def register_lvis_instances(name, metadata, json_file, image_root):
"""
Register a dataset in LVIS's json annotation format for instance detection.
Args:
name (str): a name that identifies the dataset, e.g. "lvis_v0.5_train".
metadata (dict): extra metadata associated with this dataset. It can be an empty dict.
json_file (str): path to the json instance annotation file.
image_root (str): directory which contains all the images.
"""
DatasetCatalog.register(
name, lambda: load_lvis_json(json_file, image_root, name))
MetadataCatalog.get(name).set(json_file=json_file,
image_root=image_root,
evaluator_type="lvis",
**metadata)
def build_evaluator(cls, cfg, dataset_name, output_folder=None):
"""
Create evaluator(s) for a given dataset.
This uses the special metadata "evaluator_type" associated with each builtin dataset.
For your own dataset, you can simply create an evaluator manually in your
script and do not have to worry about the hacky if-else logic here.
"""
if output_folder is None:
output_folder = os.path.join(cfg.OUTPUT_DIR, "inference")
evaluator_list = []
evaluator_type = MetadataCatalog.get(dataset_name).evaluator_type
if evaluator_type == "coco":
evaluator_list.append(
COCOEvaluator(dataset_name, cfg, True, output_folder))
if evaluator_type == "pascal_voc":
return PascalVOCDetectionEvaluator(dataset_name)
if evaluator_type == "lvis":
return LVISEvaluator(dataset_name, cfg, True, output_folder)
if len(evaluator_list) == 0:
raise NotImplementedError(
"no Evaluator for the dataset {} with the type {}".format(
dataset_name, evaluator_type))
if len(evaluator_list) == 1:
return evaluator_list[0]
return DatasetEvaluators(evaluator_list)
def __init__(self, dataset_name, cfg, distributed, output_dir=None):
"""
Args:
dataset_name (str): name of the dataset to be evaluated.
It must have the following corresponding metadata:
"json_file": the path to the LVIS format annotation
cfg (CfgNode): config instance
distributed (True): if True, will collect results from all ranks for evaluation.
Otherwise, will evaluate the results in the current process.
output_dir (str): optional, an output directory to dump results.
"""
from lvis import LVIS
self._distributed = distributed
self._output_dir = output_dir
self._cpu_device = torch.device("cpu")
self._logger = logging.getLogger(__name__)
self._metadata = MetadataCatalog.get(dataset_name)
json_file = PathManager.get_local_path(self._metadata.json_file)
self._lvis_api = LVIS(json_file)
# Test set json files do not contain annotations (evaluation must be
# performed using the LVIS evaluation server).
self._do_evaluation = len(self._lvis_api.get_ann_ids()) > 0
def register_meta_pascal_voc(
name, metadata, dirname, split, year, keepclasses, sid):
if keepclasses.startswith('base_novel'):
thing_classes = metadata["thing_classes"][sid]
elif keepclasses.startswith('base'):
thing_classes = metadata["base_classes"][sid]
elif keepclasses.startswith('novel'):
thing_classes = metadata["novel_classes"][sid]
DatasetCatalog.register(
name, lambda: load_filtered_voc_instances(
name, dirname, split, thing_classes)
)
MetadataCatalog.get(name).set(
thing_classes=thing_classes, dirname=dirname, year=year, split=split,
base_classes=metadata["base_classes"][sid],
novel_classes=metadata["novel_classes"][sid]
)
def register_meta_coco(name, metadata, imgdir, annofile):
DatasetCatalog.register(
name,
lambda: load_coco_json(annofile, imgdir, metadata, name),
)
if '_base' in name or '_novel' in name:
split = 'base' if '_base' in name else 'novel'
metadata['thing_dataset_id_to_contiguous_id'] = \
metadata['{}_dataset_id_to_contiguous_id'.format(split)]
metadata['thing_classes'] = metadata['{}_classes'.format(split)]
MetadataCatalog.get(name).set(
json_file=annofile,
image_root=imgdir,
evaluator_type="coco",
dirname="datasets/coco",
**metadata,
)
def register_coco_instances(name, metadata, json_file, image_root):
"""
Register a dataset in COCO's json annotation format for instance detection.
This is an example of how to register a new dataset.
You can do something similar to this function, to register new datasets.
Args:
name (str): the name that identifies a dataset, e.g. "coco_2014_train".
metadata (dict): extra metadata associated with this dataset. You can
leave it as an empty dict.
json_file (str): path to the json instance annotation file.
image_root (str): directory which contains all the images.
"""
# 1. register a function which returns dicts
DatasetCatalog.register(name, lambda: load_coco_json(json_file, image_root, name))
# 2. Optionally, add metadata about this dataset,
# since they might be useful in evaluation, visualization or logging
MetadataCatalog.get(name).set(
json_file=json_file, image_root=image_root, evaluator_type="coco", **metadata
)
def __init__(self, cfg):
self.cfg = cfg.clone() # cfg can be modified by model
self.model = build_model(self.cfg)
self.model.eval()
self.metadata = MetadataCatalog.get(cfg.DATASETS.TEST[0])
checkpointer = DetectionCheckpointer(self.model)
checkpointer.load(cfg.MODEL.WEIGHTS)
self.transform_gen = T.ResizeShortestEdge(
[cfg.INPUT.MIN_SIZE_TEST, cfg.INPUT.MIN_SIZE_TEST],
cfg.INPUT.MAX_SIZE_TEST)
self.input_format = cfg.INPUT.FORMAT
assert self.input_format in ["RGB", "BGR"], self.input_format
def __init__(self, dataset_name):
"""
Args:
dataset_name (str): name of the dataset, e.g., "voc_2007_test"
"""
self._dataset_name = dataset_name
meta = MetadataCatalog.get(dataset_name)
self._anno_file_template = os.path.join(meta.dirname, "Annotations",
"{}.xml")
self._image_set_path = os.path.join(meta.dirname, "ImageSets", "Main",
meta.split + ".txt")
self._class_names = meta.thing_classes
# add this two terms for calculating the mAP of different subset
self._base_classes = meta.base_classes
self._novel_classes = meta.novel_classes
assert meta.year in [2007, 2012], meta.year
self._is_2007 = meta.year == 2007
self._cpu_device = torch.device("cpu")
self._logger = logging.getLogger(__name__)
return dataset_dicts
if __name__ == "__main__":
"""
Test the LVIS json dataset loader.
Usage:
python -m fs3c.data.datasets.lvis \
path/to/json path/to/image_root dataset_name vis_limit
"""
import sys
import numpy as np
from fs3c.utils.logger import setup_logger
from PIL import Image
from fs3c.utils.visualizer import Visualizer
logger = setup_logger(name=__name__)
meta = MetadataCatalog.get(sys.argv[3])
dicts = load_lvis_json(sys.argv[1], sys.argv[2], sys.argv[3])
logger.info("Done loading {} samples.".format(len(dicts)))
dirname = "lvis-data-vis"
os.makedirs(dirname, exist_ok=True)
for d in dicts[:int(sys.argv[4])]:
img = np.array(Image.open(d["file_name"]))
visualizer = Visualizer(img, metadata=meta)
vis = visualizer.draw_dataset_dict(d)
fpath = os.path.join(dirname, os.path.basename(d["file_name"]))
vis.save(fpath)
parser.add_argument("--output", required=True, help="output directory")
parser.add_argument("--dataset", help="name of the dataset", default="coco_2017_val")
parser.add_argument("--conf-threshold", default=0.5, type=float, help="confidence threshold")
args = parser.parse_args()
logger = setup_logger()
with PathManager.open(args.input, "r") as f:
predictions = json.load(f)
pred_by_image = defaultdict(list)
for p in predictions:
pred_by_image[p["image_id"]].append(p)
dicts = list(DatasetCatalog.get(args.dataset))
metadata = MetadataCatalog.get(args.dataset)
if hasattr(metadata, "thing_dataset_id_to_contiguous_id"):
def dataset_id_map(ds_id):
return metadata.thing_dataset_id_to_contiguous_id[ds_id]
elif "lvis" in args.dataset:
# LVIS results are in the same format as COCO results, but have a different
# mapping from dataset category id to contiguous category id in [0, #categories - 1]
def dataset_id_map(ds_id):
return ds_id - 1
else:
raise ValueError("Unsupported dataset: {}".format(args.dataset))
os.makedirs(args.output, exist_ok=True)
def register_pascal_voc(name, dirname, split, year):
DatasetCatalog.register(name, lambda: load_voc_instances(dirname, split))
MetadataCatalog.get(name).set(thing_classes=CLASS_NAMES,
dirname=dirname,
year=year,
split=split)
def __init__(self, dataset_name, cfg, distributed, output_dir=None):
"""
Args:
dataset_name (str): name of the dataset to be evaluated.
It must have either the following corresponding metadata:
"json_file": the path to the COCO format annotation
Or it must be in detectron2's standard dataset format
so it can be converted to COCO format automatically.
cfg (CfgNode): config instance
distributed (True):
if True, will collect results from all ranks for evaluation.
Otherwise, will evaluate the results in the current process.
output_dir (str): optional, an output directory to dump results.
"""
self._distributed = distributed
self._output_dir = output_dir
self._dataset_name = dataset_name
self._cpu_device = torch.device("cpu")
self._logger = logging.getLogger(__name__)
self._metadata = MetadataCatalog.get(dataset_name)
if not hasattr(self._metadata, "json_file"):
self._logger.warning(
f"json_file was not found in MetaDataCatalog for '{dataset_name}'"
)
cache_path = convert_to_coco_json(dataset_name, output_dir)
self._metadata.json_file = cache_path
self._is_splits = "all" in dataset_name or "base" in dataset_name \
or "novel" in dataset_name
self._base_classes = [
8,
10,
11,
13,
14,
15,
22,
23,
24,
25,
27,
28,
31,
32,
33,
34,
35,
36,
37,
38,
39,
40,
41,
42,
43,
46,
47,
48,
49,
50,
51,
52,
53,
54,
55,
56,
57,
58,
59,
60,
61,
65,
70,
73,
74,
75,
76,
77,
78,
79,
80,
81,
82,
84,
85,
86,
87,
88,
89,
90,
]
self._novel_classes = [
1, 2, 3, 4, 5, 6, 7, 9, 16, 17, 18, 19, 20, 21, 44, 62, 63, 64, 67,
72
]
json_file = PathManager.get_local_path(self._metadata.json_file)
with contextlib.redirect_stdout(io.StringIO()):
self._coco_api = COCO(json_file)
# Test set json files do not contain annotations (evaluation must be
# performed using the COCO evaluation server).
self._do_evaluation = "annotations" in self._coco_api.dataset
def register_all_pascal_voc(root="datasets"):
SPLITS = [
("voc_2007_trainval", "VOC2007", "trainval"),
("voc_2007_train", "VOC2007", "train"),
("voc_2007_val", "VOC2007", "val"),
("voc_2007_test", "VOC2007", "test"),
("voc_2012_trainval", "VOC2012", "trainval"),
("voc_2012_train", "VOC2012", "train"),
("voc_2012_val", "VOC2012", "val"),
]
for name, dirname, split in SPLITS:
year = 2007 if "2007" in name else 2012
register_pascal_voc(name, os.path.join(root, dirname), split, year)
MetadataCatalog.get(name).evaluator_type = "pascal_voc"
# register meta datasets
METASPLITS = [
("voc_2007_trainval_base1", "VOC2007", "trainval", "base1", 1),
("voc_2007_trainval_base2", "VOC2007", "trainval", "base2", 2),
("voc_2007_trainval_base3", "VOC2007", "trainval", "base3", 3),
("voc_2012_trainval_base1", "VOC2012", "trainval", "base1", 1),
("voc_2012_trainval_base2", "VOC2012", "trainval", "base2", 2),
("voc_2012_trainval_base3", "VOC2012", "trainval", "base3", 3),
("voc_2007_trainval_all1", "VOC2007", "trainval", "base_novel_1", 1),
("voc_2007_trainval_all2", "VOC2007", "trainval", "base_novel_2", 2),
("voc_2007_trainval_all3", "VOC2007", "trainval", "base_novel_3", 3),
("voc_2012_trainval_all1", "VOC2012", "trainval", "base_novel_1", 1),
("voc_2012_trainval_all2", "VOC2012", "trainval", "base_novel_2", 2),
("voc_2012_trainval_all3", "VOC2012", "trainval", "base_novel_3", 3),
("voc_2007_test_base1", "VOC2007", "test", "base1", 1),
("voc_2007_test_base2", "VOC2007", "test", "base2", 2),
("voc_2007_test_base3", "VOC2007", "test", "base3", 3),
("voc_2007_test_novel1", "VOC2007", "test", "novel1", 1),
("voc_2007_test_novel2", "VOC2007", "test", "novel2", 2),
("voc_2007_test_novel3", "VOC2007", "test", "novel3", 3),
("voc_2007_test_all1", "VOC2007", "test", "base_novel_1", 1),
("voc_2007_test_all2", "VOC2007", "test", "base_novel_2", 2),
("voc_2007_test_all3", "VOC2007", "test", "base_novel_3", 3),
]
# register small meta datasets for fine-tuning stage
for prefix in ["all", "novel"]:
for sid in range(1, 4):
for shot in [1, 2, 3, 5, 10]:
for year in [2007, 2012]:
for seed in range(100):
seed = '' if seed == 0 else '_seed{}'.format(seed)
name = "voc_{}_trainval_{}{}_{}shot{}".format(
year, prefix, sid, shot, seed)
dirname = "VOC{}".format(year)
img_file = "{}_{}shot_split_{}_trainval".format(
prefix, shot, sid)
keepclasses = "base_novel_{}".format(sid) \
if prefix == 'all' else "novel{}".format(sid)
METASPLITS.append(
(name, dirname, img_file, keepclasses, sid))
for name, dirname, split, keepclasses, sid in METASPLITS:
year = 2007 if "2007" in name else 2012
register_meta_pascal_voc(name,
_get_builtin_metadata("pascal_voc_fewshot"),
os.path.join(root, dirname), split, year,
keepclasses, sid)
MetadataCatalog.get(name).evaluator_type = "pascal_voc"
help="Modify config options using the command-line",
default=None,
nargs=argparse.REMAINDER,
)
return parser.parse_args(in_args)
if __name__ == "__main__":
args = parse_args()
logger = setup_logger()
logger.info("Arguments: " + str(args))
cfg = setup(args)
dirname = args.output_dir
os.makedirs(dirname, exist_ok=True)
metadata = MetadataCatalog.get(cfg.DATASETS.TRAIN[0])
def output(vis, fname):
if args.show:
print(fname)
cv2.imshow("window", vis.get_image()[:, :, ::-1])
cv2.waitKey()
else:
filepath = os.path.join(dirname, fname)
print("Saving to {} ...".format(filepath))
vis.save(filepath)
scale = 2.0 if args.show else 1.0
if args.source == "dataloader":
train_data_loader = build_detection_train_loader(cfg)
for batch in train_data_loader: