScale::operator()() {
  // Create Operators
  auto dequant_op =
      pattern->NewNode(dequant_op_repr())->assert_is_op("dequantize");
  auto scale_op = pattern->NewNode(scale_op_repr())->assert_is_op("scale");

  auto dequant_out = pattern->NewNode(dequant_out_repr())
                         ->AsOutput()
                         ->assert_is_op_output("dequantize", "Output");
  auto scale_out = pattern->NewNode(scale_out_repr())
                       ->AsOutput()
                       ->assert_is_op_output("scale", "Out");

  dequant_op->LinksTo({dequant_out});
  scale_op->LinksFrom({dequant_out}).LinksTo({scale_out});

  return scale_out;
}

1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586
PDNode *patterns::ScaleMatmul::operator()() {
  auto scale_in = pattern->NewNode(scale_in_repr())
                      ->AsInput()
                      ->assert_is_op_input("scale", "X");
  auto scale_op = pattern->NewNode(scale_op_repr())->assert_is_op("scale");
  auto scale_out = pattern->NewNode(scale_out_repr())
                       ->AsOutput()
                       ->assert_is_op_output("scale", "Out");
  auto matmul_op = pattern->NewNode(matmul_op_repr())->assert_is_op("matmul");

  scale_op->LinksFrom({scale_in}).LinksTo({scale_out});
  matmul_op->LinksFrom({scale_out});
  return matmul_op;
}

1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611
PDNode *patterns::PriorBox::operator()() {
  auto prior_box_op =
      pattern->NewNode(prior_box_op_repr())->assert_is_op("prior_box");

  auto input_var = pattern->NewNode(prior_box_input_repr())
                       ->AsInput()
                       ->assert_is_op_input("prior_box", "Input");

  auto image_var = pattern->NewNode(prior_box_image_repr())
                       ->AsInput()
                       ->assert_is_op_input("prior_box", "Image");

  auto boxes_var = pattern->NewNode(prior_box_boxes_repr())
                       ->AsOutput()
                       ->assert_is_op_output("prior_box", "Boxes");

  auto variances_var = pattern->NewNode(prior_box_variances_repr())
                           ->AsOutput()
                           ->assert_is_op_output("prior_box", "Variances");

  prior_box_op->LinksFrom({input_var, image_var})
      .LinksTo({boxes_var, variances_var});
  return boxes_var;
}

H
hjchen2 已提交
1612
std::unordered_set<std::string> conv_act_set({"identity", "relu"});
1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626

PDNode *patterns::ConvElementwiseaddAct::operator()(PDNode *conv_in) {
  conv_in->AsInput();
  auto conv_op = pattern->NewNode(conv_op_repr())->assert_is_op("conv2d");
  auto conv_out = pattern->NewNode(conv_out_repr())
                      ->assert_is_op_output("conv2d")
                      ->assert_is_op_input("elementwise_add", "X")
                      ->AsIntermediate();
  auto conv_filter = pattern->NewNode(conv_filter_repr())
                         ->assert_is_op_input("conv2d", "Filter")
                         ->AsInput();
  auto elementwise_add_op = pattern->NewNode(elementwise_add_op_repr())
                                ->assert_is_op("elementwise_add");
  auto elementwise_add_in_y = pattern->NewNode(elementwise_add_in_y_repr())
1627
                                  ->assert_is_persistable_var()
1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674
                                  ->assert_is_op_input("elementwise_add", "Y")
                                  ->AsInput();
  auto elementwise_add_out = pattern->NewNode(elementwise_add_out_repr())
                                 ->assert_is_op_output("elementwise_add")
                                 ->AsIntermediate();

  auto act_op = pattern->NewNode(act_op_repr())
                    ->assert_is_op()
                    ->assert_more([&](Node *node) {
                      auto op_type = node->Name();
                      return conv_act_set.count(op_type);
                    });

  auto act_out = pattern->NewNode(act_out_repr())
                     ->assert_is_var()
                     // is activation op's output.
                     ->assert_more([&](Node *node) {
                       for (auto *in_op : node->inputs) {
                         if (conv_act_set.count(in_op->Name())) {
                           return true;
                         }
                       }
                       return false;
                     })
                     ->AsOutput();

  conv_op->LinksFrom({conv_in, conv_filter});
  conv_out->LinksFrom({conv_op});
  elementwise_add_op->LinksFrom({conv_out, elementwise_add_in_y})
      .LinksTo({elementwise_add_out});
  act_op->LinksFrom({elementwise_add_out}).LinksTo({act_out});

  return act_out;
}

PDNode *patterns::ConvElementwiseadd2Act::operator()(PDNode *conv_in) {
  auto conv_op = pattern->NewNode(conv_op_repr())->assert_is_op("conv2d");
  auto conv_filter = pattern->NewNode(conv_filter_repr())
                         ->assert_is_op_input("conv2d", "Filter")
                         ->AsInput();
  auto conv_out = pattern->NewNode(conv_out_repr())
                      ->assert_is_op_output("conv2d")
                      ->assert_is_op_input("elementwise_add", "X")
                      ->AsIntermediate();
  auto elementwise_add_op = pattern->NewNode(elementwise_add_op_repr())
                                ->assert_is_op("elementwise_add");
  auto elementwise_add_in_y = pattern->NewNode(elementwise_add_in_y_repr())
1675
                                  ->assert_is_persistable_var()
1676 1677 1678 1679
                                  ->assert_is_op_input("elementwise_add", "Y")
                                  ->AsInput();
  auto elementwise_add_out = pattern->NewNode(elementwise_add_out_repr())
                                 ->assert_is_op_output("elementwise_add")
H
hjchen2 已提交
1680
                                 ->assert_is_op_input("elementwise_add", "Y")
1681 1682 1683 1684 1685
                                 ->AsIntermediate();

  auto elementwise_add_op_1 = pattern->NewNode(elementwise_add_op_1_repr())
                                  ->assert_is_op("elementwise_add");
  auto elementwise_add_in_y_1 = pattern->NewNode(elementwise_add_in_y_1_repr())
H
hjchen2 已提交
1686
                                    ->assert_is_op_input("elementwise_add", "X")
1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713
                                    ->AsInput();
  auto elementwise_add_out_1 = pattern->NewNode(elementwise_add_out_1_repr())
                                   ->assert_is_op_output("elementwise_add")
                                   ->AsIntermediate();

  auto act_op = pattern->NewNode(act_op_repr())
                    ->assert_is_op()
                    ->assert_more([&](Node *node) {
                      auto op_type = node->Name();
                      return conv_act_set.count(op_type);
                    });
  auto act_out = pattern->NewNode(act_out_repr())
                     ->assert_is_var()
                     // is activation op's output.
                     ->assert_more([&](Node *node) {
                       for (auto *in_op : node->inputs) {
                         if (conv_act_set.count(in_op->Name())) {
                           return true;
                         }
                       }
                       return false;
                     })
                     ->AsOutput();

  conv_op->LinksFrom({conv_in, conv_filter}).LinksTo({conv_out});
  elementwise_add_op->LinksFrom({conv_out, elementwise_add_in_y})
      .LinksTo({elementwise_add_out});
H
hjchen2 已提交
1714 1715
  elementwise_add_op_1->LinksFrom({elementwise_add_out, elementwise_add_in_y_1})
      .LinksTo({elementwise_add_out_1});
1716 1717 1718 1719
  act_op->LinksFrom({elementwise_add_out_1}).LinksTo({act_out});
  return act_out;
}

N
nhzlx 已提交
1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732
PDNode *patterns::ConvElementwiseadd::operator()(PDNode *conv_in) {
  conv_in->AsInput();
  auto conv_op = pattern->NewNode(conv_op_repr())->assert_is_op("conv2d");
  auto conv_out = pattern->NewNode(conv_out_repr())
                      ->assert_is_op_output("conv2d")
                      ->assert_is_op_input("elementwise_add", "X")
                      ->AsIntermediate();
  auto conv_filter = pattern->NewNode(conv_filter_repr())
                         ->assert_is_op_input("conv2d", "Filter")
                         ->AsInput();
  auto elementwise_add_op = pattern->NewNode(elementwise_add_op_repr())
                                ->assert_is_op("elementwise_add");
  auto elementwise_add_in_y = pattern->NewNode(elementwise_add_in_y_repr())
1733
                                  ->assert_is_persistable_var()
N
nhzlx 已提交
1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747
                                  ->assert_is_op_input("elementwise_add", "Y")
                                  ->AsInput();
  auto elementwise_add_out = pattern->NewNode(elementwise_add_out_repr())
                                 ->assert_is_op_output("elementwise_add")
                                 ->AsOutput();

  conv_op->LinksFrom({conv_in, conv_filter});
  conv_out->LinksFrom({conv_op});
  elementwise_add_op->LinksFrom({conv_out, elementwise_add_in_y})
      .LinksTo({elementwise_add_out});

  return elementwise_add_out;
}

N
nhzlx 已提交
1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793
PDNode *patterns::ConvAffineChannel::operator()(
    paddle::framework::ir::PDNode *conv_input, bool with_eltwise_add) {
  // Create Operators
  conv_input->assert_is_op_input("conv2d", "Input");
  auto *conv_op = pattern->NewNode(conv_repr())->assert_is_op("conv2d");

  PDNode *eltwise_op = nullptr;
  if (with_eltwise_add) {
    eltwise_op =
        pattern->NewNode(eltwise_repr())->assert_is_op("elementwise_add");
  }

  auto *affine_channel_op =
      pattern->NewNode(affine_channel_repr())->assert_is_op("affine_channel");
  // Create variables
  // Conv Filter
  auto *conv_weight_var = pattern->NewNode(conv_weight_repr())
                              ->AsInput()
                              ->assert_is_persistable_var()
                              ->assert_is_op_input("conv2d", "Filter");

  auto *conv_out_var = pattern->NewNode(conv_out_repr())
                           ->AsIntermediate()
                           ->assert_is_only_output_of_op("conv2d");

  PDNode *eltwise_y_in_var = nullptr;
  PDNode *eltwise_out_var = nullptr;
  if (with_eltwise_add) {
    // Conv output as Bias input
    conv_out_var->assert_is_op_input("elementwise_add", "X");
    // Bias
    eltwise_y_in_var = pattern->NewNode(eltwise_y_in_repr())
                           ->assert_is_op_input("elementwise_add", "Y")
                           ->AsInput();
    eltwise_out_var = pattern->NewNode(eltwise_out_repr())
                          ->AsIntermediate()
                          ->assert_is_only_output_of_op("elementwise_add");
  } else {
    // Conv output as AffineChannel input
    conv_out_var->assert_is_op_input("affine_channel", "X");
  }

  // AC Scale
  auto *ac_scale_var = pattern->NewNode(ac_scale_repr())
                           ->AsInput()
                           ->assert_is_persistable_var()
1794
                           ->assert_has_n_outputs(1)
N
nhzlx 已提交
1795 1796 1797 1798 1799
                           ->assert_is_op_input("affine_channel", "Scale");
  // AC Bias
  auto *ac_bias_var = pattern->NewNode(ac_bias_repr())
                          ->AsInput()
                          ->assert_is_persistable_var()
1800
                          ->assert_has_n_outputs(1)
N
nhzlx 已提交
1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821
                          ->assert_is_op_input("affine_channel", "Bias");

  // AC output
  auto *ac_out_var = pattern->NewNode(ac_out_repr())
                         ->AsOutput()
                         ->assert_is_op_output("affine_channel");

  conv_op->LinksFrom({conv_input, conv_weight_var}).LinksTo({conv_out_var});

  if (with_eltwise_add) {
    eltwise_op->LinksFrom({conv_out_var, eltwise_y_in_var})
        .LinksTo({eltwise_out_var});
    affine_channel_op->LinksFrom({eltwise_out_var, ac_scale_var, ac_bias_var})
        .LinksTo({ac_out_var});
  } else {
    affine_channel_op->LinksFrom({conv_out_var, ac_scale_var, ac_bias_var})
        .LinksTo({ac_out_var});
  }
  return ac_out_var;
}

1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866
PDNode *patterns::DequantQuantAny::operator()() {
  auto *dequant_in = pattern->NewNode(dequant_in_repr())
                         ->AsInput()
                         ->assert_is_op_input("dequantize", "Input");

  auto *dequant_op =
      pattern->NewNode(dequant_op_repr())->assert_is_op("dequantize");

  auto *dequant_out = pattern->NewNode(dequant_out_repr())
                          ->AsOutput()
                          ->assert_is_op_output("dequantize", "Output");

  auto *quant_op = pattern->NewNode(quant_op_repr())
                       ->assert_is_op("quantize")
                       ->AsIntermediate();

  auto *quant_out = pattern->NewNode(quant_out_repr())
                        ->AsOutput()
                        ->assert_is_op_output("quantize");

  auto *next_op = pattern->NewNode(next_op_repr())->assert_is_op();

  dequant_op->LinksFrom({dequant_in}).LinksTo({dequant_out});
  quant_op->LinksFrom({dequant_out}).LinksTo({quant_out});
  next_op->LinksFrom({quant_out});

  return quant_out;
}

PDNode *patterns::DequantAny::operator()() {
  auto *dequant_op =
      pattern->NewNode(dequant_op_repr())->assert_is_op("dequantize");

  auto *dequant_out = pattern->NewNode(dequant_out_repr())
                          ->AsOutput()
                          ->assert_is_op_output("dequantize", "Output");

  auto *next_op = pattern->NewNode(next_op_repr())->assert_is_op();

  dequant_op->LinksTo({dequant_out});
  next_op->LinksFrom({dequant_out});

  return dequant_out;
}

1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881
PDNode *patterns::MultipleQuantize::operator()() {
  auto *prev_out = pattern->NewNode(prev_out_repr())->AsOutput();

  // find nodes that are inputs to quantize operators
  prev_out->assert_more([&](Node *node) {
    int counter = std::count_if(
        node->outputs.begin(), node->outputs.end(), [&](Node const *iter) {
          return iter && iter->IsOp() && iter->Op()->Type() == "quantize";
        });
    return (counter > 1);
  });

  return prev_out;
}

1882
PDNode *patterns::MKLDNNInPlace::operator()() {
1883
  const std::unordered_set<std::string> &supported_op_types = {
1884 1885 1886 1887 1888 1889 1890 1891 1892 1893
      "abs",
      "elementwise_mul",
      "elementwise_add",
      "gelu",
      "leaky_relu",
      "relu",
      "softmax",
      "sqrt",
      "swish",
      "tanh"};
1894 1895 1896

  auto possible_inplace_op = pattern->NewNode(inplace_to_be_op_repr())
                                 ->assert_is_ops(supported_op_types);
1897 1898

  auto input = pattern->NewNode(inplace_to_be_op_in_repr())
1899
                   ->assert_is_ops_input(supported_op_types)
1900 1901
                   ->AsInput();
  auto output = pattern->NewNode(inplace_to_be_op_out_repr())
1902
                    ->assert_is_ops_output(supported_op_types)
1903
                    ->AsOutput();
1904 1905

  auto next_op = pattern->NewNode(next_op_repr())->assert_is_op();
1906
  auto next_output = pattern->NewNode(next_op_out_repr())->AsOutput();
1907 1908 1909 1910

  // Check if op is MKL-DNN enabled
  possible_inplace_op->assert_op_attr("use_mkldnn", true);

1911
  // linked structure
1912 1913 1914
  possible_inplace_op->LinksTo({output});
  possible_inplace_op->LinksFrom({input});
  next_op->LinksFrom({output});
1915
  next_op->LinksTo({next_output});
1916 1917 1918 1919

  return possible_inplace_op;
}

1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982
// a -> transpose_op(1) -> transpose_out_a -> flatten_op(1) -> flatten_out_a
// b -> transpose_op(2) -> transpose_out_b -> flatten_op(2) -> flatten_out_b
// ...
// z -> transpose_op(n) -> transpose_out_z -> flatten_op(n) -> flatten_out_z
// flatten_out_a -> concat_op  flatten_out_b -> concat_op ... flatten_out_z ->
// concat_op
PDNode *patterns::TransposeFlattenConcat::operator()(
    std::vector<PDNode *> conv_in, int times) {
  // The times represents the repeat times of the
  // {trans, trans_out, flatten, flatten_out}
  const int kNumFields = 4;
  const int kTransOutOffset = 1;
  const int kFlattenOffset = 2;
  const int kFlattenOutOffset = 3;

  std::vector<PDNode *> nodes;

  for (int i = 0; i < times; i++) {
    nodes.push_back(
        pattern->NewNode(GetNodeName("transpose" + std::to_string(i)))
            ->assert_is_op("transpose2"));
    nodes.push_back(
        pattern->NewNode(GetNodeName("transpose_out" + std::to_string(i)))
            ->assert_is_op_output("transpose2")
            ->assert_is_op_input("flatten2", "X")
            ->AsIntermediate());
    nodes.push_back(pattern->NewNode(GetNodeName("flatten" + std::to_string(i)))
                        ->assert_is_op("flatten2"));

    nodes.push_back(
        pattern->NewNode(GetNodeName("flatten_out" + std::to_string(i)))
            ->assert_is_op_output("flatten2")
            ->assert_is_op_nth_input("concat", "X", i)
            ->AsIntermediate());
  }

  auto concat_op = pattern->NewNode(GetNodeName("concat"))
                       ->assert_is_op("concat")
                       ->assert_op_has_n_inputs("concat", times);
  auto concat_out = pattern->NewNode(GetNodeName("concat_out"))
                        ->assert_is_op_output("concat")
                        ->AsOutput();

  std::vector<PDNode *> flatten_outs;
  for (int i = 0; i < times; i++) {
    conv_in[i]->AsInput();
    // trans
    nodes[i * kNumFields]->LinksFrom({conv_in[i]});
    // trans_out
    nodes[i * kNumFields + kTransOutOffset]->LinksFrom({nodes[i * kNumFields]});
    // flatten
    nodes[i * kNumFields + kFlattenOffset]->LinksFrom(
        {nodes[i * kNumFields + kTransOutOffset]});
    // flatten_out
    nodes[i * kNumFields + kFlattenOutOffset]->LinksFrom(
        {nodes[i * kNumFields + kFlattenOffset]});
    flatten_outs.push_back(nodes[i * kNumFields + kFlattenOutOffset]);
  }

  concat_op->LinksFrom(flatten_outs).LinksTo({concat_out});
  return concat_out;
}

1983 1984 1985
void patterns::DeleteQuantOpFuse::operator()(PDNode *input_act_node,
                                             const std::string &quant_type) {
  auto *input_scale_node = pattern->NewNode(GetNodeName("input_scale_node"))
1986 1987
                               ->assert_is_op_input(quant_type, "InScale")
                               ->AsInput();
1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019
  auto *quant_node =
      pattern->NewNode(GetNodeName("quant_node"))->assert_is_op(quant_type);
  auto *output_scale_node = pattern->NewNode(GetNodeName("output_scale_node"))
                                ->assert_is_op_output(quant_type, "OutScale")
                                ->AsOutput();
  auto *output_act_node = pattern->NewNode(GetNodeName("output_act_node"))
                              ->assert_is_op_output(quant_type, "Out")
                              ->AsOutput();
  quant_node->LinksFrom({input_scale_node, input_act_node});
  output_scale_node->LinksFrom({quant_node});
  output_act_node->LinksFrom({quant_node});
}

void patterns::DequantOpFuse::operator()(PDNode *quantized_op_input,
                                         const std::string &quantized_op_type,
                                         const std::string &dequant_type,
                                         const std::string &weight_name) {
  auto *quantized_op_weight =
      pattern->NewNode(GetNodeName("quantized_op_weight"))
          ->assert_is_op_input(quantized_op_type, weight_name)
          ->AsInput();
  auto *quantized_op = pattern->NewNode(GetNodeName("quantized_op"))
                           ->assert_is_op(quantized_op_type);
  auto *quantized_op_out = pattern->NewNode(GetNodeName("quantized_op_out"))
                               ->assert_is_op_output(quantized_op_type)
                               ->assert_is_op_input(dequant_type, "X");
  auto *dequant_op =
      pattern->NewNode(GetNodeName("dequant_op"))->assert_is_op(dequant_type);
  auto *dequant_op_out = pattern->NewNode(GetNodeName("dequant_op_out"))
                             ->assert_is_op_output(dequant_type, "Out")
                             ->AsOutput();
  PDNode *dequant_channel_scale = nullptr;
2020
  if (dequant_type == "fake_channel_wise_dequantize_max_abs") {
2021 2022 2023 2024
    dequant_channel_scale =
        pattern->NewNode(GetNodeName("dequant_channel_scale"))
            ->assert_is_op_nth_input(dequant_type, "Scales", 0)
            ->AsInput();
N
nhzlx 已提交
2025
  }
2026 2027
  quantized_op->LinksFrom({quantized_op_input, quantized_op_weight});
  quantized_op_out->LinksFrom({quantized_op});
N
nhzlx 已提交
2028

2029 2030 2031 2032
  if (dequant_type == "fake_channel_wise_dequantize_max_abs") {
    dequant_op->LinksFrom({quantized_op_out, dequant_channel_scale});
  } else {
    dequant_op->LinksFrom({quantized_op_out});
N
nhzlx 已提交
2033
  }
2034
  dequant_op_out->LinksFrom({dequant_op});
N
nhzlx 已提交
2035 2036
}

2037 2038 2039
void patterns::ShuffleChannelPattern::operator()(PDNode *reshape1_in) {
  auto reshape1_op =
      pattern->NewNode(reshape1_op_repr())->assert_is_op("reshape2");
2040
  reshape1_op->assert_more([&](Node *x) {
2041 2042
    return BOOST_GET_CONST(std::vector<int>, x->Op()->GetAttr("shape"))
               .size() == 5;
2043
  });
2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071

  auto reshape1_out = pattern->NewNode(reshape1_out_repr())
                          ->assert_is_op_output("reshape2", "Out")
                          ->assert_is_op_input("transpose2")
                          ->AsIntermediate();

  auto transpose_op =
      pattern->NewNode(transpose_op_repr())->assert_is_op("transpose2");

  auto transpose_out = pattern->NewNode(transpose_out_repr())
                           ->assert_is_op_output("transpose2", "Out")
                           ->assert_is_op_input("reshape2")
                           ->AsIntermediate();

  auto reshape2_op =
      pattern->NewNode(reshape2_op_repr())->assert_is_op("reshape2");
  auto reshape2_out = pattern->NewNode(reshape2_out_repr())
                          ->assert_is_op_output("reshape2", "Out")
                          ->AsOutput();

  reshape1_op->LinksFrom({reshape1_in});
  reshape1_out->LinksFrom({reshape1_op});
  transpose_op->LinksFrom({reshape1_out});
  transpose_out->LinksFrom({transpose_op});
  reshape2_op->LinksFrom({transpose_out});
  reshape2_out->LinksFrom({reshape2_op});
}

2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106
void patterns::DeleteQuantDequantOpPattern::operator()() {
  auto any_op_out =
      pattern->NewNode(any_op_out_repr())
          ->assert_is_op_input(
              "fake_quantize_dequantize_moving_average_abs_max", "X")
          ->AsInput();

  auto quant_dequant_op_inscale =
      pattern->NewNode(quant_dequant_op_inscale_repr())
          ->assert_is_op_input(
              "fake_quantize_dequantize_moving_average_abs_max", "InScale")
          ->AsInput();
  auto quant_dequant_op =
      pattern->NewNode(quant_dequant_op_repr())
          ->assert_is_op("fake_quantize_dequantize_moving_average_abs_max");

  auto quant_dequant_out =
      pattern->NewNode(quant_dequant_op_out_repr())
          ->assert_is_op_output(
              "fake_quantize_dequantize_moving_average_abs_max", "Out")
          ->AsIntermediate();

  auto quant_dequant_op_outscale =
      pattern->NewNode(quant_dequant_op_outscale_repr())
          ->assert_is_op_output(
              "fake_quantize_dequantize_moving_average_abs_max", "OutScale")
          ->AsOutput();
  auto any_op2 = pattern->NewNode(any_op2_repr())->assert_is_op()->AsOutput();

  quant_dequant_op->LinksFrom({any_op_out, quant_dequant_op_inscale});
  quant_dequant_op_outscale->LinksFrom({quant_dequant_op});
  quant_dequant_out->LinksFrom({quant_dequant_op});
  any_op2->LinksFrom({quant_dequant_out});
}

2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157
PDNode *patterns::ReshapeTransposeMatmulPattern::operator()(
    bool with_reshape_xshape, bool with_transpose_xshape) {
  auto reshape_op =
      pattern->NewNode(reshape_op_repr())->assert_is_op("reshape2");
  auto transpose_op =
      pattern->NewNode(transpose_op_repr())->assert_is_op("transpose2");
  auto matmul_op = pattern->NewNode(matmul_op_repr())->assert_is_op("matmul");

  auto reshape_in = pattern->NewNode(reshape_in_repr())
                        ->AsInput()
                        ->assert_is_op_input("reshape2", "X");

  auto reshape_out = pattern->NewNode(reshape_out_repr())
                         ->AsIntermediate()
                         ->assert_is_op_input("transpose2", "X")
                         ->assert_is_op_output("reshape2", "Out");
  if (!with_reshape_xshape)
    reshape_out->assert_is_only_output_of_op("reshape2");

  auto reshape_xshape = with_reshape_xshape
                            ? pattern->NewNode(reshape_xshape_repr())
                                  ->AsIntermediate()
                                  ->assert_is_op_output("reshape2", "XShape")
                            : nullptr;

  auto transpose_out = pattern->NewNode(transpose_out_repr())
                           ->AsIntermediate()
                           ->assert_is_op_input("matmul")
                           ->assert_is_op_output("transpose2", "Out");
  if (!with_transpose_xshape)
    transpose_out->assert_is_only_output_of_op("transpose2");

  auto transpose_xshape =
      with_transpose_xshape
          ? pattern->NewNode(transpose_xshape_repr())
                ->AsIntermediate()
                ->assert_is_op_output("transpose2", "XShape")
          : nullptr;

  auto matmul_out = pattern->NewNode(matmul_out_repr())
                        ->AsOutput()
                        ->assert_is_op_output("matmul", "Out");

  reshape_op->LinksFrom({reshape_in}).LinksTo({reshape_out});
  if (with_reshape_xshape) reshape_op->LinksTo({reshape_xshape});
  transpose_op->LinksFrom({reshape_out}).LinksTo({transpose_out});
  if (with_transpose_xshape) transpose_op->LinksTo({transpose_xshape});
  matmul_op->LinksFrom({transpose_out}).LinksTo({matmul_out});
  return matmul_out;
}

2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194
PDNode *patterns::MatmulTransposeReshapePattern::operator()() {
  auto reshape_op =
      pattern->NewNode(reshape_op_repr())->assert_is_op("reshape2");
  auto transpose_op =
      pattern->NewNode(transpose_op_repr())->assert_is_op("transpose2");
  auto matmul_op = pattern->NewNode(matmul_op_repr())->assert_is_op("matmul");

  auto matmul_out = pattern->NewNode(matmul_out_repr())
                        ->AsInput()
                        ->assert_is_op_output("matmul", "Out")
                        ->assert_is_op_input("transpose2", "X");

  auto transpose_out = pattern->NewNode(transpose_out_repr())
                           ->AsIntermediate()
                           ->assert_is_op_output("transpose2", "Out")
                           ->assert_is_op_input("reshape2", "X");

  auto transpose_out_xshape = pattern->NewNode(transpose_out_xshape_repr())
                                  ->AsIntermediate()
                                  ->assert_is_op_output("transpose2", "XShape");

  auto reshape_out = pattern->NewNode(reshape_out_repr())
                         ->AsOutput()
                         ->assert_is_op_output("reshape2");

  auto reshape_out_xshape = pattern->NewNode(reshape_out_xshape_repr())
                                ->AsIntermediate()
                                ->assert_is_op_output("reshape2", "XShape");

  matmul_op->LinksTo({matmul_out});
  transpose_op->LinksTo({transpose_out_xshape});
  reshape_op->LinksTo({reshape_out_xshape});
  transpose_op->LinksFrom({matmul_out}).LinksTo({transpose_out});
  reshape_op->LinksFrom({transpose_out}).LinksTo({reshape_out});
  return reshape_out;
}

2195 2196 2197
}  // namespace ir
}  // namespace framework
}  // namespace paddle