callbacks.py 9.6 KB
Newer Older
K
Kaipeng Deng 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
# 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   
# limitations under the License.

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

import os
20
import sys
K
Kaipeng Deng 已提交
21
import datetime
22
import six
K
Kaipeng Deng 已提交
23

W
wangguanzhong 已提交
24
import paddle.distributed as dist
K
Kaipeng Deng 已提交
25 26 27 28

from ppdet.utils.checkpoint import save_model

from ppdet.utils.logger import setup_logger
29
logger = setup_logger('ppdet.engine')
K
Kaipeng Deng 已提交
30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

__all__ = ['Callback', 'ComposeCallback', 'LogPrinter', 'Checkpointer']


class Callback(object):
    def __init__(self, model):
        self.model = model

    def on_step_begin(self, status):
        pass

    def on_step_end(self, status):
        pass

    def on_epoch_begin(self, status):
        pass

    def on_epoch_end(self, status):
        pass


class ComposeCallback(object):
    def __init__(self, callbacks):
53 54 55 56
        callbacks = [c for c in list(callbacks) if c is not None]
        for c in callbacks:
            assert isinstance(
                c, Callback), "callback should be subclass of Callback"
K
Kaipeng Deng 已提交
57 58 59
        self._callbacks = callbacks

    def on_step_begin(self, status):
60 61
        for c in self._callbacks:
            c.on_step_begin(status)
K
Kaipeng Deng 已提交
62 63

    def on_step_end(self, status):
64 65
        for c in self._callbacks:
            c.on_step_end(status)
K
Kaipeng Deng 已提交
66 67

    def on_epoch_begin(self, status):
68 69
        for c in self._callbacks:
            c.on_epoch_begin(status)
K
Kaipeng Deng 已提交
70 71

    def on_epoch_end(self, status):
72 73
        for c in self._callbacks:
            c.on_epoch_end(status)
K
Kaipeng Deng 已提交
74 75 76 77 78 79 80


class LogPrinter(Callback):
    def __init__(self, model):
        super(LogPrinter, self).__init__(model)

    def on_step_end(self, status):
W
wangguanzhong 已提交
81
        if dist.get_world_size() < 2 or dist.get_rank() == 0:
K
Kaipeng Deng 已提交
82 83
            mode = status['mode']
            if mode == 'train':
K
Kaipeng Deng 已提交
84 85 86 87 88 89 90 91
                epoch_id = status['epoch_id']
                step_id = status['step_id']
                steps_per_epoch = status['steps_per_epoch']
                training_staus = status['training_staus']
                batch_time = status['batch_time']
                data_time = status['data_time']

                epoches = self.model.cfg.epoch
K
Kaipeng Deng 已提交
92 93
                batch_size = self.model.cfg['{}Reader'.format(mode.capitalize(
                ))]['batch_size']
K
Kaipeng Deng 已提交
94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122

                logs = training_staus.log()
                space_fmt = ':' + str(len(str(steps_per_epoch))) + 'd'
                if step_id % self.model.cfg.log_iter == 0:
                    eta_steps = (epoches - epoch_id) * steps_per_epoch - step_id
                    eta_sec = eta_steps * batch_time.global_avg
                    eta_str = str(datetime.timedelta(seconds=int(eta_sec)))
                    ips = float(batch_size) / batch_time.avg
                    fmt = ' '.join([
                        'Epoch: [{}]',
                        '[{' + space_fmt + '}/{}]',
                        'learning_rate: {lr:.6f}',
                        '{meters}',
                        'eta: {eta}',
                        'batch_cost: {btime}',
                        'data_cost: {dtime}',
                        'ips: {ips:.4f} images/s',
                    ])
                    fmt = fmt.format(
                        epoch_id,
                        step_id,
                        steps_per_epoch,
                        lr=status['learning_rate'],
                        meters=logs,
                        eta=eta_str,
                        btime=str(batch_time),
                        dtime=str(data_time),
                        ips=ips)
                    logger.info(fmt)
K
Kaipeng Deng 已提交
123
            if mode == 'eval':
K
Kaipeng Deng 已提交
124 125 126 127 128
                step_id = status['step_id']
                if step_id % 100 == 0:
                    logger.info("Eval iter: {}".format(step_id))

    def on_epoch_end(self, status):
W
wangguanzhong 已提交
129
        if dist.get_world_size() < 2 or dist.get_rank() == 0:
K
Kaipeng Deng 已提交
130 131
            mode = status['mode']
            if mode == 'eval':
K
Kaipeng Deng 已提交
132 133 134 135 136 137 138 139 140
                sample_num = status['sample_num']
                cost_time = status['cost_time']
                logger.info('Total sample number: {}, averge FPS: {}'.format(
                    sample_num, sample_num / cost_time))


class Checkpointer(Callback):
    def __init__(self, model):
        super(Checkpointer, self).__init__(model)
W
wangxinxin08 已提交
141
        cfg = self.model.cfg
142 143 144
        self.best_ap = 0.
        self.save_dir = os.path.join(self.model.cfg.save_dir,
                                     self.model.cfg.filename)
145 146 147 148
        if hasattr(self.model.model, 'student_model'):
            self.weight = self.model.model.student_model
        else:
            self.weight = self.model.model
K
Kaipeng Deng 已提交
149 150

    def on_epoch_end(self, status):
K
Kaipeng Deng 已提交
151 152
        # Checkpointer only performed during training
        mode = status['mode']
153 154 155
        epoch_id = status['epoch_id']
        weight = None
        save_name = None
W
wangguanzhong 已提交
156
        if dist.get_world_size() < 2 or dist.get_rank() == 0:
157 158
            if mode == 'train':
                end_epoch = self.model.cfg.epoch
159 160 161
                if (
                        epoch_id + 1
                ) % self.model.cfg.snapshot_epoch == 0 or epoch_id == end_epoch - 1:
162 163
                    save_name = str(
                        epoch_id) if epoch_id != end_epoch - 1 else "model_final"
164
                    weight = self.weight
165 166 167 168
            elif mode == 'eval':
                if 'save_best_model' in status and status['save_best_model']:
                    for metric in self.model._metrics:
                        map_res = metric.get_results()
169 170 171 172 173 174
                        if 'bbox' in map_res:
                            key = 'bbox'
                        elif 'keypoint' in map_res:
                            key = 'keypoint'
                        else:
                            key = 'mask'
175
                        if key not in map_res:
176
                            logger.warning("Evaluation results empty, this may be due to " \
177 178 179
                                        "training iterations being too few or not " \
                                        "loading the correct weights.")
                            return
180 181 182
                        if map_res[key][0] > self.best_ap:
                            self.best_ap = map_res[key][0]
                            save_name = 'best_model'
183
                            weight = self.weight
184 185 186 187 188
                        logger.info("Best test {} ap is {:0.3f}.".format(
                            key, self.best_ap))
            if weight:
                save_model(weight, self.model.optimizer, self.save_dir,
                           save_name, epoch_id + 1)
189 190 191 192 193 194 195 196 197 198 199 200


class WiferFaceEval(Callback):
    def __init__(self, model):
        super(WiferFaceEval, self).__init__(model)

    def on_epoch_begin(self, status):
        assert self.model.mode == 'eval', \
            "WiferFaceEval can only be set during evaluation"
        for metric in self.model._metrics:
            metric.update(self.model.model)
        sys.exit()
201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217


class VisualDLWriter(Callback):
    """
    Use VisualDL to log data or image
    """

    def __init__(self, model):
        super(VisualDLWriter, self).__init__(model)

        assert six.PY3, "VisualDL requires Python >= 3.5"
        try:
            from visualdl import LogWriter
        except Exception as e:
            logger.error('visualdl not found, plaese install visualdl. '
                         'for example: `pip install visualdl`.')
            raise e
M
Manuel Garcia 已提交
218 219
        self.vdl_writer = LogWriter(
            model.cfg.get('vdl_log_dir', 'vdl_log_dir/scalar'))
220 221 222 223 224 225 226
        self.vdl_loss_step = 0
        self.vdl_mAP_step = 0
        self.vdl_image_step = 0
        self.vdl_image_frame = 0

    def on_step_end(self, status):
        mode = status['mode']
W
wangguanzhong 已提交
227
        if dist.get_world_size() < 2 or dist.get_rank() == 0:
228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250
            if mode == 'train':
                training_staus = status['training_staus']
                for loss_name, loss_value in training_staus.get().items():
                    self.vdl_writer.add_scalar(loss_name, loss_value,
                                               self.vdl_loss_step)
                    self.vdl_loss_step += 1
            elif mode == 'test':
                ori_image = status['original_image']
                result_image = status['result_image']
                self.vdl_writer.add_image(
                    "original/frame_{}".format(self.vdl_image_frame), ori_image,
                    self.vdl_image_step)
                self.vdl_writer.add_image(
                    "result/frame_{}".format(self.vdl_image_frame),
                    result_image, self.vdl_image_step)
                self.vdl_image_step += 1
                # each frame can display ten pictures at most.
                if self.vdl_image_step % 10 == 0:
                    self.vdl_image_step = 0
                    self.vdl_image_frame += 1

    def on_epoch_end(self, status):
        mode = status['mode']
W
wangguanzhong 已提交
251
        if dist.get_world_size() < 2 or dist.get_rank() == 0:
252 253 254 255 256 257 258
            if mode == 'eval':
                for metric in self.model._metrics:
                    for key, map_value in metric.get_results().items():
                        self.vdl_writer.add_scalar("{}-mAP".format(key),
                                                   map_value[0],
                                                   self.vdl_mAP_step)
                self.vdl_mAP_step += 1