train.py

# 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.

import argparse

import paddle.fluid as fluid
from paddle.fluid.dygraph.parallel import ParallelEnv

import dygraph
from dygraph.cvlibs import manager
from dygraph.utils import get_environ_info
from dygraph.utils import logger
from dygraph.utils import Config
from dygraph.core import train


def parse_args():
    parser = argparse.ArgumentParser(description='Model training')
    # params of training
    parser.add_argument(
        "--config", dest="cfg", help="The config file.", default=None, type=str)
    parser.add_argument(
        '--iters',
        dest='iters',
        help='iters for training',
        type=int,
        default=None)
    parser.add_argument(
        '--batch_size',
        dest='batch_size',
        help='Mini batch size of one gpu or cpu',
        type=int,
        default=None)
    parser.add_argument(
        '--learning_rate',
        dest='learning_rate',
        help='Learning rate',
        type=float,
        default=None)
    parser.add_argument(
        '--save_interval_iters',
        dest='save_interval_iters',
        help='The interval iters for save a model snapshot',
        type=int,
        default=5)
    parser.add_argument(
        '--save_dir',
        dest='save_dir',
        help='The directory for saving the model snapshot',
        type=str,
        default='./output')
    parser.add_argument(
        '--num_workers',
        dest='num_workers',
        help='Num workers for data loader',
        type=int,
        default=0)
    parser.add_argument(
        '--do_eval',
        dest='do_eval',
        help='Eval while training',
        action='store_true')
    parser.add_argument(
        '--log_iters',
        dest='log_iters',
        help='Display logging information at every log_iters',
        default=10,
        type=int)
    parser.add_argument(
        '--use_vdl',
        dest='use_vdl',
        help='Whether to record the data to VisualDL during training',
        action='store_true')

    return parser.parse_args()


def main(args):
    env_info = get_environ_info()
    info = ['{}: {}'.format(k, v) for k, v in env_info.items()]
    info = '\n'.join(['\n', format('Environment Information', '-^48s')] + info +
                     ['-' * 48])
    logger.info(info)

    places = fluid.CUDAPlace(ParallelEnv().dev_id) \
        if env_info['Paddle compiled with cuda'] and env_info['GPUs used'] \
        else fluid.CPUPlace()

    with fluid.dygraph.guard(places):
        if not args.cfg:
            raise RuntimeError('No configuration file specified.')

        cfg = Config(args.cfg)
        train_dataset = cfg.train_dataset
        if not train_dataset:
            raise RuntimeError(
                'The training dataset is not specified in the configuration file.'
            )

        val_dataset = cfg.val_dataset if args.do_eval else None

        train(
            cfg.model,
            train_dataset,
            places=places,
            eval_dataset=val_dataset,
            optimizer=cfg.optimizer,
            save_dir=args.save_dir,
            iters=cfg.iters,
            batch_size=cfg.batch_size,
            save_interval_iters=args.save_interval_iters,
            log_iters=args.log_iters,
            num_classes=train_dataset.num_classes,
            num_workers=args.num_workers,
            use_vdl=args.use_vdl)


if __name__ == '__main__':
    args = parse_args()
    main(args)