def register_voc_data(args):
register_pascal_voc("pascal_trainval_2007",
args.DATASETS.CLASSIFIER_DATAROOT + 'VOC2007/',
'trainval', 2007)
register_pascal_voc("pascal_trainval_2012",
args.DATASETS.CLASSIFIER_DATAROOT + 'VOC2012/',
'trainval', 2012)
register_pascal_voc("pascal_test_2007",
args.DATASETS.CLASSIFIER_DATAROOT + 'VOC2007/', 'test',
2007)
from detectron2.data import MetadataCatalog, DatasetCatalog
from detectron2.engine import DefaultTrainer, default_argument_parser, default_setup, hooks, launch
from detectron2.evaluation import (
DatasetEvaluators,
PascalVOCDetectionEvaluator,
verify_results,
)
from detectron2.modeling import GeneralizedRCNNWithTTA
from detectron2.data.datasets.pascal_voc import register_pascal_voc
_datasets_root = "datasets"
for d in ["trainval", "test"]:
register_pascal_voc(name=f'100DOH_hand_{d}',
dirname=_datasets_root,
split=d,
year=2007,
class_names=["hand"])
MetadataCatalog.get(f'100DOH_hand_{d}').set(evaluator_type='pascal_voc')
class Trainer(DefaultTrainer):
"""
We use the "DefaultTrainer" which contains pre-defined default logic for
standard training workflow. They may not work for you, especially if you
are working on a new research project. In that case you can use the cleaner
"SimpleTrainer", or write your own training loop. You can use
"tools/plain_train_net.py" as an example.
"""
@classmethod
def build_evaluator(cls, cfg, dataset_name, output_folder=None):
COCOPanopticEvaluator,
DatasetEvaluators,
LVISEvaluator,
PascalVOCDetectionEvaluator,
SemSegEvaluator,
verify_results,
)
from detectron2.modeling import GeneralizedRCNNWithTTA
# more import
from detectron2.data.datasets.pascal_voc import register_pascal_voc
_datasets_root = "datasets"
for d in ["train", "val"]:
register_pascal_voc(name=f'100DOH_hand_{d}',
dirname=_datasets_root,
split=d,
year=2007)
MetadataCatalog.get(f'100DOH_hand_{d}').set(evaluator_type='pascal_voc')
class Trainer(DefaultTrainer):
"""
We use the "DefaultTrainer" which contains pre-defined default logic for
standard training workflow. They may not work for you, especially if you
are working on a new research project. In that case you can use the cleaner
"SimpleTrainer", or write your own training loop. You can use
"tools/plain_train_net.py" as an example.
"""
@classmethod
def build_evaluator(cls, cfg, dataset_name, output_folder=None):
"""