program.py 18.2 KB
Newer Older
M
refine  
MissPenguin 已提交
1
# Copyright (c) 2021 PaddlePaddle Authors. All Rights Reserved.
L
LDOUBLEV 已提交
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
#
# 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.

from __future__ import absolute_import
from __future__ import division
from __future__ import print_function

W
WenmuZhou 已提交
19
import os
L
LDOUBLEV 已提交
20
import sys
21
import platform
L
LDOUBLEV 已提交
22 23
import yaml
import time
W
WenmuZhou 已提交
24 25 26 27 28 29
import shutil
import paddle
import paddle.distributed as dist
from tqdm import tqdm
from argparse import ArgumentParser, RawDescriptionHelpFormatter

L
LDOUBLEV 已提交
30 31
from ppocr.utils.stats import TrainingStats
from ppocr.utils.save_load import save_model
D
dyning 已提交
32 33
from ppocr.utils.utility import print_dict
from ppocr.utils.logging import get_logger
L
LDOUBLEV 已提交
34
from ppocr.utils import profiler
D
dyning 已提交
35 36
from ppocr.data import build_dataloader
import numpy as np
L
LDOUBLEV 已提交
37

D
dyning 已提交
38

L
LDOUBLEV 已提交
39 40 41 42 43 44 45
class ArgsParser(ArgumentParser):
    def __init__(self):
        super(ArgsParser, self).__init__(
            formatter_class=RawDescriptionHelpFormatter)
        self.add_argument("-c", "--config", help="configuration file to use")
        self.add_argument(
            "-o", "--opt", nargs='+', help="set configuration options")
L
LDOUBLEV 已提交
46 47 48 49 50 51 52
        self.add_argument(
            '-p',
            '--profiler_options',
            type=str,
            default=None,
            help='The option of profiler, which should be in format \"key1=value1;key2=value2;key3=value3\".'
        )
L
LDOUBLEV 已提交
53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86

    def parse_args(self, argv=None):
        args = super(ArgsParser, self).parse_args(argv)
        assert args.config is not None, \
            "Please specify --config=configure_file_path."
        args.opt = self._parse_opt(args.opt)
        return args

    def _parse_opt(self, opts):
        config = {}
        if not opts:
            return config
        for s in opts:
            s = s.strip()
            k, v = s.split('=')
            config[k] = yaml.load(v, Loader=yaml.Loader)
        return config


class AttrDict(dict):
    """Single level attribute dict, NOT recursive"""

    def __init__(self, **kwargs):
        super(AttrDict, self).__init__()
        super(AttrDict, self).update(kwargs)

    def __getattr__(self, key):
        if key in self:
            return self[key]
        raise AttributeError("object has no attribute '{}'".format(key))


global_config = AttrDict()

农夫三拳_'s avatar
农夫三拳_ 已提交
87 88
default_config = {'Global': {'debug': False, }}

L
LDOUBLEV 已提交
89 90 91 92 93 94 95 96

def load_config(file_path):
    """
    Load config from yml/yaml file.
    Args:
        file_path (str): Path of the config file to be loaded.
    Returns: global config
    """
农夫三拳_'s avatar
农夫三拳_ 已提交
97
    merge_config(default_config)
L
LDOUBLEV 已提交
98 99
    _, ext = os.path.splitext(file_path)
    assert ext in ['.yml', '.yaml'], "only support yaml files for now"
W
WenmuZhou 已提交
100
    merge_config(yaml.load(open(file_path, 'rb'), Loader=yaml.Loader))
L
LDOUBLEV 已提交
101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118
    return global_config


def merge_config(config):
    """
    Merge config into global config.
    Args:
        config (dict): Config to be merged.
    Returns: global config
    """
    for key, value in config.items():
        if "." not in key:
            if isinstance(value, dict) and key in global_config:
                global_config[key].update(value)
            else:
                global_config[key] = value
        else:
            sub_keys = key.split('.')
T
tink2123 已提交
119 120 121 122
            assert (
                sub_keys[0] in global_config
            ), "the sub_keys can only be one of global_config: {}, but get: {}, please check your running command".format(
                global_config.keys(), sub_keys[0])
L
LDOUBLEV 已提交
123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142
            cur = global_config[sub_keys[0]]
            for idx, sub_key in enumerate(sub_keys[1:]):
                if idx == len(sub_keys) - 2:
                    cur[sub_key] = value
                else:
                    cur = cur[sub_key]


def check_gpu(use_gpu):
    """
    Log error and exit when set use_gpu=true in paddlepaddle
    cpu version.
    """
    err = "Config use_gpu cannot be set as true while you are " \
          "using paddlepaddle cpu version ! \nPlease try: \n" \
          "\t1. Install paddlepaddle-gpu to run model on GPU \n" \
          "\t2. Set use_gpu as false in config file to run " \
          "model on CPU"

    try:
W
WenmuZhou 已提交
143
        if use_gpu and not paddle.is_compiled_with_cuda():
W
WenmuZhou 已提交
144
            print(err)
L
LDOUBLEV 已提交
145 146 147 148 149
            sys.exit(1)
    except Exception as e:
        pass


W
WenmuZhou 已提交
150
def train(config,
D
dyning 已提交
151 152 153
          train_dataloader,
          valid_dataloader,
          device,
W
WenmuZhou 已提交
154 155 156 157 158 159 160 161 162 163 164
          model,
          loss_class,
          optimizer,
          lr_scheduler,
          post_process_class,
          eval_class,
          pre_best_model_dict,
          logger,
          vdl_writer=None):
    cal_metric_during_train = config['Global'].get('cal_metric_during_train',
                                                   False)
L
LDOUBLEV 已提交
165 166 167 168
    log_smooth_window = config['Global']['log_smooth_window']
    epoch_num = config['Global']['epoch_num']
    print_batch_step = config['Global']['print_batch_step']
    eval_batch_step = config['Global']['eval_batch_step']
L
LDOUBLEV 已提交
169
    profiler_options = config['profiler_options']
W
WenmuZhou 已提交
170

D
dyning 已提交
171
    global_step = 0
172 173
    if 'global_step' in pre_best_model_dict:
        global_step = pre_best_model_dict['global_step']
L
LDOUBLEV 已提交
174 175 176 177
    start_eval_step = 0
    if type(eval_batch_step) == list and len(eval_batch_step) >= 2:
        start_eval_step = eval_batch_step[0]
        eval_batch_step = eval_batch_step[1]
W
WenmuZhou 已提交
178 179 180 181 182
        if len(valid_dataloader) == 0:
            logger.info(
                'No Images in eval dataset, evaluation during training will be disabled'
            )
            start_eval_step = 1e111
L
LDOUBLEV 已提交
183 184 185
        logger.info(
            "During the training process, after the {}th iteration, an evaluation is run every {} iterations".
            format(start_eval_step, eval_batch_step))
L
LDOUBLEV 已提交
186 187
    save_epoch_step = config['Global']['save_epoch_step']
    save_model_dir = config['Global']['save_model_dir']
188 189
    if not os.path.exists(save_model_dir):
        os.makedirs(save_model_dir)
W
WenmuZhou 已提交
190 191 192 193
    main_indicator = eval_class.main_indicator
    best_model_dict = {main_indicator: 0}
    best_model_dict.update(pre_best_model_dict)
    train_stats = TrainingStats(log_smooth_window, ['lr'])
T
tink2123 已提交
194
    model_average = False
W
WenmuZhou 已提交
195 196
    model.train()

T
tink2123 已提交
197
    use_srn = config['Architecture']['algorithm'] == "SRN"
T
tink2123 已提交
198 199
    extra_input = config['Architecture'][
        'algorithm'] in ["SRN", "NRTR", "SAR", "SEED"]
200
    try:
L
fix bug  
LDOUBLEV 已提交
201
        model_type = config['Architecture']['model_type']
202
    except:
L
fix bug  
LDOUBLEV 已提交
203
        model_type = None
T
tink2123 已提交
204
    algorithm = config['Architecture']['algorithm']
M
refine  
MissPenguin 已提交
205

W
WenmuZhou 已提交
206 207 208
    if 'start_epoch' in best_model_dict:
        start_epoch = best_model_dict['start_epoch']
    else:
T
tink2123 已提交
209
        start_epoch = 1
W
WenmuZhou 已提交
210

T
tink2123 已提交
211
    for epoch in range(start_epoch, epoch_num + 1):
212 213
        train_dataloader = build_dataloader(
            config, 'Train', device, logger, seed=epoch)
W
WenmuZhou 已提交
214 215 216 217
        train_batch_cost = 0.0
        train_reader_cost = 0.0
        batch_sum = 0
        batch_start = time.time()
J
Jane-Ding 已提交
218 219 220
        max_iter = len(train_dataloader) - 1 if platform.system(
        ) == "Windows" else len(train_dataloader)
        for idx, batch in enumerate(train_dataloader):
L
LDOUBLEV 已提交
221
            profiler.add_profiler_step(profiler_options)
W
WenmuZhou 已提交
222
            train_reader_cost += time.time() - batch_start
J
Jane-Ding 已提交
223 224
            if idx >= max_iter:
                break
W
WenmuZhou 已提交
225 226
            lr = optimizer.get_lr()
            images = batch[0]
T
tink2123 已提交
227
            if use_srn:
T
tink2123 已提交
228
                model_average = True
T
tink2123 已提交
229
            if model_type == 'table' or extra_input:
M
refine  
MissPenguin 已提交
230 231 232
                preds = model(images, data=batch[1:])
            else:
                preds = model(images)
W
WenmuZhou 已提交
233 234
            loss = loss_class(preds, batch)
            avg_loss = loss['loss']
D
dyning 已提交
235
            avg_loss.backward()
W
WenmuZhou 已提交
236 237
            optimizer.step()
            optimizer.clear_grad()
W
WenmuZhou 已提交
238 239 240 241

            train_batch_cost += time.time() - batch_start
            batch_sum += len(images)

D
dyning 已提交
242 243
            if not isinstance(lr_scheduler, float):
                lr_scheduler.step()
W
WenmuZhou 已提交
244 245 246 247 248 249

            # logger and visualdl
            stats = {k: v.numpy().mean() for k, v in loss.items()}
            stats['lr'] = lr
            train_stats.update(stats)

L
LDOUBLEV 已提交
250
            if cal_metric_during_train:  # only rec and cls need
W
WenmuZhou 已提交
251
                batch = [item.numpy() for item in batch]
M
MissPenguin 已提交
252 253 254 255 256
                if model_type == 'table':
                    eval_class(preds, batch)
                else:
                    post_result = post_process_class(preds, batch[1])
                    eval_class(post_result, batch)
littletomatodonkey's avatar
fix doc  
littletomatodonkey 已提交
257 258
                metric = eval_class.get_metric()
                train_stats.update(metric)
W
WenmuZhou 已提交
259 260 261 262 263 264

            if vdl_writer is not None and dist.get_rank() == 0:
                for k, v in train_stats.get().items():
                    vdl_writer.add_scalar('TRAIN/{}'.format(k), v, global_step)
                vdl_writer.add_scalar('TRAIN/lr', lr, global_step)

265 266 267
            if dist.get_rank() == 0 and (
                (global_step > 0 and global_step % print_batch_step == 0) or
                (idx >= len(train_dataloader) - 1)):
W
WenmuZhou 已提交
268
                logs = train_stats.log()
W
WenmuZhou 已提交
269
                strs = 'epoch: [{}/{}], iter: {}, {}, reader_cost: {:.5f} s, batch_cost: {:.5f} s, samples: {}, ips: {:.5f}'.format(
W
WenmuZhou 已提交
270 271 272
                    epoch, epoch_num, global_step, logs, train_reader_cost /
                    print_batch_step, train_batch_cost / print_batch_step,
                    batch_sum, batch_sum / train_batch_cost)
W
WenmuZhou 已提交
273
                logger.info(strs)
W
WenmuZhou 已提交
274 275 276
                train_batch_cost = 0.0
                train_reader_cost = 0.0
                batch_sum = 0
W
WenmuZhou 已提交
277 278 279
            # eval
            if global_step > start_eval_step and \
                    (global_step - start_eval_step) % eval_batch_step == 0 and dist.get_rank() == 0:
T
tink2123 已提交
280 281 282 283 284 285 286
                if model_average:
                    Model_Average = paddle.incubate.optimizer.ModelAverage(
                        0.15,
                        parameters=model.parameters(),
                        min_average_window=10000,
                        max_average_window=15625)
                    Model_Average.apply()
T
tink2123 已提交
287 288 289 290 291
                cur_metric = eval(
                    model,
                    valid_dataloader,
                    post_process_class,
                    eval_class,
M
refine  
MissPenguin 已提交
292
                    model_type,
T
tink2123 已提交
293
                    extra_input=extra_input)
L
LDOUBLEV 已提交
294 295 296
                cur_metric_str = 'cur metric, {}'.format(', '.join(
                    ['{}: {}'.format(k, v) for k, v in cur_metric.items()]))
                logger.info(cur_metric_str)
W
WenmuZhou 已提交
297 298 299

                # logger metric
                if vdl_writer is not None:
L
LDOUBLEV 已提交
300
                    for k, v in cur_metric.items():
W
WenmuZhou 已提交
301 302
                        if isinstance(v, (float, int)):
                            vdl_writer.add_scalar('EVAL/{}'.format(k),
L
LDOUBLEV 已提交
303 304
                                                  cur_metric[k], global_step)
                if cur_metric[main_indicator] >= best_model_dict[
W
WenmuZhou 已提交
305
                        main_indicator]:
L
LDOUBLEV 已提交
306
                    best_model_dict.update(cur_metric)
W
WenmuZhou 已提交
307 308 309 310 311 312 313 314 315
                    best_model_dict['best_epoch'] = epoch
                    save_model(
                        model,
                        optimizer,
                        save_model_dir,
                        logger,
                        is_best=True,
                        prefix='best_accuracy',
                        best_model_dict=best_model_dict,
316 317
                        epoch=epoch,
                        global_step=global_step)
L
LDOUBLEV 已提交
318
                best_str = 'best metric, {}'.format(', '.join([
W
WenmuZhou 已提交
319 320 321 322 323 324 325 326 327
                    '{}: {}'.format(k, v) for k, v in best_model_dict.items()
                ]))
                logger.info(best_str)
                # logger best metric
                if vdl_writer is not None:
                    vdl_writer.add_scalar('EVAL/best_{}'.format(main_indicator),
                                          best_model_dict[main_indicator],
                                          global_step)
            global_step += 1
T
tink2123 已提交
328
            optimizer.clear_grad()
329
            batch_start = time.time()
W
WenmuZhou 已提交
330 331 332 333 334 335 336 337 338
        if dist.get_rank() == 0:
            save_model(
                model,
                optimizer,
                save_model_dir,
                logger,
                is_best=False,
                prefix='latest',
                best_model_dict=best_model_dict,
339 340
                epoch=epoch,
                global_step=global_step)
W
WenmuZhou 已提交
341 342 343 344 345 346 347 348 349
        if dist.get_rank() == 0 and epoch > 0 and epoch % save_epoch_step == 0:
            save_model(
                model,
                optimizer,
                save_model_dir,
                logger,
                is_best=False,
                prefix='iter_epoch_{}'.format(epoch),
                best_model_dict=best_model_dict,
350 351
                epoch=epoch,
                global_step=global_step)
L
LDOUBLEV 已提交
352
    best_str = 'best metric, {}'.format(', '.join(
W
WenmuZhou 已提交
353 354 355 356
        ['{}: {}'.format(k, v) for k, v in best_model_dict.items()]))
    logger.info(best_str)
    if dist.get_rank() == 0 and vdl_writer is not None:
        vdl_writer.close()
L
LDOUBLEV 已提交
357 358 359
    return


M
refine  
MissPenguin 已提交
360 361 362 363
def eval(model,
         valid_dataloader,
         post_process_class,
         eval_class,
L
LDOUBLEV 已提交
364
         model_type=None,
T
tink2123 已提交
365
         extra_input=False):
W
WenmuZhou 已提交
366 367 368 369
    model.eval()
    with paddle.no_grad():
        total_frame = 0.0
        total_time = 0.0
W
fix bug  
WenmuZhou 已提交
370
        pbar = tqdm(total=len(valid_dataloader), desc='eval model:')
371 372
        max_iter = len(valid_dataloader) - 1 if platform.system(
        ) == "Windows" else len(valid_dataloader)
W
WenmuZhou 已提交
373
        for idx, batch in enumerate(valid_dataloader):
374
            if idx >= max_iter:
W
WenmuZhou 已提交
375
                break
W
fix bug  
WenmuZhou 已提交
376
            images = batch[0]
W
WenmuZhou 已提交
377
            start = time.time()
T
tink2123 已提交
378
            if model_type == 'table' or extra_input:
M
refine  
MissPenguin 已提交
379 380 381
                preds = model(images, data=batch[1:])
            else:
                preds = model(images)
W
WenmuZhou 已提交
382 383 384 385
            batch = [item.numpy() for item in batch]
            # Obtain usable results from post-processing methods
            total_time += time.time() - start
            # Evaluate the results of the current batch
M
MissPenguin 已提交
386 387 388 389 390
            if model_type == 'table':
                eval_class(preds, batch)
            else:
                post_result = post_process_class(preds, batch[1])
                eval_class(post_result, batch)
W
fix bug  
WenmuZhou 已提交
391
            pbar.update(1)
W
WenmuZhou 已提交
392
            total_frame += len(images)
L
LDOUBLEV 已提交
393 394
        # Get final metric,eg. acc or hmean
        metric = eval_class.get_metric()
D
dyning 已提交
395

W
fix bug  
WenmuZhou 已提交
396
    pbar.close()
W
WenmuZhou 已提交
397
    model.train()
L
LDOUBLEV 已提交
398 399
    metric['fps'] = total_frame / total_time
    return metric
L
licx 已提交
400

T
tink2123 已提交
401

B
Bin Lu 已提交
402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452
def update_center(char_center, post_result, preds):
    result, label = post_result
    feats, logits = preds
    logits = paddle.argmax(logits, axis=-1)
    feats = feats.numpy()
    logits = logits.numpy()

    for idx_sample in range(len(label)):
        if result[idx_sample][0] == label[idx_sample][0]:
            feat = feats[idx_sample]
            logit = logits[idx_sample]
            for idx_time in range(len(logit)):
                index = logit[idx_time]
                if index in char_center.keys():
                    char_center[index][0] = (
                        char_center[index][0] * char_center[index][1] +
                        feat[idx_time]) / (char_center[index][1] + 1)
                    char_center[index][1] += 1
                else:
                    char_center[index] = [feat[idx_time], 1]
    return char_center


def get_center(model, eval_dataloader, post_process_class):
    pbar = tqdm(total=len(eval_dataloader), desc='get center:')
    max_iter = len(eval_dataloader) - 1 if platform.system(
    ) == "Windows" else len(eval_dataloader)
    char_center = dict()
    for idx, batch in enumerate(eval_dataloader):
        if idx >= max_iter:
            break
        images = batch[0]
        start = time.time()
        preds = model(images)

        batch = [item.numpy() for item in batch]
        # Obtain usable results from post-processing methods
        total_time += time.time() - start
        # Evaluate the results of the current batch
        post_result = post_process_class(preds, batch[1])

        #update char_center
        char_center = update_center(char_center, post_result, preds)
        pbar.update(1)

    pbar.close()
    for key in char_center.keys():
        char_center[key] = char_center[key][0]
    return char_center


453
def preprocess(is_train=False):
L
licx 已提交
454
    FLAGS = ArgsParser().parse_args()
L
LDOUBLEV 已提交
455
    profiler_options = FLAGS.profiler_options
L
licx 已提交
456 457
    config = load_config(FLAGS.config)
    merge_config(FLAGS.opt)
L
LDOUBLEV 已提交
458 459
    profile_dic = {"profiler_options": FLAGS.profiler_options}
    merge_config(profile_dic)
L
licx 已提交
460

W
WenmuZhou 已提交
461 462 463 464 465 466 467 468 469 470 471 472
    if is_train:
        # save_config
        save_model_dir = config['Global']['save_model_dir']
        os.makedirs(save_model_dir, exist_ok=True)
        with open(os.path.join(save_model_dir, 'config.yml'), 'w') as f:
            yaml.dump(
                dict(config), f, default_flow_style=False, sort_keys=False)
        log_file = '{}/train.log'.format(save_model_dir)
    else:
        log_file = None
    logger = get_logger(name='root', log_file=log_file)

L
licx 已提交
473 474 475 476
    # check if set use_gpu=True in paddlepaddle cpu version
    use_gpu = config['Global']['use_gpu']
    check_gpu(use_gpu)

W
WenmuZhou 已提交
477 478
    alg = config['Architecture']['algorithm']
    assert alg in [
J
Jethong 已提交
479
        'EAST', 'DB', 'SAST', 'Rosetta', 'CRNN', 'STARNet', 'RARE', 'SRN',
T
tink2123 已提交
480
        'CLS', 'PGNet', 'Distillation', 'NRTR', 'TableAttn', 'SAR', 'PSE',
B
Bin Lu 已提交
481 482
        'SEED'
    ]
W
WenmuZhou 已提交
483 484 485 486 487
    windows_not_support_list = ['PSE']
    if platform.system() == "Windows" and alg in windows_not_support_list:
        logger.warning('{} is not support in Windows now'.format(
            windows_not_support_list))
        sys.exit()
L
licx 已提交
488

W
WenmuZhou 已提交
489 490
    device = 'gpu:{}'.format(dist.ParallelEnv().dev_id) if use_gpu else 'cpu'
    device = paddle.set_device(device)
D
dyning 已提交
491

D
dyning 已提交
492
    config['Global']['distributed'] = dist.get_world_size() != 1
W
WenmuZhou 已提交
493

D
dyning 已提交
494 495
    if config['Global']['use_visualdl']:
        from visualdl import LogWriter
L
fix bug  
LDOUBLEV 已提交
496
        save_model_dir = config['Global']['save_model_dir']
D
dyning 已提交
497 498 499 500 501 502 503 504 505
        vdl_writer_path = '{}/vdl/'.format(save_model_dir)
        os.makedirs(vdl_writer_path, exist_ok=True)
        vdl_writer = LogWriter(logdir=vdl_writer_path)
    else:
        vdl_writer = None
    print_dict(config, logger)
    logger.info('train with paddle {} and device {}'.format(paddle.__version__,
                                                            device))
    return config, device, logger, vdl_writer