trainer.py 14.9 KB
Newer Older
F
Feng Ni 已提交
1 2 3 4 5 6 7 8 9 10 11 12
# Copyright (c) 2020 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
K
Kaipeng Deng 已提交
13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50
# limitations under the License.

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

import os
import time
import random
import datetime
import numpy as np
from PIL import Image

import paddle
from paddle.distributed import ParallelEnv
from paddle.static import InputSpec

from ppdet.core.workspace import create
from ppdet.utils.checkpoint import load_weight, load_pretrain_weight
from ppdet.utils.visualizer import visualize_results
from ppdet.metrics import Metric, COCOMetric, VOCMetric, get_categories, get_infer_results
import ppdet.utils.stats as stats

from .callbacks import Callback, ComposeCallback, LogPrinter, Checkpointer
from .export_utils import _dump_infer_config

from ppdet.utils.logger import setup_logger
logger = setup_logger(__name__)

__all__ = ['Trainer']


class Trainer(object):
    def __init__(self, cfg, mode='train'):
        self.cfg = cfg
        assert mode.lower() in ['train', 'eval', 'test'], \
                "mode should be 'train', 'eval' or 'test'"
        self.mode = mode.lower()
51
        self.optimizer = None
K
Kaipeng Deng 已提交
52 53 54

        # build model
        self.model = create(cfg.architecture)
55 56

        # model slim build
57
        if 'slim' in cfg and cfg.slim:
58 59 60 61 62
            if self.mode == 'train':
                self.load_weights(cfg.pretrain_weights, cfg.weight_type)
            slim = create(cfg.slim)
            slim(self.model)

K
Kaipeng Deng 已提交
63 64
        # build data loader
        self.dataset = cfg['{}Dataset'.format(self.mode.capitalize())]
K
Kaipeng Deng 已提交
65
        if self.mode == 'train':
K
Kaipeng Deng 已提交
66 67
            self.loader = create('{}Reader'.format(self.mode.capitalize()))(
                self.dataset, cfg.worker_num)
K
Kaipeng Deng 已提交
68 69 70 71 72 73 74 75
        # EvalDataset build with BatchSampler to evaluate in single device
        # TODO: multi-device evaluate
        if self.mode == 'eval':
            self._eval_batch_sampler = paddle.io.BatchSampler(
                self.dataset, batch_size=self.cfg.EvalReader['batch_size'])
            self.loader = create('{}Reader'.format(self.mode.capitalize()))(
                self.dataset, cfg.worker_num, self._eval_batch_sampler)
        # TestDataset build after user set images, skip loader creation here
K
Kaipeng Deng 已提交
76 77 78 79 80 81 82 83

        # build optimizer in train mode
        if self.mode == 'train':
            steps_per_epoch = len(self.loader)
            self.lr = create('LearningRate')(steps_per_epoch)
            self.optimizer = create('OptimizerBuilder')(self.lr,
                                                        self.model.parameters())

K
Kaipeng Deng 已提交
84 85 86
        self._nranks = ParallelEnv().nranks
        self._local_rank = ParallelEnv().local_rank

K
Kaipeng Deng 已提交
87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112
        self.status = {}

        self.start_epoch = 0
        self.end_epoch = cfg.epoch

        self._weights_loaded = False

        # initial default callbacks
        self._init_callbacks()

        # initial default metrics
        self._init_metrics()
        self._reset_metrics()

    def _init_callbacks(self):
        if self.mode == 'train':
            self._callbacks = [LogPrinter(self), Checkpointer(self)]
            self._compose_callback = ComposeCallback(self._callbacks)
        elif self.mode == 'eval':
            self._callbacks = [LogPrinter(self)]
            self._compose_callback = ComposeCallback(self._callbacks)
        else:
            self._callbacks = []
            self._compose_callback = None

    def _init_metrics(self):
G
Guanghua Yu 已提交
113 114 115
        if self.mode == 'test':
            self._metrics = []
            return
K
Kaipeng Deng 已提交
116
        if self.cfg.metric == 'COCO':
W
wangxinxin08 已提交
117
            # TODO: bias should be unified
118
            bias = self.cfg['bias'] if 'bias' in self.cfg else 0
W
wangxinxin08 已提交
119 120
            self._metrics = [
                COCOMetric(
121
                    anno_file=self.dataset.get_anno(), bias=bias)
W
wangxinxin08 已提交
122
            ]
K
Kaipeng Deng 已提交
123 124 125 126 127 128 129
        elif self.cfg.metric == 'VOC':
            self._metrics = [
                VOCMetric(
                    anno_file=self.dataset.get_anno(),
                    class_num=self.cfg.num_classes,
                    map_type=self.cfg.map_type)
            ]
K
Kaipeng Deng 已提交
130
        else:
K
Kaipeng Deng 已提交
131 132
            logger.warn("Metric not support for metric type {}".format(
                self.cfg.metric))
K
Kaipeng Deng 已提交
133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168
            self._metrics = []

    def _reset_metrics(self):
        for metric in self._metrics:
            metric.reset()

    def register_callbacks(self, callbacks):
        callbacks = [h for h in list(callbacks) if h is not None]
        for c in callbacks:
            assert isinstance(c, Callback), \
                    "metrics shoule be instances of subclass of Metric"
        self._callbacks.extend(callbacks)
        self._compose_callback = ComposeCallback(self._callbacks)

    def register_metrics(self, metrics):
        metrics = [m for m in list(metrics) if m is not None]
        for m in metrics:
            assert isinstance(m, Metric), \
                    "metrics shoule be instances of subclass of Metric"
        self._metrics.extend(metrics)

    def load_weights(self, weights, weight_type='pretrain'):
        assert weight_type in ['pretrain', 'resume', 'finetune'], \
                "weight_type can only be 'pretrain', 'resume', 'finetune'"
        if weight_type == 'resume':
            self.start_epoch = load_weight(self.model, weights, self.optimizer)
            logger.debug("Resume weights of epoch {}".format(self.start_epoch))
        else:
            self.start_epoch = 0
            load_pretrain_weight(self.model, weights,
                                 self.cfg.get('load_static_weights', False),
                                 weight_type)
            logger.debug("Load {} weights {} to start training".format(
                weight_type, weights))
        self._weights_loaded = True

K
Kaipeng Deng 已提交
169
    def train(self, validate=False):
K
Kaipeng Deng 已提交
170 171 172 173 174 175
        assert self.mode == 'train', "Model not in 'train' mode"

        # if no given weights loaded, load backbone pretrain weights as default
        if not self._weights_loaded:
            self.load_weights(self.cfg.pretrain_weights)

176
        model = self.model
K
Kaipeng Deng 已提交
177 178
        if self._nranks > 1:
            model = paddle.DataParallel(self.model)
179 180
        else:
            model = self.model
K
Kaipeng Deng 已提交
181

K
Kaipeng Deng 已提交
182 183 184 185 186 187 188 189 190 191 192 193 194
        self.status.update({
            'epoch_id': self.start_epoch,
            'step_id': 0,
            'steps_per_epoch': len(self.loader)
        })

        self.status['batch_time'] = stats.SmoothedValue(
            self.cfg.log_iter, fmt='{avg:.4f}')
        self.status['data_time'] = stats.SmoothedValue(
            self.cfg.log_iter, fmt='{avg:.4f}')
        self.status['training_staus'] = stats.TrainingStats(self.cfg.log_iter)

        for epoch_id in range(self.start_epoch, self.cfg.epoch):
K
Kaipeng Deng 已提交
195
            self.status['mode'] = 'train'
K
Kaipeng Deng 已提交
196 197 198
            self.status['epoch_id'] = epoch_id
            self._compose_callback.on_epoch_begin(self.status)
            self.loader.dataset.set_epoch(epoch_id)
K
Kaipeng Deng 已提交
199
            model.train()
K
Kaipeng Deng 已提交
200 201 202 203 204 205 206
            iter_tic = time.time()
            for step_id, data in enumerate(self.loader):
                self.status['data_time'].update(time.time() - iter_tic)
                self.status['step_id'] = step_id
                self._compose_callback.on_step_begin(self.status)

                # model forward
K
Kaipeng Deng 已提交
207
                outputs = model(data)
K
Kaipeng Deng 已提交
208 209 210 211 212 213 214 215 216 217
                loss = outputs['loss']

                # model backward
                loss.backward()
                self.optimizer.step()
                curr_lr = self.optimizer.get_lr()
                self.lr.step()
                self.optimizer.clear_grad()
                self.status['learning_rate'] = curr_lr

K
Kaipeng Deng 已提交
218
                if self._nranks < 2 or self._local_rank == 0:
K
Kaipeng Deng 已提交
219 220 221 222
                    self.status['training_staus'].update(outputs)

                self.status['batch_time'].update(time.time() - iter_tic)
                self._compose_callback.on_step_end(self.status)
F
Feng Ni 已提交
223
                iter_tic = time.time()
K
Kaipeng Deng 已提交
224

K
Kaipeng Deng 已提交
225 226
            self._compose_callback.on_epoch_end(self.status)

K
Kaipeng Deng 已提交
227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244
            if validate and (self._nranks < 2 or self._local_rank == 0) \
                    and (epoch_id % self.cfg.snapshot_epoch == 0 \
                             or epoch_id == self.end_epoch - 1):
                if not hasattr(self, '_eval_loader'):
                    # build evaluation dataset and loader
                    self._eval_dataset = self.cfg.EvalDataset
                    self._eval_batch_sampler = \
                        paddle.io.BatchSampler(
                            self._eval_dataset,
                            batch_size=self.cfg.EvalReader['batch_size'])
                    self._eval_loader = create('EvalReader')(
                        self._eval_dataset,
                        self.cfg.worker_num,
                        batch_sampler=self._eval_batch_sampler)
                with paddle.no_grad():
                    self._eval_with_loader(self._eval_loader)

    def _eval_with_loader(self, loader):
K
Kaipeng Deng 已提交
245 246 247
        sample_num = 0
        tic = time.time()
        self._compose_callback.on_epoch_begin(self.status)
K
Kaipeng Deng 已提交
248 249 250
        self.status['mode'] = 'eval'
        self.model.eval()
        for step_id, data in enumerate(loader):
K
Kaipeng Deng 已提交
251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273
            self.status['step_id'] = step_id
            self._compose_callback.on_step_begin(self.status)
            # forward
            outs = self.model(data)

            # update metrics
            for metric in self._metrics:
                metric.update(data, outs)

            sample_num += data['im_id'].numpy().shape[0]
            self._compose_callback.on_step_end(self.status)

        self.status['sample_num'] = sample_num
        self.status['cost_time'] = time.time() - tic
        self._compose_callback.on_epoch_end(self.status)

        # accumulate metric to log out
        for metric in self._metrics:
            metric.accumulate()
            metric.log()
        # reset metric states for metric may performed multiple times
        self._reset_metrics()

K
Kaipeng Deng 已提交
274 275 276
    def evaluate(self):
        self._eval_with_loader(self.loader)

K
Kaipeng Deng 已提交
277 278 279 280 281 282 283
    def predict(self, images, draw_threshold=0.5, output_dir='output'):
        self.dataset.set_images(images)
        loader = create('TestReader')(self.dataset, 0)

        imid2path = self.dataset.get_imid2path()

        anno_file = self.dataset.get_anno()
284
        clsid2catid, catid2name = get_categories(self.cfg.metric, anno_file)
K
Kaipeng Deng 已提交
285

K
Kaipeng Deng 已提交
286 287 288
        # Run Infer 
        self.status['mode'] = 'test'
        self.model.eval()
K
Kaipeng Deng 已提交
289 290 291 292 293 294
        for step_id, data in enumerate(loader):
            self.status['step_id'] = step_id
            # forward
            outs = self.model(data)
            for key in ['im_shape', 'scale_factor', 'im_id']:
                outs[key] = data[key]
G
Guanghua Yu 已提交
295 296
            for key, value in outs.items():
                outs[key] = value.numpy()
K
Kaipeng Deng 已提交
297 298 299 300 301 302 303 304 305 306 307 308 309

            batch_res = get_infer_results(outs, clsid2catid)
            bbox_num = outs['bbox_num']
            start = 0
            for i, im_id in enumerate(outs['im_id']):
                image_path = imid2path[int(im_id)]
                image = Image.open(image_path).convert('RGB')
                end = start + bbox_num[i]

                bbox_res = batch_res['bbox'][start:end] \
                        if 'bbox' in batch_res else None
                mask_res = batch_res['mask'][start:end] \
                        if 'mask' in batch_res else None
G
Guanghua Yu 已提交
310 311 312
                segm_res = batch_res['segm'][start:end] \
                        if 'segm' in batch_res else None
                image = visualize_results(image, bbox_res, mask_res, segm_res,
K
Kaipeng Deng 已提交
313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333
                                          int(outs['im_id']), catid2name,
                                          draw_threshold)

                # save image with detection
                save_name = self._get_save_image_name(output_dir, image_path)
                logger.info("Detection bbox results save in {}".format(
                    save_name))
                image.save(save_name, quality=95)
                start = end

    def _get_save_image_name(self, output_dir, image_path):
        """
        Get save image name from source image path.
        """
        if not os.path.exists(output_dir):
            os.makedirs(output_dir)
        image_name = os.path.split(image_path)[-1]
        name, ext = os.path.splitext(image_name)
        return os.path.join(output_dir, "{}".format(name)) + ext

    def export(self, output_dir='output_inference'):
334
        self.model.eval()
K
Kaipeng Deng 已提交
335 336 337 338 339 340 341 342 343 344 345
        model_name = os.path.splitext(os.path.split(self.cfg.filename)[-1])[0]
        save_dir = os.path.join(output_dir, model_name)
        if not os.path.exists(save_dir):
            os.makedirs(save_dir)
        image_shape = None
        if 'inputs_def' in self.cfg['TestReader']:
            inputs_def = self.cfg['TestReader']['inputs_def']
            image_shape = inputs_def.get('image_shape', None)
        if image_shape is None:
            image_shape = [3, None, None]

K
Kaipeng Deng 已提交
346 347
        self.model.eval()

K
Kaipeng Deng 已提交
348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363
        # Save infer cfg
        _dump_infer_config(self.cfg,
                           os.path.join(save_dir, 'infer_cfg.yml'), image_shape,
                           self.model)

        input_spec = [{
            "image": InputSpec(
                shape=[None] + image_shape, name='image'),
            "im_shape": InputSpec(
                shape=[None, 2], name='im_shape'),
            "scale_factor": InputSpec(
                shape=[None, 2], name='scale_factor')
        }]

        # dy2st and save model
        static_model = paddle.jit.to_static(self.model, input_spec=input_spec)
364 365 366 367 368 369 370 371 372
        # NOTE: dy2st do not pruned program, but jit.save will prune program
        # input spec, prune input spec here and save with pruned input spec
        pruned_input_spec = self._prune_input_spec(
            input_spec, static_model.forward.main_program,
            static_model.forward.outputs)
        paddle.jit.save(
            static_model,
            os.path.join(save_dir, 'model'),
            input_spec=pruned_input_spec)
K
Kaipeng Deng 已提交
373
        logger.info("Export model and saved in {}".format(save_dir))
374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390

    def _prune_input_spec(self, input_spec, program, targets):
        # try to prune static program to figure out pruned input spec
        # so we perform following operations in static mode
        paddle.enable_static()
        pruned_input_spec = [{}]
        program = program.clone()
        program = program._prune(targets=targets)
        global_block = program.global_block()
        for name, spec in input_spec[0].items():
            try:
                v = global_block.var(name)
                pruned_input_spec[0][name] = spec
            except Exception:
                pass
        paddle.disable_static()
        return pruned_input_spec