data_feed.cc 30.2 KB
Newer Older
W
Wang Guibao 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14
/* Copyright (c) 2016 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. */

D
dongdaxiang 已提交
15 16 17 18 19
#if defined _WIN32 || defined __APPLE__
#else
#define _LINUX
#endif

20
#include "paddle/fluid/framework/data_feed.h"
D
dongdaxiang 已提交
21
#ifdef _LINUX
D
dongdaxiang 已提交
22
#include <stdio_ext.h>
D
dongdaxiang 已提交
23
#endif
24
#include <utility>
25
#include "gflags/gflags.h"
W
Wang Guibao 已提交
26 27 28
#include "google/protobuf/io/zero_copy_stream_impl.h"
#include "google/protobuf/message.h"
#include "google/protobuf/text_format.h"
29 30
#include "io/fs.h"
#include "io/shell.h"
W
Wang Guibao 已提交
31 32
#include "paddle/fluid/framework/feed_fetch_method.h"
#include "paddle/fluid/framework/feed_fetch_type.h"
33
#include "paddle/fluid/platform/timer.h"
W
Wang Guibao 已提交
34 35 36 37 38 39 40 41

namespace paddle {
namespace framework {

void DataFeed::AddFeedVar(Variable* var, const std::string& name) {
  CheckInit();
  for (size_t i = 0; i < use_slots_.size(); ++i) {
    if (name == use_slots_[i]) {
42
      feed_vec_[i] = var->GetMutable<LoDTensor>();
W
Wang Guibao 已提交
43 44 45 46 47
    }
  }
}

bool DataFeed::SetFileList(const std::vector<std::string>& files) {
48
  std::unique_lock<std::mutex> lock(*mutex_for_pick_file_);
W
Wang Guibao 已提交
49
  CheckInit();
50 51
  // Do not set finish_set_filelist_ flag,
  // since a user may set file many times after init reader
W
Wang Guibao 已提交
52 53 54 55 56 57 58 59 60 61 62 63
  filelist_.assign(files.begin(), files.end());

  finish_set_filelist_ = true;
  return true;
}

void DataFeed::SetBatchSize(int batch_size) {
  PADDLE_ENFORCE(batch_size > 0, "Illegal batch size: %d.", batch_size);
  default_batch_size_ = batch_size;
}

bool DataFeed::PickOneFile(std::string* filename) {
64 65 66 67 68 69
  PADDLE_ENFORCE(mutex_for_pick_file_ != nullptr,
                 "should call SetFileListMutex before PickOneFile");
  PADDLE_ENFORCE(file_idx_ != nullptr,
                 "should call SetFileListIndex before PickOneFile");
  std::unique_lock<std::mutex> lock(*mutex_for_pick_file_);
  if (*file_idx_ == filelist_.size()) {
70
    VLOG(3) << "DataFeed::PickOneFile no more file to pick";
W
Wang Guibao 已提交
71 72
    return false;
  }
73 74
  VLOG(3) << "file_idx_=" << *file_idx_;
  *filename = filelist_[(*file_idx_)++];
D
dongdaxiang 已提交
75
  // LOG(ERROR) << "pick file:" << *filename;
W
Wang Guibao 已提交
76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101
  return true;
}

void DataFeed::CheckInit() {
  PADDLE_ENFORCE(finish_init_, "Initialization did not succeed.");
}

void DataFeed::CheckSetFileList() {
  PADDLE_ENFORCE(finish_set_filelist_, "Set filelist did not succeed.");
}

void DataFeed::CheckStart() {
  PADDLE_ENFORCE(finish_start_, "Datafeed has not started running yet.");
}

template <typename T>
void PrivateQueueDataFeed<T>::SetQueueSize(int queue_size) {
  PADDLE_ENFORCE(queue_size > 0, "Illegal queue size: %d.", queue_size);
  queue_size_ = queue_size;
  queue_ = std::unique_ptr<paddle::operators::reader::BlockingQueue<T>>(
      new paddle::operators::reader::BlockingQueue<T>(queue_size_));
}

template <typename T>
bool PrivateQueueDataFeed<T>::Start() {
  CheckSetFileList();
102 103
  read_thread_ = std::thread(&PrivateQueueDataFeed::ReadThread, this);
  read_thread_.detach();
W
Wang Guibao 已提交
104 105 106 107 108 109 110

  finish_start_ = true;
  return true;
}

template <typename T>
void PrivateQueueDataFeed<T>::ReadThread() {
D
dongdaxiang 已提交
111
#ifdef _LINUX
112 113 114 115 116 117 118 119 120
  std::string filename;
  while (PickOneFile(&filename)) {
    int err_no = 0;
    fp_ = fs_open_read(filename, &err_no, pipe_command_);
    __fsetlocking(&*fp_, FSETLOCKING_BYCALLER);
    T instance;
    while (ParseOneInstanceFromPipe(&instance)) {
      queue_->Send(instance);
    }
W
Wang Guibao 已提交
121
  }
122
  queue_->Close();
D
dongdaxiang 已提交
123
#endif
W
Wang Guibao 已提交
124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144
}

template <typename T>
int PrivateQueueDataFeed<T>::Next() {
  CheckStart();
  int index = 0;
  T instance;
  T ins_vec;
  while (index < default_batch_size_) {
    if (!queue_->Receive(&instance)) {
      break;
    }
    AddInstanceToInsVec(&ins_vec, instance, index++);
  }
  batch_size_ = index;
  if (batch_size_ != 0) {
    PutToFeedVec(ins_vec);
  }
  return batch_size_;
}

145
// explicit instantiation
W
Wang Guibao 已提交
146 147
template class PrivateQueueDataFeed<std::vector<MultiSlotType>>;

148 149 150
template <typename T>
InMemoryDataFeed<T>::InMemoryDataFeed() {
  cur_channel_ = 0;
151 152
  shuffled_ins_ = std::make_shared<paddle::framework::BlockingQueue<T>>();
  shuffled_ins_out_ = std::make_shared<paddle::framework::BlockingQueue<T>>();
D
dongdaxiang 已提交
153
  fleet_send_batch_size_ = 80000;  // hard code here
154 155 156 157
  memory_data_ = nullptr;
  mutex_for_update_memory_data_ = nullptr;
  this->file_idx_ = nullptr;
  this->mutex_for_pick_file_ = nullptr;
158 159 160 161 162
}

template <typename T>
bool InMemoryDataFeed<T>::Start() {
  DataFeed::CheckSetFileList();
163 164
  if (shuffled_ins_->Size() == 0 && shuffled_ins_out_->Size() == 0) {
    FillMemoryDataToChannel();
165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183
  }
  DataFeed::finish_start_ = true;
  return true;
}

template <typename T>
int InMemoryDataFeed<T>::Next() {
  DataFeed::CheckStart();
  std::shared_ptr<paddle::framework::BlockingQueue<T>> in_channel = nullptr;
  std::shared_ptr<paddle::framework::BlockingQueue<T>> out_channel = nullptr;
  if (cur_channel_ == 0) {
    in_channel = shuffled_ins_;
    out_channel = shuffled_ins_out_;
  } else {
    in_channel = shuffled_ins_out_;
    out_channel = shuffled_ins_;
  }
  CHECK(in_channel != nullptr);
  CHECK(out_channel != nullptr);
X
xujiaqi01 已提交
184 185 186
  VLOG(3) << "in_channel size=" << in_channel->Size()
          << ", out_channel size=" << out_channel->Size()
          << ", thread_id=" << thread_id_;
187
  int index = 0;
D
dongdaxiang 已提交
188 189 190 191 192
  T instance;
  T ins_vec;
  while (index < DataFeed::default_batch_size_) {
    if (in_channel->Size() == 0) {
      break;
193
    }
194 195
    in_channel->Pop(&instance);

D
dongdaxiang 已提交
196 197 198 199
    AddInstanceToInsVec(&ins_vec, instance, index++);
    out_channel->Push(std::move(instance));
  }
  DataFeed::batch_size_ = index;
200 201
  VLOG(3) << "batch_size_=" << DataFeed::batch_size_
          << ", thread_id=" << thread_id_;
D
dongdaxiang 已提交
202 203 204 205 206 207
  if (DataFeed::batch_size_ != 0) {
    PutToFeedVec(ins_vec);
  } else {
    cur_channel_ = 1 - cur_channel_;
  }
  return DataFeed::batch_size_;
208 209
}

210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234
template <typename T>
void InMemoryDataFeed<T>::SetMemoryData(void* memory_data) {
  memory_data_ = static_cast<std::vector<T>*>(memory_data);
}

template <typename T>
void InMemoryDataFeed<T>::SetMemoryDataMutex(std::mutex* mutex) {
  mutex_for_update_memory_data_ = mutex;
}

template <typename T>
void InMemoryDataFeed<T>::SetThreadId(int thread_id) {
  thread_id_ = thread_id;
}

template <typename T>
void InMemoryDataFeed<T>::SetThreadNum(int thread_num) {
  thread_num_ = thread_num;
}

template <typename T>
void InMemoryDataFeed<T>::SetTrainerNum(int trainer_num) {
  trainer_num_ = trainer_num;
}

235 236
template <typename T>
void InMemoryDataFeed<T>::PutInsToChannel(const std::string& ins_str) {
237
  std::vector<T> ins;
X
xujiaqi01 已提交
238
  DeserializeIns(&ins, ins_str);
239 240 241 242
  shuffled_ins_->Extend(std::move(ins));
  VLOG(3) << "PutInsToChannel put ins num=" << ins.size()
          << " to channel, channel size=" << shuffled_ins_->Size()
          << " thread_id=" << thread_id_;
243 244
}

245 246
template <typename T>
void InMemoryDataFeed<T>::FillMemoryDataToChannel() {
X
xujiaqi01 已提交
247
  VLOG(3) << "FillMemoryDataToChannel, thread_id=" << thread_id_;
248 249
  auto interval = GetMemoryDataInterval();
  VLOG(3) << "memory data size=" << memory_data_->size()
250 251
          << ", fill data from  [" << interval.first << ", " << interval.second
          << "), thread_id=" << thread_id_;
252
  for (int64_t i = interval.first; i < interval.second; ++i) {
253 254 255 256 257 258 259
    T& t = (*memory_data_)[i];
    shuffled_ins_->Push(std::move(t));
  }
}

template <typename T>
void InMemoryDataFeed<T>::FillChannelToMemoryData() {
D
dongdaxiang 已提交
260
#ifdef _LINUX
X
xujiaqi01 已提交
261
  VLOG(3) << "FillChannelToMemoryData, thread_id=" << thread_id_;
262 263
  std::vector<T> local_vec;
  std::shared_ptr<paddle::framework::BlockingQueue<T>> channel = nullptr;
264
  std::shared_ptr<paddle::framework::BlockingQueue<T>> pre_channel = nullptr;
265 266
  if (cur_channel_ == 0) {
    channel = shuffled_ins_;
267
    pre_channel = shuffled_ins_out_;
268 269
  } else {
    channel = shuffled_ins_out_;
270
    pre_channel = shuffled_ins_;
271 272
  }
  CHECK(channel != nullptr);
273
  CHECK(pre_channel != nullptr);
274
  CHECK_EQ(pre_channel->Size(), 0);
X
xujiaqi01 已提交
275
  local_vec.resize(channel->Size());
276
  for (int64_t i = 0; i < local_vec.size(); ++i) {
277
    channel->Pop(&local_vec[i]);
278
  }
279
  VLOG(3) << "local_vec size=" << local_vec.size()
280
          << ", thread_id=" << thread_id_;
X
xujiaqi01 已提交
281 282 283 284
  {
    std::lock_guard<std::mutex> g(*mutex_for_update_memory_data_);
    VLOG(3) << "before insert, memory_data_ size=" << memory_data_->size()
            << ", thread_id=" << thread_id_;
285
    memory_data_->insert(memory_data_->end(), local_vec.begin(),
286
                         local_vec.end());
X
xujiaqi01 已提交
287 288 289
    VLOG(3) << "after insert memory_data_ size=" << memory_data_->size()
            << ", thread_id=" << thread_id_;
  }
290
  std::vector<T>().swap(local_vec);
D
dongdaxiang 已提交
291
#endif
292 293
}

294 295
template <typename T>
void InMemoryDataFeed<T>::LoadIntoMemory() {
D
dongdaxiang 已提交
296
#ifdef _LINUX
X
xujiaqi01 已提交
297
  VLOG(3) << "LoadIntoMemory() begin, thread_id=" << thread_id_;
298 299 300
  std::vector<T> local_vec;
  std::string filename;
  while (DataFeed::PickOneFile(&filename)) {
X
xujiaqi01 已提交
301 302
    VLOG(3) << "PickOneFile, filename=" << filename
            << ", thread_id=" << thread_id_;
303
    int err_no = 0;
D
dongdaxiang 已提交
304 305
    PrivateQueueDataFeed<T>::fp_ =
        fs_open_read(filename, &err_no, PrivateQueueDataFeed<T>::pipe_command_);
306
    CHECK(PrivateQueueDataFeed<T>::fp_ != nullptr);
307 308
    __fsetlocking(&*PrivateQueueDataFeed<T>::fp_, FSETLOCKING_BYCALLER);
    T instance;
309 310
    platform::Timer timeline;
    timeline.Start();
D
dongdaxiang 已提交
311
    while (ParseOneInstanceFromPipe(&instance)) {
312 313
      local_vec.push_back(instance);
    }
314
    timeline.Pause();
315 316
    VLOG(3) << "LoadIntoMemory() read all lines, file=" << filename
            << ", cost time=" << timeline.ElapsedSec()
317
            << " seconds, thread_id=" << thread_id_;
318 319
    {
      std::lock_guard<std::mutex> lock(*mutex_for_update_memory_data_);
320
      timeline.Start();
X
xujiaqi01 已提交
321
      memory_data_->insert(memory_data_->end(),
322 323 324 325
                           std::make_move_iterator(local_vec.begin()),
                           std::make_move_iterator(local_vec.end()));
      timeline.Pause();
      VLOG(3) << "LoadIntoMemory() memory_data insert, cost time="
326
              << timeline.ElapsedSec() << " seconds, thread_id=" << thread_id_;
327
    }
328
    local_vec.clear();
329
  }
330
  std::vector<T>().swap(local_vec);
X
xujiaqi01 已提交
331
  VLOG(3) << "LoadIntoMemory() end, thread_id=" << thread_id_;
D
dongdaxiang 已提交
332
#endif
333 334 335 336
}

template <typename T>
void InMemoryDataFeed<T>::LocalShuffle() {
X
xujiaqi01 已提交
337
  VLOG(3) << "LocalShuffle() begin, thread_id=" << thread_id_;
338
  FillMemoryDataToChannel();
X
xujiaqi01 已提交
339
  VLOG(3) << "LocalShuffle() end, thread_id=" << thread_id_;
340 341
}

342
template <typename T>
343
void InMemoryDataFeed<T>::GlobalShuffle() {
D
dongdaxiang 已提交
344
#ifdef _LINUX
345
  VLOG(3) << "GlobalShuffle() begin, thread_id=" << thread_id_;
346
  auto fleet_ptr = FleetWrapper::GetInstance();
347 348 349 350 351 352 353 354 355 356
  std::vector<std::vector<T*>> send_vec(trainer_num_);
  for (auto& vec : send_vec) {
    vec.reserve(fleet_send_batch_size_);
  }
  std::vector<std::future<int32_t>> total_status;
  auto interval = GetMemoryDataInterval();
  VLOG(3) << "global shuffle data from  [" << interval.first << ", "
          << interval.second << "), thread_id=" << thread_id_;
  for (int64_t i = interval.first; i < interval.second; ++i) {
    // if get ins id, can also use hash
X
xujiaqi01 已提交
357
    // std::string ins_id = memory_data_[i].ins_id;
358
    int64_t random_num = rand_r(&rand_seed);
X
xujiaqi01 已提交
359
    int64_t node_id = random_num % trainer_num_;
360
    send_vec[node_id].push_back(&((*memory_data_)[i]));
361
    if (i % fleet_send_batch_size_ == 0 && i != 0) {
362 363 364 365
      for (int j = 0; j < send_vec.size(); ++j) {
        std::string send_str;
        SerializeIns(send_vec[j], &send_str);
        VLOG(3) << "send str_length=" << send_str.length()
366 367
                << ", ins num=" << send_vec[j].size() << " to node_id=" << j
                << ", thread_id=" << thread_id_;
368 369 370 371
        auto ret = fleet_ptr->SendClientToClientMsg(0, j, send_str);
        VLOG(3) << "end send, thread_id=" << thread_id_;
        send_vec[j].clear();
        total_status.push_back(std::move(ret));
372 373 374
      }
    }
  }
375 376 377 378
  for (int j = 0; j < send_vec.size(); ++j) {
    if (send_vec[j].size() != 0) {
      std::string send_str;
      SerializeIns(send_vec[j], &send_str);
379 380
      VLOG(3) << "send str_length=" << send_str.length() << " to node_id=" << j
              << ", thread_id=" << thread_id_;
381 382 383
      auto ret = fleet_ptr->SendClientToClientMsg(0, j, send_str);
      VLOG(3) << "end send, thread_id=" << thread_id_;
      total_status.push_back(std::move(ret));
384
    }
385 386 387 388
    std::vector<T*>().swap(send_vec[j]);
  }
  for (auto& t : total_status) {
    t.wait();
389
  }
390
  VLOG(3) << "GlobalShuffle() end, thread_id=" << thread_id_;
D
dongdaxiang 已提交
391
#endif
392 393 394 395 396 397 398 399 400 401 402 403 404 405
}

template <typename T>
std::pair<int64_t, int64_t> InMemoryDataFeed<T>::GetMemoryDataInterval() {
  int64_t start = 0;
  int64_t end = 0;
  int64_t size = memory_data_->size();
  for (int64_t i = 0; i <= static_cast<int64_t>(thread_id_); ++i) {
    int64_t len = size / static_cast<int64_t>(thread_num_) +
                  (i < (size % static_cast<int64_t>(thread_num_)));
    start = end;
    end += len;
  }
  return std::make_pair(start, end);
406 407
}

408 409 410
// explicit instantiation
template class InMemoryDataFeed<std::vector<MultiSlotType>>;

W
Wang Guibao 已提交
411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439
void MultiSlotDataFeed::Init(
    const paddle::framework::DataFeedDesc& data_feed_desc) {
  finish_init_ = false;
  finish_set_filelist_ = false;
  finish_start_ = false;

  PADDLE_ENFORCE(data_feed_desc.has_multi_slot_desc(),
                 "Multi_slot_desc has not been set.");
  paddle::framework::MultiSlotDesc multi_slot_desc =
      data_feed_desc.multi_slot_desc();
  SetBatchSize(data_feed_desc.batch_size());
  SetQueueSize(data_feed_desc.batch_size());
  size_t all_slot_num = multi_slot_desc.slots_size();
  all_slots_.resize(all_slot_num);
  all_slots_type_.resize(all_slot_num);
  use_slots_index_.resize(all_slot_num);
  use_slots_.clear();
  use_slots_is_dense_.clear();
  for (size_t i = 0; i < all_slot_num; ++i) {
    const auto& slot = multi_slot_desc.slots(i);
    all_slots_[i] = slot.name();
    all_slots_type_[i] = slot.type();
    use_slots_index_[i] = slot.is_used() ? use_slots_.size() : -1;
    if (slot.is_used()) {
      use_slots_.push_back(all_slots_[i]);
      use_slots_is_dense_.push_back(slot.is_dense());
    }
  }
  feed_vec_.resize(use_slots_.size());
440
  pipe_command_ = data_feed_desc.pipe_command();
W
Wang Guibao 已提交
441 442 443
  finish_init_ = true;
}

D
dongdaxiang 已提交
444
void MultiSlotDataFeed::ReadThread() {
445
#ifdef _LINUX
446 447 448 449
  std::string filename;
  while (PickOneFile(&filename)) {
    int err_no = 0;
    fp_ = fs_open_read(filename, &err_no, pipe_command_);
D
dongdaxiang 已提交
450
    CHECK(fp_ != nullptr);
451 452 453 454 455 456 457
    __fsetlocking(&*fp_, FSETLOCKING_BYCALLER);
    std::vector<MultiSlotType> instance;
    int ins_num = 0;
    while (ParseOneInstanceFromPipe(&instance)) {
      ins_num++;
      queue_->Send(instance);
    }
D
dongdaxiang 已提交
458
    VLOG(3) << "filename: " << filename << " inst num: " << ins_num;
D
dongdaxiang 已提交
459
  }
460
  queue_->Close();
461
#endif
D
dongdaxiang 已提交
462 463
}

W
Wang Guibao 已提交
464
bool MultiSlotDataFeed::CheckFile(const char* filename) {
465
#ifdef _LINUX
W
Wang Guibao 已提交
466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493
  CheckInit();  // get info of slots
  std::ifstream fin(filename);
  if (!fin.good()) {
    VLOG(1) << "error: open file<" << filename << "> fail";
    return false;
  }
  std::string line;
  int instance_cout = 0;
  std::string all_slots_alias = "";
  for (const auto& alias : all_slots_) {
    all_slots_alias += alias + " ";
  }
  std::string use_slots_alias = "";
  for (const auto& alias : use_slots_) {
    use_slots_alias += alias + " ";
  }
  VLOG(3) << "total slots num: " << all_slots_.size();
  VLOG(3) << "total slots alias: " << all_slots_alias;
  VLOG(3) << "used slots num: " << use_slots_.size();
  VLOG(3) << "used slots alias: " << use_slots_alias;
  while (getline(fin, line)) {
    ++instance_cout;
    const char* str = line.c_str();
    char* endptr = const_cast<char*>(str);
    int len = line.length();
    for (size_t i = 0; i < all_slots_.size(); ++i) {
      int num = strtol(endptr, &endptr, 10);
      if (num < 0) {
494 495
        VLOG(0) << "error: the number of ids is a negative number: " << num;
        VLOG(0) << "please check line<" << instance_cout << "> in file<"
W
Wang Guibao 已提交
496 497 498
                << filename << ">";
        return false;
      } else if (num == 0) {
499
        VLOG(0)
W
Wang Guibao 已提交
500 501 502 503
            << "error: the number of ids can not be zero, you need "
               "padding it in data generator; or if there is something wrong"
               " with the data, please check if the data contains unresolvable "
               "characters.";
504
        VLOG(0) << "please check line<" << instance_cout << "> in file<"
W
Wang Guibao 已提交
505 506 507
                << filename << ">";
        return false;
      } else if (errno == ERANGE || num > INT_MAX) {
508 509
        VLOG(0) << "error: the number of ids greater than INT_MAX";
        VLOG(0) << "please check line<" << instance_cout << "> in file<"
W
Wang Guibao 已提交
510 511 512 513 514 515 516
                << filename << ">";
        return false;
      }
      if (all_slots_type_[i] == "float") {
        for (int i = 0; i < num; ++i) {
          strtof(endptr, &endptr);
          if (errno == ERANGE) {
517
            VLOG(0) << "error: the value is out of the range of "
W
Wang Guibao 已提交
518
                       "representable values for float";
519
            VLOG(0) << "please check line<" << instance_cout << "> in file<"
W
Wang Guibao 已提交
520 521 522 523
                    << filename << ">";
            return false;
          }
          if (i + 1 != num && endptr - str == len) {
524 525
            VLOG(0) << "error: there is a wrong with the number of ids.";
            VLOG(0) << "please check line<" << instance_cout << "> in file<"
W
Wang Guibao 已提交
526 527 528 529 530 531 532 533
                    << filename << ">";
            return false;
          }
        }
      } else if (all_slots_type_[i] == "uint64") {
        for (int i = 0; i < num; ++i) {
          strtoull(endptr, &endptr, 10);
          if (errno == ERANGE) {
534
            VLOG(0) << "error: the value is out of the range of "
W
Wang Guibao 已提交
535
                       "representable values for uint64_t";
536
            VLOG(0) << "please check line<" << instance_cout << "> in file<"
W
Wang Guibao 已提交
537 538 539 540
                    << filename << ">";
            return false;
          }
          if (i + 1 != num && endptr - str == len) {
541 542
            VLOG(0) << "error: there is a wrong with the number of ids.";
            VLOG(0) << "please check line<" << instance_cout << "> in file<"
W
Wang Guibao 已提交
543 544 545 546 547
                    << filename << ">";
            return false;
          }
        }
      } else {
548
        VLOG(0) << "error: this type<" << all_slots_type_[i]
W
Wang Guibao 已提交
549 550 551 552
                << "> is not supported";
        return false;
      }
    }
553 554 555
    // It may be added '\t' character to the end of the output of reduce
    // task when processes data by Hadoop(when the output of the reduce
    // task of Hadoop has only one field, it will add a '\t' at the end
556 557 558 559 560
    // of the line by default, and you can use this option to avoid it:
    // `-D mapred.textoutputformat.ignoreseparator=true`), which does
    // not affect the correctness of the data. Therefore, it should be
    // judged that the data is not normal when the end of each line of
    // data contains characters which are not spaces.
561 562 563 564 565 566 567 568
    while (endptr - str != len) {
      if (!isspace(*(endptr++))) {
        VLOG(0)
            << "error: there is some extra characters at the end of the line.";
        VLOG(0) << "please check line<" << instance_cout << "> in file<"
                << filename << ">";
        return false;
      }
W
Wang Guibao 已提交
569 570 571 572
    }
  }
  VLOG(3) << "instances cout: " << instance_cout;
  VLOG(3) << "The file format is correct";
573
#endif
W
Wang Guibao 已提交
574 575 576
  return true;
}

D
dongdaxiang 已提交
577 578
bool MultiSlotDataFeed::ParseOneInstanceFromPipe(
    std::vector<MultiSlotType>* instance) {
579
#ifdef _LINUX
580 581 582
  thread_local string::LineFileReader reader;

  if (!reader.getline(&*(fp_.get()))) {
D
dongdaxiang 已提交
583 584
    return false;
  } else {
585 586 587
    int use_slots_num = use_slots_.size();
    instance->resize(use_slots_num);

D
dongdaxiang 已提交
588 589
    const char* str = reader.get();
    std::string line = std::string(str);
590
    // VLOG(3) << line;
D
dongdaxiang 已提交
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
    char* endptr = const_cast<char*>(str);
    int pos = 0;
    for (size_t i = 0; i < use_slots_index_.size(); ++i) {
      int idx = use_slots_index_[i];
      int num = strtol(&str[pos], &endptr, 10);
      PADDLE_ENFORCE(
          num,
          "The number of ids can not be zero, you need padding "
          "it in data generator; or if there is something wrong with "
          "the data, please check if the data contains unresolvable "
          "characters.\nplease check this error line: %s",
          str);
      if (idx != -1) {
        (*instance)[idx].Init(all_slots_type_[i]);
        if ((*instance)[idx].GetType()[0] == 'f') {  // float
          for (int j = 0; j < num; ++j) {
            float feasign = strtof(endptr, &endptr);
            (*instance)[idx].AddValue(feasign);
          }
        } else if ((*instance)[idx].GetType()[0] == 'u') {  // uint64
          for (int j = 0; j < num; ++j) {
            uint64_t feasign = (uint64_t)strtoull(endptr, &endptr, 10);
            (*instance)[idx].AddValue(feasign);
          }
        }
        pos = endptr - str;
      } else {
        for (int j = 0; j <= num; ++j) {
D
dongdaxiang 已提交
619 620 621 622
          // pos = line.find_first_of(' ', pos + 1);
          while (line[pos + 1] != ' ') {
            pos++;
          }
D
dongdaxiang 已提交
623 624 625 626 627
        }
      }
    }
    return true;
  }
628 629 630
#else
  return true;
#endif
D
dongdaxiang 已提交
631 632
}

W
Wang Guibao 已提交
633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651
bool MultiSlotDataFeed::ParseOneInstance(std::vector<MultiSlotType>* instance) {
  std::string line;
  if (getline(file_, line)) {
    int use_slots_num = use_slots_.size();
    instance->resize(use_slots_num);
    // parse line
    const char* str = line.c_str();
    char* endptr = const_cast<char*>(str);
    int pos = 0;
    for (size_t i = 0; i < use_slots_index_.size(); ++i) {
      int idx = use_slots_index_[i];
      int num = strtol(&str[pos], &endptr, 10);
      PADDLE_ENFORCE(
          num,
          "The number of ids can not be zero, you need padding "
          "it in data generator; or if there is something wrong with "
          "the data, please check if the data contains unresolvable "
          "characters.\nplease check this error line: %s",
          str);
652

W
Wang Guibao 已提交
653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688
      if (idx != -1) {
        (*instance)[idx].Init(all_slots_type_[i]);
        if ((*instance)[idx].GetType()[0] == 'f') {  // float
          for (int j = 0; j < num; ++j) {
            float feasign = strtof(endptr, &endptr);
            (*instance)[idx].AddValue(feasign);
          }
        } else if ((*instance)[idx].GetType()[0] == 'u') {  // uint64
          for (int j = 0; j < num; ++j) {
            uint64_t feasign = (uint64_t)strtoull(endptr, &endptr, 10);
            (*instance)[idx].AddValue(feasign);
          }
        }
        pos = endptr - str;
      } else {
        for (int j = 0; j <= num; ++j) {
          pos = line.find_first_of(' ', pos + 1);
        }
      }
    }
  } else {
    return false;
  }
  return true;
}

void MultiSlotDataFeed::AddInstanceToInsVec(
    std::vector<MultiSlotType>* ins_vec,
    const std::vector<MultiSlotType>& instance, int index) {
  if (index == 0) {
    ins_vec->resize(instance.size());
    for (size_t i = 0; i < instance.size(); ++i) {
      (*ins_vec)[i].Init(instance[i].GetType());
      (*ins_vec)[i].InitOffset();
    }
  }
689

W
Wang Guibao 已提交
690 691 692 693 694 695 696 697 698 699 700
  for (size_t i = 0; i < instance.size(); ++i) {
    (*ins_vec)[i].AddIns(instance[i]);
  }
}

void MultiSlotDataFeed::PutToFeedVec(
    const std::vector<MultiSlotType>& ins_vec) {
  for (size_t i = 0; i < use_slots_.size(); ++i) {
    const auto& type = ins_vec[i].GetType();
    const auto& offset = ins_vec[i].GetOffset();
    int total_instance = static_cast<int>(offset.back());
701

W
Wang Guibao 已提交
702 703
    if (type[0] == 'f') {  // float
      const auto& feasign = ins_vec[i].GetFloatData();
704 705 706
      float* tensor_ptr = feed_vec_[i]->mutable_data<float>(
          {total_instance, 1}, platform::CPUPlace());
      memcpy(tensor_ptr, &feasign[0], total_instance * sizeof(float));
W
Wang Guibao 已提交
707 708 709
    } else if (type[0] == 'u') {  // uint64
      // no uint64_t type in paddlepaddle
      const auto& feasign = ins_vec[i].GetUint64Data();
710 711 712 713
      int64_t* tensor_ptr = feed_vec_[i]->mutable_data<int64_t>(
          {total_instance, 1}, platform::CPUPlace());
      memcpy(tensor_ptr, &feasign[0], total_instance * sizeof(int64_t));
    }
714

715 716 717 718 719
    LoD data_lod{offset};
    feed_vec_[i]->set_lod(data_lod);
    if (use_slots_is_dense_[i]) {
      int dim = total_instance / batch_size_;
      feed_vec_[i]->Resize({batch_size_, dim});
W
Wang Guibao 已提交
720 721 722 723
    }
  }
}

724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768
void MultiSlotInMemoryDataFeed::Init(
    const paddle::framework::DataFeedDesc& data_feed_desc) {
  finish_init_ = false;
  finish_set_filelist_ = false;
  finish_start_ = false;

  PADDLE_ENFORCE(data_feed_desc.has_multi_slot_desc(),
                 "Multi_slot_desc has not been set.");
  paddle::framework::MultiSlotDesc multi_slot_desc =
      data_feed_desc.multi_slot_desc();
  SetBatchSize(data_feed_desc.batch_size());
  SetQueueSize(data_feed_desc.batch_size());
  size_t all_slot_num = multi_slot_desc.slots_size();
  all_slots_.resize(all_slot_num);
  all_slots_type_.resize(all_slot_num);
  use_slots_index_.resize(all_slot_num);
  use_slots_.clear();
  use_slots_is_dense_.clear();
  for (size_t i = 0; i < all_slot_num; ++i) {
    const auto& slot = multi_slot_desc.slots(i);
    all_slots_[i] = slot.name();
    all_slots_type_[i] = slot.type();
    use_slots_index_[i] = slot.is_used() ? use_slots_.size() : -1;
    if (slot.is_used()) {
      use_slots_.push_back(all_slots_[i]);
      use_slots_is_dense_.push_back(slot.is_dense());
    }
  }
  feed_vec_.resize(use_slots_.size());
  pipe_command_ = data_feed_desc.pipe_command();
  finish_init_ = true;
}

bool MultiSlotInMemoryDataFeed::ParseOneInstanceFromPipe(
    std::vector<MultiSlotType>* instance) {
  thread_local string::LineFileReader reader;

  if (!reader.getline(&*(fp_.get()))) {
    return false;
  } else {
    int use_slots_num = use_slots_.size();
    instance->resize(use_slots_num);

    const char* str = reader.get();
    std::string line = std::string(str);
769
    // VLOG(3) << line;
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
    char* endptr = const_cast<char*>(str);
    int pos = 0;
    for (size_t i = 0; i < use_slots_index_.size(); ++i) {
      int idx = use_slots_index_[i];
      int num = strtol(&str[pos], &endptr, 10);
      PADDLE_ENFORCE(
          num,
          "The number of ids can not be zero, you need padding "
          "it in data generator; or if there is something wrong with "
          "the data, please check if the data contains unresolvable "
          "characters.\nplease check this error line: %s",
          str);
      if (idx != -1) {
        (*instance)[idx].Init(all_slots_type_[i]);
        if ((*instance)[idx].GetType()[0] == 'f') {  // float
          for (int j = 0; j < num; ++j) {
            float feasign = strtof(endptr, &endptr);
            (*instance)[idx].AddValue(feasign);
          }
        } else if ((*instance)[idx].GetType()[0] == 'u') {  // uint64
          for (int j = 0; j < num; ++j) {
            uint64_t feasign = (uint64_t)strtoull(endptr, &endptr, 10);
            (*instance)[idx].AddValue(feasign);
          }
        }
        pos = endptr - str;
      } else {
        for (int j = 0; j <= num; ++j) {
          // pos = line.find_first_of(' ', pos + 1);
          while (line[pos + 1] != ' ') {
            pos++;
          }
        }
      }
    }
    return true;
  }
}

D
dongdaxiang 已提交
809 810
bool MultiSlotInMemoryDataFeed::ParseOneInstance(
    std::vector<MultiSlotType>* instance) {
811 812 813 814
  std::string line;
  if (getline(file_, line)) {
    int use_slots_num = use_slots_.size();
    instance->resize(use_slots_num);
815
    VLOG(3) << line;
816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902
    // parse line
    const char* str = line.c_str();
    char* endptr = const_cast<char*>(str);
    int pos = 0;
    for (size_t i = 0; i < use_slots_index_.size(); ++i) {
      int idx = use_slots_index_[i];
      int num = strtol(&str[pos], &endptr, 10);
      PADDLE_ENFORCE(
          num,
          "The number of ids can not be zero, you need padding "
          "it in data generator; or if there is something wrong with "
          "the data, please check if the data contains unresolvable "
          "characters.\nplease check this error line: %s",
          str);

      if (idx != -1) {
        (*instance)[idx].Init(all_slots_type_[i]);
        if ((*instance)[idx].GetType()[0] == 'f') {  // float
          for (int j = 0; j < num; ++j) {
            float feasign = strtof(endptr, &endptr);
            (*instance)[idx].AddValue(feasign);
          }
        } else if ((*instance)[idx].GetType()[0] == 'u') {  // uint64
          for (int j = 0; j < num; ++j) {
            uint64_t feasign = (uint64_t)strtoull(endptr, &endptr, 10);
            (*instance)[idx].AddValue(feasign);
          }
        }
        pos = endptr - str;
      } else {
        for (int j = 0; j <= num; ++j) {
          pos = line.find_first_of(' ', pos + 1);
        }
      }
    }
  } else {
    return false;
  }
  return true;
}

void MultiSlotInMemoryDataFeed::AddInstanceToInsVec(
    std::vector<MultiSlotType>* ins_vec,
    const std::vector<MultiSlotType>& instance, int index) {
  if (index == 0) {
    ins_vec->resize(instance.size());
    for (size_t i = 0; i < instance.size(); ++i) {
      (*ins_vec)[i].Init(instance[i].GetType());
      (*ins_vec)[i].InitOffset();
    }
  }

  for (size_t i = 0; i < instance.size(); ++i) {
    (*ins_vec)[i].AddIns(instance[i]);
  }
}

void MultiSlotInMemoryDataFeed::PutToFeedVec(
    const std::vector<MultiSlotType>& ins_vec) {
  for (size_t i = 0; i < use_slots_.size(); ++i) {
    const auto& type = ins_vec[i].GetType();
    const auto& offset = ins_vec[i].GetOffset();
    int total_instance = static_cast<int>(offset.back());

    if (type[0] == 'f') {  // float
      const auto& feasign = ins_vec[i].GetFloatData();
      float* tensor_ptr = feed_vec_[i]->mutable_data<float>(
          {total_instance, 1}, platform::CPUPlace());
      memcpy(tensor_ptr, &feasign[0], total_instance * sizeof(float));
    } else if (type[0] == 'u') {  // uint64
      // no uint64_t type in paddlepaddle
      const auto& feasign = ins_vec[i].GetUint64Data();
      int64_t* tensor_ptr = feed_vec_[i]->mutable_data<int64_t>(
          {total_instance, 1}, platform::CPUPlace());
      memcpy(tensor_ptr, &feasign[0], total_instance * sizeof(int64_t));
    }

    LoD data_lod{offset};
    feed_vec_[i]->set_lod(data_lod);
    if (use_slots_is_dense_[i]) {
      int dim = total_instance / batch_size_;
      feed_vec_[i]->Resize({batch_size_, dim});
    }
  }
}

// todo serialize ins in global shuffle
D
dongdaxiang 已提交
903
void MultiSlotInMemoryDataFeed::SerializeIns(
904
    const std::vector<std::vector<MultiSlotType>*>& ins, std::string* str) {
905 906
  auto fleet_ptr = FleetWrapper::GetInstance();
  fleet_ptr->Serialize(ins, str);
907 908
}
// todo deserialize ins in global shuffle
909
void MultiSlotInMemoryDataFeed::DeserializeIns(
910
    std::vector<std::vector<MultiSlotType>>* ins, const std::string& str) {
911 912
  auto fleet_ptr = FleetWrapper::GetInstance();
  fleet_ptr->Deserialize(ins, str);
913 914
}

W
Wang Guibao 已提交
915 916
}  // namespace framework
}  // namespace paddle