program.py 16.5 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

    def parse_args(self, argv=None):
        args = super(ArgsParser, self).parse_args(argv)
L
add kie  
LDOUBLEV 已提交
56
        # args.config = "/Users/hongyongjie/project/PaddleOCR/configs/kie/kie_unet_sdmgr.yml"
L
LDOUBLEV 已提交
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 87
        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
88 89
default_config = {'Global': {'debug': False, }}

L
LDOUBLEV 已提交
90 91 92 93 94 95 96 97

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
98
    merge_config(default_config)
L
LDOUBLEV 已提交
99 100
    _, ext = os.path.splitext(file_path)
    assert ext in ['.yml', '.yaml'], "only support yaml files for now"
W
WenmuZhou 已提交
101
    merge_config(yaml.load(open(file_path, 'rb'), Loader=yaml.Loader))
L
LDOUBLEV 已提交
102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119
    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
120 121 122 123
            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
124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143
            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
144
        if use_gpu and not paddle.is_compiled_with_cuda():
W
WenmuZhou 已提交
145
            print(err)
L
LDOUBLEV 已提交
146 147 148 149 150
            sys.exit(1)
    except Exception as e:
        pass


W
WenmuZhou 已提交
151
def train(config,
D
dyning 已提交
152 153 154
          train_dataloader,
          valid_dataloader,
          device,
W
WenmuZhou 已提交
155 156 157 158 159 160 161 162 163 164 165
          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
166 167 168 169
    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
170
    profiler_options = config['profiler_options']
W
WenmuZhou 已提交
171

D
dyning 已提交
172
    global_step = 0
173 174
    if 'global_step' in pre_best_model_dict:
        global_step = pre_best_model_dict['global_step']
L
LDOUBLEV 已提交
175 176 177 178
    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
179 180 181 182 183
        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
184 185 186
        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
187 188
    save_epoch_step = config['Global']['save_epoch_step']
    save_model_dir = config['Global']['save_model_dir']
189 190
    if not os.path.exists(save_model_dir):
        os.makedirs(save_model_dir)
W
WenmuZhou 已提交
191 192 193 194
    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
195
    model_average = False
W
WenmuZhou 已提交
196 197
    model.train()

T
tink2123 已提交
198
    use_srn = config['Architecture']['algorithm'] == "SRN"
T
tink2123 已提交
199 200
    extra_input = config['Architecture'][
        'algorithm'] in ["SRN", "NRTR", "SAR", "SEED"]
201
    try:
L
fix bug  
LDOUBLEV 已提交
202
        model_type = config['Architecture']['model_type']
203
    except:
L
fix bug  
LDOUBLEV 已提交
204
        model_type = None
T
tink2123 已提交
205
    algorithm = config['Architecture']['algorithm']
T
tink2123 已提交
206

W
WenmuZhou 已提交
207 208 209
    if 'start_epoch' in best_model_dict:
        start_epoch = best_model_dict['start_epoch']
    else:
T
tink2123 已提交
210
        start_epoch = 1
W
WenmuZhou 已提交
211

T
tink2123 已提交
212
    for epoch in range(start_epoch, epoch_num + 1):
213 214
        train_dataloader = build_dataloader(
            config, 'Train', device, logger, seed=epoch)
W
WenmuZhou 已提交
215 216 217 218
        train_batch_cost = 0.0
        train_reader_cost = 0.0
        batch_sum = 0
        batch_start = time.time()
J
Jane-Ding 已提交
219 220
        max_iter = len(train_dataloader) - 1 if platform.system(
        ) == "Windows" else len(train_dataloader)
W
WenmuZhou 已提交
221
        for idx, batch in enumerate(train_dataloader):
L
LDOUBLEV 已提交
222
            profiler.add_profiler_step(profiler_options)
W
WenmuZhou 已提交
223
            train_reader_cost += time.time() - batch_start
J
Jane-Ding 已提交
224
            if idx >= max_iter:
W
WenmuZhou 已提交
225 226 227
                break
            lr = optimizer.get_lr()
            images = batch[0]
T
tink2123 已提交
228
            if use_srn:
T
tink2123 已提交
229
                model_average = True
T
tink2123 已提交
230
            if model_type == 'table' or extra_input:
M
refine  
MissPenguin 已提交
231
                preds = model(images, data=batch[1:])
T
tink2123 已提交
232
            else:
L
add kie  
LDOUBLEV 已提交
233
                preds = model(batch)
W
WenmuZhou 已提交
234 235
            loss = loss_class(preds, batch)
            avg_loss = loss['loss']
D
dyning 已提交
236
            avg_loss.backward()
W
WenmuZhou 已提交
237 238
            optimizer.step()
            optimizer.clear_grad()
W
WenmuZhou 已提交
239 240 241 242

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

D
dyning 已提交
243 244
            if not isinstance(lr_scheduler, float):
                lr_scheduler.step()
W
WenmuZhou 已提交
245 246 247 248 249 250

            # 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
251
            if cal_metric_during_train:  # only rec and cls need
W
WenmuZhou 已提交
252
                batch = [item.numpy() for item in batch]
M
MissPenguin 已提交
253 254 255 256 257
                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
258 259
                metric = eval_class.get_metric()
                train_stats.update(metric)
W
WenmuZhou 已提交
260 261 262 263 264 265

            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)

266 267 268
            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
269
                logs = train_stats.log()
W
WenmuZhou 已提交
270
                strs = 'epoch: [{}/{}], iter: {}, {}, reader_cost: {:.5f} s, batch_cost: {:.5f} s, samples: {}, ips: {:.5f}'.format(
W
WenmuZhou 已提交
271 272 273
                    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
274
                logger.info(strs)
W
WenmuZhou 已提交
275 276 277
                train_batch_cost = 0.0
                train_reader_cost = 0.0
                batch_sum = 0
W
WenmuZhou 已提交
278 279 280
            # 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
281 282 283 284 285 286 287
                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
288 289 290 291 292
                cur_metric = eval(
                    model,
                    valid_dataloader,
                    post_process_class,
                    eval_class,
M
refine  
MissPenguin 已提交
293
                    model_type,
T
tink2123 已提交
294
                    extra_input=extra_input)
L
LDOUBLEV 已提交
295 296 297
                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
298 299 300

                # logger metric
                if vdl_writer is not None:
L
LDOUBLEV 已提交
301
                    for k, v in cur_metric.items():
W
WenmuZhou 已提交
302 303
                        if isinstance(v, (float, int)):
                            vdl_writer.add_scalar('EVAL/{}'.format(k),
L
LDOUBLEV 已提交
304 305
                                                  cur_metric[k], global_step)
                if cur_metric[main_indicator] >= best_model_dict[
W
WenmuZhou 已提交
306
                        main_indicator]:
L
LDOUBLEV 已提交
307
                    best_model_dict.update(cur_metric)
W
WenmuZhou 已提交
308 309 310 311 312 313 314 315 316
                    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,
317 318
                        epoch=epoch,
                        global_step=global_step)
L
LDOUBLEV 已提交
319
                best_str = 'best metric, {}'.format(', '.join([
W
WenmuZhou 已提交
320 321 322 323 324 325 326 327 328
                    '{}: {}'.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
329
            optimizer.clear_grad()
330
            batch_start = time.time()
W
WenmuZhou 已提交
331 332 333 334 335 336 337 338 339
        if dist.get_rank() == 0:
            save_model(
                model,
                optimizer,
                save_model_dir,
                logger,
                is_best=False,
                prefix='latest',
                best_model_dict=best_model_dict,
340 341
                epoch=epoch,
                global_step=global_step)
W
WenmuZhou 已提交
342 343 344 345 346 347 348 349 350
        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,
351 352
                epoch=epoch,
                global_step=global_step)
L
LDOUBLEV 已提交
353
    best_str = 'best metric, {}'.format(', '.join(
W
WenmuZhou 已提交
354 355 356 357
        ['{}: {}'.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
358 359 360
    return


M
refine  
MissPenguin 已提交
361 362 363 364
def eval(model,
         valid_dataloader,
         post_process_class,
         eval_class,
L
LDOUBLEV 已提交
365
         model_type=None,
T
tink2123 已提交
366
         extra_input=False):
W
WenmuZhou 已提交
367 368 369 370
    model.eval()
    with paddle.no_grad():
        total_frame = 0.0
        total_time = 0.0
W
fix bug  
WenmuZhou 已提交
371
        pbar = tqdm(total=len(valid_dataloader), desc='eval model:')
372 373
        max_iter = len(valid_dataloader) - 1 if platform.system(
        ) == "Windows" else len(valid_dataloader)
W
WenmuZhou 已提交
374
        for idx, batch in enumerate(valid_dataloader):
375
            if idx >= max_iter:
W
WenmuZhou 已提交
376
                break
W
fix bug  
WenmuZhou 已提交
377
            images = batch[0]
W
WenmuZhou 已提交
378
            start = time.time()
T
tink2123 已提交
379
            if model_type == 'table' or extra_input:
M
refine  
MissPenguin 已提交
380
                preds = model(images, data=batch[1:])
X
xiaoting 已提交
381
            else:
L
add kie  
LDOUBLEV 已提交
382
                preds = model(batch)
W
WenmuZhou 已提交
383 384 385 386
            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
387 388 389 390 391
            if model_type == 'table':
                eval_class(preds, batch)
            else:
                post_result = post_process_class(preds, batch[1])
                eval_class(post_result, batch)
L
LDOUBLEV 已提交
392

W
fix bug  
WenmuZhou 已提交
393
            pbar.update(1)
W
WenmuZhou 已提交
394
            total_frame += len(images)
L
LDOUBLEV 已提交
395 396
        # Get final metric,eg. acc or hmean
        metric = eval_class.get_metric()
D
dyning 已提交
397

W
fix bug  
WenmuZhou 已提交
398
    pbar.close()
W
WenmuZhou 已提交
399
    model.train()
L
LDOUBLEV 已提交
400 401
    metric['fps'] = total_frame / total_time
    return metric
L
licx 已提交
402

T
tink2123 已提交
403

404
def preprocess(is_train=False):
L
licx 已提交
405
    FLAGS = ArgsParser().parse_args()
L
LDOUBLEV 已提交
406
    profiler_options = FLAGS.profiler_options
L
licx 已提交
407 408
    config = load_config(FLAGS.config)
    merge_config(FLAGS.opt)
L
LDOUBLEV 已提交
409 410
    profile_dic = {"profiler_options": FLAGS.profiler_options}
    merge_config(profile_dic)
L
licx 已提交
411

W
WenmuZhou 已提交
412 413 414 415 416 417 418 419 420 421 422
    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
423 424 425 426 427

    # 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
428 429
    alg = config['Architecture']['algorithm']
    assert alg in [
J
Jethong 已提交
430
        'EAST', 'DB', 'SAST', 'Rosetta', 'CRNN', 'STARNet', 'RARE', 'SRN',
T
tink2123 已提交
431
        'CLS', 'PGNet', 'Distillation', 'NRTR', 'TableAttn', 'SAR', 'PSE',
L
LDOUBLEV 已提交
432
        'SEED', 'SDMGR'
W
WenmuZhou 已提交
433
    ]
W
WenmuZhou 已提交
434 435 436 437 438
    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
439

W
WenmuZhou 已提交
440 441
    device = 'gpu:{}'.format(dist.ParallelEnv().dev_id) if use_gpu else 'cpu'
    device = paddle.set_device(device)
D
dyning 已提交
442

D
dyning 已提交
443
    config['Global']['distributed'] = dist.get_world_size() != 1
W
WenmuZhou 已提交
444

D
dyning 已提交
445 446
    if config['Global']['use_visualdl']:
        from visualdl import LogWriter
L
fix bug  
LDOUBLEV 已提交
447
        save_model_dir = config['Global']['save_model_dir']
D
dyning 已提交
448 449 450 451 452 453 454 455 456
        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