def _non_dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    data_loaders = [
        build_dataloader(
            dataset,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            cfg.gpus,
            dist=False)
    ]
    # put model on gpus
    model = MMDataParallel(model, device_ids=range(cfg.gpus)).cuda()
    # build runner
    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(model, batch_processor, optimizer, cfg.work_dir,
                    cfg.log_level)
    runner.register_training_hooks(cfg.lr_config, cfg.optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _non_dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    data_loaders = [
        build_dataloader(
            dataset,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            cfg.gpus,
            dist=False)
    ]
    # put model on gpus
    model = MMDataParallel(model, device_ids=range(cfg.gpus)).cuda()
    # build runner
    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(model, batch_processor, optimizer, cfg.work_dir,
                    cfg.log_level)
    runner.register_training_hooks(cfg.lr_config, cfg.optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _non_dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    data_loaders = [
        build_dataloader(
            dataset,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            cfg.gpus,
            dist=False)
    ]
    # put model on gpus
    model = MMDataParallel(model, device_ids=range(cfg.gpus)).cuda()
    # build runner
    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(model, batch_processor, optimizer, cfg.work_dir,
                    cfg.log_level)
    runner.register_training_hooks(cfg.lr_config, cfg.optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _non_dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    data_loaders = [
        build_dataloader(
            dataset,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            cfg.gpus,
            dist=False)
    ]
    # put model on gpus
    model = MMDataParallel(model, device_ids=range(cfg.gpus)).cuda()
    # build runner
    runner = Runner(model, batch_processor, cfg.optimizer, cfg.work_dir,
                    cfg.log_level)
    runner.register_training_hooks(cfg.lr_config, cfg.optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _non_dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    data_loaders = [
        build_dataloader(
            dataset,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            cfg.gpus,
            dist=False)
    ]
    # put model on gpus
    model = MMDataParallel(model, device_ids=range(cfg.gpus)).cuda()
    # build runner
    runner = Runner(model, batch_processor, cfg.optimizer, cfg.work_dir,
                    cfg.log_level)
    runner.register_training_hooks(cfg.lr_config, cfg.optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _non_dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    data_loaders = [
        build_dataloader(
            dataset,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            cfg.gpus,
            dist=False)
    ]
    # put model on gpus
    model = MMDataParallel(model, device_ids=range(cfg.gpus)).cuda()
    # build runner
    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(model, batch_processor, optimizer, cfg.work_dir,
                    cfg.log_level)
    runner.register_training_hooks(cfg.lr_config, cfg.optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _non_dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    data_loaders = [
        build_dataloader(
            dataset,
            cfg.data.videos_per_gpu,
            cfg.data.workers_per_gpu,
            cfg.gpus,
            dist=False)
    ]
    # put model on gpus
    model = MMDataParallel(model, device_ids=range(cfg.gpus)).cuda()
    # build runner
    runner = Runner(model, batch_processor, cfg.optimizer, cfg.work_dir,
                    cfg.log_level)
    runner.register_training_hooks(cfg.lr_config, cfg.optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _non_dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    data_loaders = [
        build_dataloader(
            dataset,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            cfg.gpus,
            dist=False)
    ]
    # put model on gpus
    model = MMDataParallel(model, device_ids=range(cfg.gpus)).cuda()
    # build runner
    runner = Runner(model, batch_processor, cfg.optimizer, cfg.work_dir,
                    cfg.log_level)
    runner.register_training_hooks(cfg.lr_config, cfg.optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _non_dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    data_loaders = [
        build_dataloader(
            dataset,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            cfg.gpus,
            dist=False)
    ]
    # put model on gpus
    model = MMDataParallel(model, device_ids=range(cfg.gpus)).cuda()
    # build runner
    runner = Runner(model, batch_processor, cfg.optimizer, cfg.work_dir,
                    cfg.log_level)
    runner.register_training_hooks(cfg.lr_config, cfg.optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _single_train(model, data_loaders, cfg):
    if cfg.gpus > 1:
        raise NotImplemented
    # put model on gpus
    model = MMDataParallel(model, device_ids=range(cfg.gpus)).cuda()
    # build runner
    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(model, batch_processor, optimizer, cfg.work_dir,
                    cfg.log_level)
    runner.register_training_hooks(cfg.lr_config, cfg.optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    data_loaders = [
        build_dataloader(
            dataset,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            dist=True)
    ]
    # put model on gpus
    model = MMDistributedDataParallel(model.cuda())
    # build runner
    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(model, batch_processor, optimizer, cfg.work_dir,
                    cfg.log_level)
    # register hooks
    optimizer_config = DistOptimizerHook(**cfg.optimizer_config)
    runner.register_training_hooks(cfg.lr_config, optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)
    runner.register_hook(DistSamplerSeedHook())
    # register eval hooks
    if validate:
        val_dataset_cfg = cfg.data.val
        if isinstance(model.module, RPN):
            # TODO: implement recall hooks for other datasets
            runner.register_hook(CocoDistEvalRecallHook(val_dataset_cfg))
        else:
            dataset_type = getattr(datasets, val_dataset_cfg.type)
            if issubclass(dataset_type, datasets.CocoDataset):
                runner.register_hook(CocoDistEvalmAPHook(val_dataset_cfg))
            else:
                runner.register_hook(DistEvalmAPHook(val_dataset_cfg))

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _non_dist_train(
        model, train_dataset, cfg,
        eval_dataset=None, vis_dataset=None, validate=False, logger=None
):
    # prepare data loaders
    data_loaders = [
        build_data_loader(
            train_dataset,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            cfg.gpus,
            dist=False)
    ]
    # put model on gpus
    model = MMDataParallel(model, device_ids=range(cfg.gpus)).cuda()
    # build runner
    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(
        model, batch_processor, optimizer, cfg.work_dir, cfg.log_level, logger
    )
    logger.info("Register Optimizer Hook...")
    runner.register_training_hooks(
        cfg.lr_config, cfg.optimizer_config, cfg.checkpoint_config, cfg.log_config
    )
    logger.info("Register EmptyCache Hook...")
    runner.register_hook(
        EmptyCacheHook(before_epoch=True, after_iter=False, after_epoch=True),
        priority='VERY_LOW'
    )

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)

    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _non_dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    data_loaders = [
        build_dataloader(
            dataset,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            cfg.gpus,
            dist=False)
    ]
    # put model on gpus
    model = MMDataParallel(model, device_ids=range(cfg.gpus)).cuda()

    # build runner
    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(model, batch_processor, optimizer, cfg.work_dir,
                    cfg.log_level)
    # fp16 setting
    fp16_cfg = cfg.get('fp16', None)
    if fp16_cfg is not None:
        optimizer_config = Fp16OptimizerHook(
            **cfg.optimizer_config, **fp16_cfg, distributed=False)
    else:
        optimizer_config = cfg.optimizer_config
    runner.register_training_hooks(cfg.lr_config, optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _non_dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    # 返回dataloader的迭代器，采用pytorch的DataLoader方法封装数据集
    data_loaders = [
        build_dataloader(
            dataset,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            cfg.gpus,
            dist=False)
    ]
    # put model on gpus 这里多GPU输入没用list而是迭代器，注意单GPU是range(0,1),遍历的时候只有0
    model = MMDataParallel(model, device_ids=range(cfg.gpus)).cuda()

    # build runner 
    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(model, batch_processor, optimizer, cfg.work_dir,
                    cfg.log_level)
    # fp16 setting
    fp16_cfg = cfg.get('fp16', None)
    if fp16_cfg is not None:
        optimizer_config = Fp16OptimizerHook(
            **cfg.optimizer_config, **fp16_cfg, distributed=False)
    else:
        optimizer_config = cfg.optimizer_config

    # 注册钩子    
    runner.register_training_hooks(cfg.lr_config, optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)
    # 断点加载或文件加载数据
    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _non_dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    data_loaders = [
        build_dataloader(
            dataset,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            len(cfg.gpus.train),
            dist=False)
    ]
    print('dataloader built')

    model = MMDataParallel(model, device_ids=cfg.gpus.train).cuda()
    print('model paralleled')

    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(model, batch_processor, optimizer, cfg.work_dir,
                    cfg.log_level)

    runner.register_training_hooks(cfg.lr_config, cfg.optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _non_dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    data_loaders = [
        build_dataloader(
            dataset,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            len(cfg.gpus.train),
            dist=False)
    ]
    print('dataloader built')

    # put model on gpus
    model = MMDataParallel(model, device_ids=cfg.gpus.train).cuda()
    print('model paralleled')

    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(model, batch_processor, optimizer, cfg.work_dir,
                    cfg.log_level)

    runner.register_training_hooks(cfg.lr_config, cfg.optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _non_dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    data_loaders = [
        build_dataloader(
            dataset,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            len(cfg.gpus.train),
            drop_last=cfg.data.drop_last,
            dist=False)
    ]
    print('dataloader built')

    model = MMDataParallel(model, device_ids=cfg.gpus.train).cuda()
    print('model paralleled')

    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(model, batch_processor, optimizer, cfg.work_dir,
                    cfg.log_level)

    runner.register_training_hooks(cfg.lr_config, cfg.optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _non_dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    data_loaders = [
        build_dataloader(
            dataset,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            len(cfg.gpus.train),
            dist=False)
    ]
    print('dataloader built')

    # put model on gpus
    model = MMDataParallel(model, device_ids=cfg.gpus.train).cuda()
    print('model paralleled')

    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(model, batch_processor, optimizer, cfg.work_dir,
                    cfg.log_level)

    runner.register_training_hooks(cfg.lr_config, cfg.optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    data_loaders = [
        build_dataloader(
            dataset,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            dist=True)
    ]
    # put model on gpus
    model = MMDistributedDataParallel(model.cuda())
    # build runner
    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(model, batch_processor, optimizer, cfg.work_dir,
                    cfg.log_level)
    # register hooks
    optimizer_config = DistOptimizerHook(**cfg.optimizer_config)
    runner.register_training_hooks(cfg.lr_config, optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)
    runner.register_hook(DistSamplerSeedHook())
    # register eval hooks
    if validate:
        val_dataset_cfg = cfg.data.val
        if isinstance(model.module, RPN):
            # TODO: implement recall hooks for other datasets
            runner.register_hook(CocoDistEvalRecallHook(val_dataset_cfg))
        else:
            dataset_type = getattr(datasets, val_dataset_cfg.type)
            if issubclass(dataset_type, datasets.CocoDataset):
                runner.register_hook(CocoDistEvalmAPHook(val_dataset_cfg))
            else:
                runner.register_hook(DistEvalmAPHook(val_dataset_cfg))

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _non_dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    dataset = dataset if isinstance(dataset, (list, tuple)) else [dataset]
    data_loaders = [
        build_dataloader(
            ds,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            cfg.gpus,
            dist=False) for ds in dataset
    ]
    # put model on gpus
    model = MMDataParallel(model, device_ids=range(cfg.gpus)).cuda()

    # build runner
    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(model, batch_processor, optimizer, cfg.work_dir,
                    cfg.log_level)
    # fp16 setting
    fp16_cfg = cfg.get('fp16', None)
    if fp16_cfg is not None:
        optimizer_config = Fp16OptimizerHook(
            **cfg.optimizer_config, **fp16_cfg, distributed=False)
    else:
        optimizer_config = cfg.optimizer_config
    runner.register_training_hooks(cfg.lr_config, optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _non_dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    dataset = dataset if isinstance(dataset, (list, tuple)) else [dataset]
    data_loaders = [
        build_dataloader(
            ds,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            cfg.gpus,
            dist=False) for ds in dataset
    ]
    # put model on gpus
    model = MMDataParallel(model, device_ids=range(cfg.gpus)).cuda()

    # build runner
    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(model, batch_processor, optimizer, cfg.work_dir,
                    cfg.log_level)
    # fp16 setting
    fp16_cfg = cfg.get('fp16', None)
    if fp16_cfg is not None:
        optimizer_config = Fp16OptimizerHook(
            **cfg.optimizer_config, **fp16_cfg, distributed=False)
    else:
        optimizer_config = cfg.optimizer_config
    runner.register_training_hooks(cfg.lr_config, optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    data_loaders = [
        build_dataloader(
            dataset,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            dist=True)
    ]
    # put model on gpus
    model = MMDistributedDataParallel(model.cuda())
    # build runner
    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(model, batch_processor, optimizer, cfg.work_dir,
                    cfg.log_level)
    # register hooks
    optimizer_config = DistOptimizerHook(**cfg.optimizer_config)
    runner.register_training_hooks(cfg.lr_config, optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)
    runner.register_hook(DistSamplerSeedHook())
    # register eval hooks
    if validate:
        val_dataset_cfg = cfg.data.val
        if isinstance(model.module, RPN):
            # TODO: implement recall hooks for other datasets
            runner.register_hook(CocoDistEvalRecallHook(val_dataset_cfg))
        else:
            dataset_type = getattr(datasets, val_dataset_cfg.type)
            if issubclass(dataset_type, datasets.CocoDataset):
                runner.register_hook(CocoDistEvalmAPHook(val_dataset_cfg))
            else:
                runner.register_hook(DistEvalmAPHook(val_dataset_cfg))

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _non_dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    dataset = dataset if isinstance(dataset, (list, tuple)) else [dataset]
    data_loaders = [
        build_dataloader(
            ds,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            cfg.gpus,
            dist=False) for ds in dataset
    ]
    # put model on gpus
    model = MMDataParallel(model, device_ids=range(cfg.gpus)).cuda()

    # build runner
    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(model, batch_processor, optimizer, cfg.work_dir,
                    cfg.log_level)
    # fp16 setting
    fp16_cfg = cfg.get('fp16', None)
    if fp16_cfg is not None:
        optimizer_config = Fp16OptimizerHook(
            **cfg.optimizer_config, **fp16_cfg, distributed=False)
    else:
        optimizer_config = cfg.optimizer_config
    runner.register_training_hooks(cfg.lr_config, optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)
    #####
    runner.register_hook(CheckpointHook(interval=500))

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _non_dist_train(model, dataset, cfg, validate=False):
    if validate:
        raise NotImplementedError('Built-in validation is not implemented '
                                  'yet in not-distributed training. Use '
                                  'distributed training or test.py and '
                                  '*eval.py scripts instead.')
    # prepare data loaders
    dataset = dataset if isinstance(dataset, (list, tuple)) else [dataset]
    data_loaders = [
        build_dataloader(
            ds,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            cfg.gpus,
            dist=False) for ds in dataset
    ]
    # put model on gpus
    model = MMDataParallel(model, device_ids=range(cfg.gpus)).cuda()

    # build runner
    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(model, batch_processor, optimizer, cfg.work_dir,
                    cfg.log_level)
    # fp16 setting
    fp16_cfg = cfg.get('fp16', None)
    if fp16_cfg is not None:
        optimizer_config = Fp16OptimizerHook(
            **cfg.optimizer_config, **fp16_cfg, distributed=False)
    else:
        optimizer_config = cfg.optimizer_config
    runner.register_training_hooks(cfg.lr_config, optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    data_loaders = [
        build_dataloader(
            dataset,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            dist=True)
    ]
    # put model on gpus
    model = MMDistributedDataParallel(model.cuda())
    # build runner
    runner = Runner(model, batch_processor, cfg.optimizer, cfg.work_dir,
                    cfg.log_level)
    # register hooks
    optimizer_config = DistOptimizerHook(**cfg.optimizer_config)
    runner.register_training_hooks(cfg.lr_config, optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)
    runner.register_hook(DistSamplerSeedHook())
    # register eval hooks
    if validate:
        if isinstance(model.module, RPN):
            # TODO: implement recall hooks for other datasets
            runner.register_hook(CocoDistEvalRecallHook(cfg.data.val))
        else:
            if cfg.data.val.type == 'CocoDataset':
                runner.register_hook(CocoDistEvalmAPHook(cfg.data.val))
            else:
                runner.register_hook(DistEvalmAPHook(cfg.data.val))

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _non_dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    data_loaders = [
        build_dataloader(
            dataset,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            cfg.gpus,
            dist=False)
    ]
    # put model on gpus
    model = MMDataParallel(model, device_ids=range(cfg.gpus)).cuda()

    # build runner
    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(model, batch_processor, optimizer, cfg.work_dir,
                    cfg.log_level)
    # fp16 setting
    fp16_cfg = cfg.get('fp16', None)
    if fp16_cfg is not None:
        optimizer_config = Fp16OptimizerHook(
            **cfg.optimizer_config, **fp16_cfg, distributed=False)
    else:
        optimizer_config = cfg.optimizer_config
    runner.register_training_hooks(cfg.lr_config, optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _non_dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    dataset = dataset if isinstance(dataset, (list, tuple)) else [dataset]
    data_loaders = [
        build_dataloader(
            ds,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            cfg.gpus,
            dist=False) for ds in dataset
    ]
    # put model on gpus
    model = MMDataParallel(model, device_ids=range(cfg.gpus)).cuda()

    # build runner
    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(model, batch_processor, optimizer, cfg.work_dir,
                    cfg.log_level)
    # fp16 setting
    fp16_cfg = cfg.get('fp16', None)
    if fp16_cfg is not None:
        optimizer_config = Fp16OptimizerHook(
            **cfg.optimizer_config, **fp16_cfg, distributed=False)
    else:
        optimizer_config = cfg.optimizer_config
    runner.register_training_hooks(cfg.lr_config, optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _non_dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    dataset = dataset if isinstance(dataset, (list, tuple)) else [dataset]
    data_loaders = [
        build_dataloader(
            ds,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            cfg.gpus,
            dist=False) for ds in dataset
    ]
    # put model on gpus
    model = MMDataParallel(model, device_ids=range(cfg.gpus)).cuda()

    # build runner
    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(model, batch_processor, optimizer, cfg.work_dir,
                    cfg.log_level)
    # fp16 setting
    fp16_cfg = cfg.get('fp16', None)
    if fp16_cfg is not None:
        optimizer_config = Fp16OptimizerHook(
            **cfg.optimizer_config, **fp16_cfg, distributed=False)
    else:
        optimizer_config = cfg.optimizer_config
    runner.register_training_hooks(cfg.lr_config, optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _non_dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    dataset = dataset if isinstance(dataset, (list, tuple)) else [dataset]
    data_loaders = [
        build_dataloader(
            ds,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            cfg.gpus,
            dist=False) for ds in dataset
    ]
    # put model on gpus
    model = MMDataParallel(model, device_ids=range(cfg.gpus)).cuda()

    # build runner
    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(model, batch_processor, optimizer, cfg.work_dir,
                    cfg.log_level)
    # fp16 setting
    fp16_cfg = cfg.get('fp16', None)
    if fp16_cfg is not None:
        optimizer_config = Fp16OptimizerHook(
            **cfg.optimizer_config, **fp16_cfg, distributed=False)
    else:
        optimizer_config = cfg.optimizer_config
    runner.register_training_hooks(cfg.lr_config, optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs) 

def _dist_train(model, dataset, cfg, validate=False):
    # prepare data loaders
    data_loaders = [
        build_dataloader(
            dataset,
            cfg.data.imgs_per_gpu,
            cfg.data.workers_per_gpu,
            dist=True)
    ]
    # put model on gpus
    model = MMDistributedDataParallel(model.cuda())
    # build runner
    optimizer = build_optimizer(model, cfg.optimizer)
    runner = Runner(model, batch_processor, optimizer, cfg.work_dir,
                    cfg.log_level)
    # register hooks
    optimizer_config = DistOptimizerHook(**cfg.optimizer_config)
    runner.register_training_hooks(cfg.lr_config, optimizer_config,
                                   cfg.checkpoint_config, cfg.log_config)
    runner.register_hook(DistSamplerSeedHook())
    # register eval hooks
    if validate:
        val_dataset_cfg = cfg.data.val
        if isinstance(model.module, RPN):
            # TODO: implement recall hooks for other datasets
            runner.register_hook(CocoDistEvalRecallHook(val_dataset_cfg))
        else:
            dataset_type = getattr(datasets, val_dataset_cfg.type)
            if issubclass(dataset_type, datasets.CocoDataset):
                runner.register_hook(CocoDistEvalmAPHook(val_dataset_cfg))
            else:
                runner.register_hook(DistEvalmAPHook(val_dataset_cfg))

    if cfg.resume_from:
        runner.resume(cfg.resume_from)
    elif cfg.load_from:
        runner.load_checkpoint(cfg.load_from)
    runner.run(data_loaders, cfg.workflow, cfg.total_epochs)