train.py

import copy
import time
from os import path as osp

import mmcv
import torch
from mmcv import Config
from mmcv.runner import get_dist_info, init_dist
from mmdet import __version__ as mmdet_version
from mmdet3d import __version__ as mmdet3d_version
from mmdet3d.apis import train_model
from mmdet3d.datasets import build_dataset
from mmdet3d.models import build_model
from mmdet3d.utils import collect_env, get_root_logger
from mmdet.apis import set_random_seed
from mmseg import __version__ as mmseg_version

from demf import parse_args


def main():
    args = parse_args()

    cfg = Config.fromfile(args.config)
    if args.cfg_options is not None:
        cfg.merge_from_dict(args.cfg_options)
    # import modules from string list.
    if cfg.get('custom_imports', None):
        from mmcv.utils import import_modules_from_strings
        import_modules_from_strings(**cfg['custom_imports'])

    # set cudnn_benchmark
    if cfg.get('cudnn_benchmark', False):
        torch.backends.cudnn.benchmark = True

    # work_dir is determined in this priority: CLI > segment in file > filename
    if args.work_dir is not None:
        # update configs according to CLI args if args.work_dir is not None
        cfg.work_dir = args.work_dir
    elif cfg.get('work_dir', None) is None:
        # use config filename as default work_dir if cfg.work_dir is None
        cfg.work_dir = osp.join('./work_dirs',
                                osp.splitext(osp.basename(args.config))[0])
    if args.resume_from is not None:
        cfg.resume_from = args.resume_from
    if args.gpu_ids is not None:
        cfg.gpu_ids = args.gpu_ids
    else:
        cfg.gpu_ids = range(1) if args.gpus is None else range(args.gpus)

    if args.autoscale_lr:
        # apply the linear scaling rule (https://arxiv.org/abs/1706.02677)
        cfg.optimizer['lr'] = cfg.optimizer['lr'] * len(cfg.gpu_ids) / 8

    # init distributed env first, since logger depends on the dist info.
    if args.launcher == 'none':
        distributed = False
    else:
        distributed = True
        init_dist(args.launcher, **cfg.dist_params)
        # re-set gpu_ids with distributed training mode
        _, world_size = get_dist_info()
        cfg.gpu_ids = range(world_size)

    # create work_dir
    mmcv.mkdir_or_exist(osp.abspath(cfg.work_dir))
    # dump config
    cfg.dump(osp.join(cfg.work_dir, osp.basename(args.config)))
    # init the logger before other steps
    timestamp = time.strftime('%Y%m%d_%H%M%S', time.localtime())
    log_file = osp.join(cfg.work_dir, f'{timestamp}.log')
    # specify logger name, if we still use 'mmdet', the output info will be
    # filtered and won't be saved in the log_file
    # TODO: ugly workaround to judge whether we are training det or seg model
    if cfg.model.type in ['EncoderDecoder3D']:
        logger_name = 'mmseg'
    else:
        logger_name = 'mmdet'
    logger = get_root_logger(
        log_file=log_file, log_level=cfg.log_level, name=logger_name)

    # init the meta dict to record some important information such as
    # environment info and seed, which will be logged
    meta = dict()
    # log env info
    env_info_dict = collect_env()
    env_info = '\n'.join([(f'{k}: {v}') for k, v in env_info_dict.items()])
    dash_line = '-' * 60 + '\n'
    logger.info('Environment info:\n' + dash_line + env_info + '\n' +
                dash_line)
    meta['env_info'] = env_info
    meta['config'] = cfg.pretty_text

    # log some basic info
    logger.info(f'Distributed training: {distributed}')
    logger.info(f'Config:\n{cfg.pretty_text}')

    # set random seeds
    if args.seed is not None:
        logger.info(f'Set random seed to {args.seed}, '
                    f'deterministic: {args.deterministic}')
        set_random_seed(args.seed, deterministic=args.deterministic)
    cfg.seed = args.seed
    meta['seed'] = args.seed
    meta['exp_name'] = osp.basename(args.config)

    model = build_model(
        cfg.model,
        train_cfg=cfg.get('train_cfg'),
        test_cfg=cfg.get('test_cfg'))
    model.init_weights()

    logger.info(f'Model:\n{model}')
    datasets = [build_dataset(cfg.data.train)]
    if len(cfg.workflow) == 2:
        val_dataset = copy.deepcopy(cfg.data.val)
        # in case we use a dataset wrapper
        if 'dataset' in cfg.data.train:
            val_dataset.pipeline = cfg.data.train.dataset.pipeline
        else:
            val_dataset.pipeline = cfg.data.train.pipeline
        # set test_mode=False here in deep copied config
        # which do not affect AP/AR calculation later
        # refer to https://mmdetection3d.readthedocs.io/en/latest/tutorials/customize_runtime.html#customize-workflow  # noqa
        val_dataset.test_mode = False
        datasets.append(build_dataset(val_dataset))
    if cfg.checkpoint_config is not None:
        # save mmdet version, config file content and class names in
        # checkpoints as meta data
        cfg.checkpoint_config.meta = dict(
            mmdet_version=mmdet_version,
            mmseg_version=mmseg_version,
            mmdet3d_version=mmdet3d_version,
            config=cfg.pretty_text,
            CLASSES=datasets[0].CLASSES,
            PALETTE=datasets[0].PALETTE  # for segmentors
            if hasattr(datasets[0], 'PALETTE') else None)
    # add an attribute for visualization convenience
    model.CLASSES = datasets[0].CLASSES
    train_model(
        model,
        datasets,
        cfg,
        distributed=distributed,
        validate=(not args.no_validate),
        timestamp=timestamp,
        meta=meta)


if __name__ == '__main__':
    main()