train_net.py

# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved
import os

from detectron2.checkpoint import DetectionCheckpointer
from detectron2.config import get_cfg
from detectron2.engine import default_argument_parser, default_setup, launch
from detectron2.evaluation import COCOEvaluator
from detectron2.evaluation.testing import verify_results
from detectron2.utils import comm

from configs import add_centernet_config
from engine import CenterTrainer

import warnings
warnings.filterwarnings("ignore")

class Trainer(CenterTrainer):
    @classmethod
    def build_evaluator(cls, cfg, dataset_name, output_folder=None):
        if output_folder is None:
            output_folder = os.path.join(cfg.OUTPUT_DIR, "inference")
        return COCOEvaluator(dataset_name, cfg, True, output_folder)

def setup(args):
    """
    Create configs and perform basic setups.
    """
    cfg = get_cfg()
    add_centernet_config(cfg)
    cfg.merge_from_file(args.config_file)
    cfg.merge_from_list(args.opts)
    cfg.freeze()
    default_setup(cfg, args)
    return cfg

def main(args):
    cfg = setup(args)

    if args.eval_only:
        model = Trainer.build_model(cfg)
        DetectionCheckpointer(model, save_dir=cfg.OUTPUT_DIR).resume_or_load(
            cfg.MODEL.WEIGHTS, resume=args.resume
        )
        res = Trainer.test(cfg, model)
        if comm.is_main_process():
            verify_results(cfg, res)
        return res
    """
    If you'd like to do anything fancier than the standard training logic,
    consider writing your own training loop or subclassing the trainer.
    """
    trainer = Trainer(cfg)
    trainer.resume_or_load(resume=args.resume)

    return trainer.train()

if __name__ == "__main__":
    args = default_argument_parser().parse_args()
    print("Command Line Args:", args)
    launch(
        main,
        args.num_gpus,
        num_machines=args.num_machines,
        machine_rank=args.machine_rank,
        dist_url=args.dist_url,
        args=(args,),
    )