def kitti_eval(det_results, dataset, iou_thr=0.5):
gt_bboxes = []
gt_labels = []
gt_ignore = []
for i in range(len(dataset)):
ann = dataset.get_ann_info(i)
bboxes = ann['bboxes']
labels = ann['labels']
gt_bboxes.append(bboxes)
gt_labels.append(labels)
# if i>10:
# break
if not gt_ignore:
gt_ignore = None
dataset_name = 'kitti'
eval_map(det_results,
gt_bboxes,
gt_labels,
gt_ignore=gt_ignore,
scale_ranges=None,
iou_thr=iou_thr,
dataset=dataset_name,
print_summary=True)
def voc_eval(result_file, dataset, iou_thr=0.5):
det_results = mmcv.load(result_file)
gt_bboxes = []
gt_labels = []
gt_ignore = []
for i in range(len(dataset)):
ann = dataset.get_ann_info(i)
bboxes = ann['bboxes']
labels = ann['labels']
#if 'bboxes_ignore' in ann:
#ignore = np.concatenate([
# np.zeros(bboxes.shape[0], dtype=np.bool),
# np.ones(ann['bboxes_ignore'].shape[0], dtype=np.bool)
#])
#gt_ignore.append(ignore)
#bboxes = np.vstack([bboxes, ann['bboxes_ignore']])
#print(ann)
#labels = np.concatenate([labels, ann['labels_ignore']])
gt_bboxes.append(bboxes)
gt_labels.append(labels)
if not gt_ignore:
gt_ignore = None
eval_map(det_results,
gt_bboxes,
gt_labels,
gt_ignore=gt_ignore,
scale_ranges=None,
iou_thr=iou_thr,
dataset=dataset.CLASSES,
print_summary=True)
def voc_eval(result_file, dataset, iou_thr=0.5):
det_results = mmcv.load(result_file)
gt_bboxes = []
gt_labels = []
gt_ignore = []
for i in range(len(dataset)):
ann = dataset.get_ann_info(i)
bboxes = ann['bboxes']
labels = ann['labels']
if 'bboxes_ignore' in ann:
print("!")
ignore = np.concatenate([
np.zeros(bboxes.shape[0], dtype=np.bool),
np.ones(ann['bboxes_ignore'].shape[0], dtype=np.bool)
])
gt_ignore.append(ignore)
bboxes = np.vstack([bboxes, ann['bboxes_ignore']])
labels = np.concatenate([labels, ann['labels_ignore']])
gt_bboxes.append(bboxes)
gt_labels.append(labels)
if not gt_ignore:
gt_ignore = None
if hasattr(dataset, 'year') and dataset.year == 2007:
dataset_name = 'voc07'
else:
dataset_name = dataset.CLASSES
eval_map(det_results,
gt_bboxes,
gt_labels,
gt_ignore=gt_ignore,
scale_ranges=None,
iou_thr=iou_thr,
dataset=dataset_name,
print_summary=True)
def voc_eval(result_file, dataset, iou_thr=0.5, nproc=4):
det_results = mmcv.load(result_file)
annotations = [dataset.get_ann_info(i) for i in range(len(dataset))]
if hasattr(dataset, 'year') and dataset.year == 2007:
dataset_name = 'voc07'
else:
dataset_name = dataset.CLASSES
eval_map(det_results, annotations, scale_ranges=None, iou_thr=iou_thr, dataset=dataset_name)
def voc_eval(result_file, dataset, iou_thr=0.5):
det_results = mmcv.load(result_file)
gt_bboxes = []
gt_labels = []
gt_ignore = []
for i in range(len(dataset)):
ann = dataset.get_ann_info(i)
bboxes = ann['bboxes']
labels = ann['labels']
if 'bboxes_ignore' in ann:
ignore = np.concatenate([
np.zeros(bboxes.shape[0], dtype=np.bool),
np.ones(ann['bboxes_ignore'].shape[0], dtype=np.bool)
])
gt_ignore.append(ignore)
bboxes = np.vstack([bboxes, ann['bboxes_ignore']])
labels = np.concatenate([labels, ann['labels_ignore']])
gt_bboxes.append(bboxes)
gt_labels.append(labels)
if not gt_ignore:
gt_ignore = gt_ignore
if hasattr(dataset, 'year') and dataset.year == 2007:
dataset_name = 'voc07'
else:
dataset_name = dataset.CLASSES
if iou_thr == -1:
print("Start COCO-style evaluation")
mean_ap_list = {}
for iou_thr in np.arange(0.5, 1.0, 0.05):
mean_ap, eval_results = eval_map(det_results,
gt_bboxes,
gt_labels,
gt_ignore=gt_ignore,
scale_ranges=None,
iou_thr=iou_thr,
dataset=dataset_name,
print_summary=False)
mean_ap_list[iou_thr] = mean_ap
mean_ap = np.mean(list(mean_ap_list.values()))
mean_ap_list[-1] = mean_ap
print("Path: {} mAP: {}".format(result_file, mean_ap_list))
return mean_ap_list
else:
mean_ap, eval_results = eval_map(det_results,
gt_bboxes,
gt_labels,
gt_ignore=gt_ignore,
scale_ranges=None,
iou_thr=iou_thr,
dataset=dataset_name,
print_summary=True)
print("mAP: {}".format({iou_thr: mean_ap}))
return {iou_thr: mean_ap}
def voc_eval(result_file, dataset, iou_thr=0.5):
det_results = mmcv.load(result_file)
gt_bboxes = []
gt_labels = []
gt_ignore = []
for i in range(len(dataset)):
ann = dataset.get_ann_info(i)
bboxes = ann['bboxes']
labels = ann['labels']
if 'bboxes_ignore' in ann:
ignore = np.concatenate([
np.zeros(bboxes.shape[0], dtype=np.bool),
np.ones(ann['bboxes_ignore'].shape[0], dtype=np.bool)
])
gt_ignore.append(ignore)
bboxes = np.vstack([bboxes, ann['bboxes_ignore']])
labels = np.concatenate([labels, ann['labels_ignore']])
gt_bboxes.append(bboxes)
gt_labels.append(labels)
if not gt_ignore:
gt_ignore = gt_ignore
if hasattr(dataset, 'year') and dataset.year == 2007:
dataset_name = 'voc07'
else:
dataset_name = dataset.CLASSES
eval_map(
det_results,
gt_bboxes,
gt_labels,
gt_ignore=gt_ignore,
scale_ranges=None,
iou_thr=iou_thr,
dataset=dataset_name,
print_summary=True)
# added by WSK
# this is added to compute the C0C0-style AP
iou_thr_list = [0.5, 0.55, 0.60, 0.65, 0.70, 0.75, 0.80, 0.85, 0.90, 0.95]
mean_ap_list = []
ap_COCO = 0
for thr in iou_thr_list:
mean_ap, eval_results = eval_map(
det_results,
gt_bboxes,
gt_labels,
gt_ignore=gt_ignore,
scale_ranges=None,
iou_thr=thr,
dataset=dataset_name,
print_summary=True)
ap_COCO = ap_COCO + mean_ap*0.1
mean_ap_list.append(mean_ap)
print('the ap_COCO is ', ap_COCO)
print('the mean_ap_list is ', mean_ap_list)
def voc_eval(result_file, dataset, cfg, iou_thr=0.5):
det_results = mmcv.load(result_file)
gt_bboxes = []
gt_labels = []
gt_ignore = []
for i in range(len(dataset)):
ann = dataset.get_ann_info(i)
#print(ann.keys())
print(ann['filename'])
filename = ann['filename']
bboxes = ann['bboxes']
labels = ann['labels']
#print("bboxes----")
#print(bboxes)
if 'bboxes_ignore' in ann:
ignore = np.concatenate([
np.zeros(bboxes.shape[0], dtype=np.bool),
np.ones(ann['bboxes_ignore'].shape[0], dtype=np.bool)
])
gt_ignore.append(ignore)
bboxes = np.vstack([bboxes, ann['bboxes_ignore']])
labels = np.concatenate([labels, ann['labels_ignore']])
gt_bboxes.append(bboxes)
gt_labels.append(labels)
# resize back
for class_id in range(len(det_results[i])):
det_bboxes = det_results[i][class_id]
det_bboxes[:, 0] = det_bboxes[:, 0] / cfg.input_size * ann['width']
det_bboxes[:,
1] = det_bboxes[:, 1] / cfg.input_size * ann['height']
det_bboxes[:, 2] = det_bboxes[:, 2] / cfg.input_size * ann['width']
det_bboxes[:,
3] = det_bboxes[:, 3] / cfg.input_size * ann['height']
det_results[i][class_id] = det_bboxes
if not gt_ignore:
gt_ignore = gt_ignore
gt_ignore = None
#print(len(gt_ignore))
print("*************************************")
#print(det_results)
dataset_name = 'aicourt'
eval_map(det_results,
gt_bboxes,
gt_labels,
gt_ignore=gt_ignore,
scale_ranges=None,
iou_thr=iou_thr,
dataset=dataset_name,
print_summary=True)
def eval(result_file, dataset, iou_thr=0.5, dataset_name='voc', split='16_4'):
det_results = mmcv.load(result_file)
gt_bboxes = []
gt_labels = []
gt_ignore = []
for i in range(len(dataset)):
ann = dataset.get_ann_info(i)
bboxes = ann['bboxes']
labels = ann['labels']
if 'bboxes_ignore' in ann:
ignore = np.concatenate([
np.zeros(bboxes.shape[0], dtype=np.bool),
np.ones(ann['bboxes_ignore'].shape[0], dtype=np.bool)
])
gt_ignore.append(ignore)
bboxes = np.vstack([bboxes, ann['bboxes_ignore']])
labels = np.concatenate([labels, ann['labels_ignore']])
gt_bboxes.append(bboxes)
gt_labels.append(labels)
if not gt_ignore:
gt_ignore = None
mean_ap, eval_results = eval_map(det_results,
gt_bboxes,
gt_labels,
gt_ignore=gt_ignore,
scale_ranges=None,
iou_thr=iou_thr,
dataset=dataset.CLASSES,
dataset_name=dataset_name,
print_summary=True,
split=split)
return mean_ap, eval_results
def evaluate(self,
results,
metric='mAP',
logger=None,
iou_thr=0.5,
nproc=4,
**kwargs):
if not isinstance(metric, str):
assert len(metric) == 1
metric = metric[0]
if 'logger' in kwargs.keys():
logger = kwargs['logger']
ds_name = self.CLASSES
allowed_metrics = ['mAP']
if metric not in allowed_metrics:
raise KeyError('metric {} is not supported'.format(metric))
annotations = [self.get_ann_info(i) for i in range(len(self))]
eval_results = {}
assert isinstance(iou_thr, float)
mean_ap, _ = eval_map(results,
annotations,
scale_ranges=None,
iou_thr=iou_thr,
dataset=ds_name,
logger=logger)
eval_results['mAP'] = mean_ap
return eval_results
def voc_eval_with_return(result_file,
dataset,
iou_thr=0.5,
logger='print',
only_ap=True):
det_results = mmcv.load(result_file)
annotations = [dataset.get_ann_info(i) for i in range(len(dataset))]
if hasattr(dataset, 'year') and dataset.year == 2007:
dataset_name = 'voc07'
else:
dataset_name = dataset.CLASSES
mean_ap, eval_results = eval_map(
det_results,
annotations,
scale_ranges=None,
iou_thr=iou_thr,
dataset=dataset_name,
logger=logger)
if only_ap:
eval_results = [{
'ap': eval_results[i]['ap']
} for i in range(len(eval_results))]
return mean_ap, eval_results
def evaluate(self,
results,
metric='mAP',
logger=None,
proposal_nums=(100, 300, 1000),
iou_thr=0.5,
scale_ranges=None):
if not isinstance(metric, str):
assert len(metric) == 1
metric = metric[0]
allowed_metrics = [
'mAP',
]
if metric not in allowed_metrics:
raise KeyError(f'metric {metric} is not supported')
annotations = [self.get_ann_info(i) for i in range(len(self))]
eval_results = {}
if metric == 'mAP':
assert isinstance(iou_thr, float)
for stage in range(3):
if stage == 0:
ds_name = self.dataset.CLASSES_1
label = 'category'
else:
ds_name = self.dataset.CLASSES_2
label = 'pests'
stage_annotations = []
for anno in annotations:
anno['labels'] = anno[label]
stage_annotations.append(anno)
'''
result (temp) : [[det_results_1], [det_results_2], [det_results_3]]
det_results (list[list]): [[cls1_det, cls2_det, ...], ...].
The outer list indicates images, and the inner list indicates
per-class detected bboxes.
annotations (list[dict]): Ground truth annotations where each item of
the list indicates an image. Keys of annotations are:
- `bboxes`: numpy array of shape (n, 4)
- `labels`: numpy array of shape (n, )
- `bboxes_ignore` (optional): numpy array of shape (k, 4)
- `labels_ignore` (optional): numpy array of shape (k, )
'''
print('eval stage %d mAP from outputs' % (stage + 1))
mean_ap, _ = eval_map(results[stage],
stage_annotations,
scale_ranges=None,
iou_thr=iou_thr,
dataset=ds_name,
logger=logger)
eval_results['mAP'] = mean_ap
return eval_results
def evaluate(self,
results,
metric='mAP',
logger=None,
proposal_nums=(100, 300, 1000),
iou_thr=0.5,
scale_ranges=None):
"""Evaluate the dataset.
Args:
results (list): Testing results of the dataset.
metric (str | list[str]): Metrics to be evaluated.
logger (logging.Logger | None | str): Logger used for printing
related information during evaluation. Default: None.
proposal_nums (Sequence[int]): Proposal number used for evaluating
recalls, such as recall@100, recall@1000.
Default: (100, 300, 1000).
iou_thr (float | list[float]): IoU threshold. Default: 0.5.
scale_ranges (list[tuple] | None): Scale ranges for evaluating mAP.
Default: None.
"""
if not isinstance(metric, str):
assert len(metric) == 1
metric = metric[0]
allowed_metrics = ['mAP', 'recall']
if metric not in allowed_metrics:
raise KeyError(f'metric {metric} is not supported')
annotations = [self.get_ann_info(i) for i in range(len(self))]
eval_results = OrderedDict()
iou_thrs = [iou_thr] if isinstance(iou_thr, float) else iou_thr
if metric == 'mAP':
assert isinstance(iou_thrs, list)
mean_aps = []
for iou_thr in iou_thrs:
print_log(f'\n{"-" * 15}iou_thr: {iou_thr}{"-" * 15}')
mean_ap, _ = eval_map(results,
annotations,
scale_ranges=scale_ranges,
iou_thr=iou_thr,
dataset=self.CLASSES,
logger=logger)
mean_aps.append(mean_ap)
eval_results[f'AP{int(iou_thr * 100):02d}'] = round(mean_ap, 3)
eval_results['mAP'] = sum(mean_aps) / len(mean_aps)
elif metric == 'recall':
gt_bboxes = [ann['bboxes'] for ann in annotations]
recalls = eval_recalls(gt_bboxes,
results,
proposal_nums,
iou_thr,
logger=logger)
for i, num in enumerate(proposal_nums):
for j, iou in enumerate(iou_thrs):
eval_results[f'recall@{num}@{iou}'] = recalls[i, j]
if recalls.shape[1] > 1:
ar = recalls.mean(axis=1)
for i, num in enumerate(proposal_nums):
eval_results[f'AR@{num}'] = ar[i]
return eval_results
def evaluate(self,
results,
metric=None,
iou_thr=(0.25, 0.5),
iou_thr_2d=(0.5, ),
logger=None,
show=False,
out_dir=None):
# evaluate 3D detection performance
if isinstance(results[0], dict):
return super().evaluate(results, metric, iou_thr, logger, show,
out_dir)
# evaluate 2D detection performance
else:
eval_results = OrderedDict()
annotations = [self.get_ann_info(i) for i in range(len(self))]
iou_thr_2d = (iou_thr_2d) if isinstance(iou_thr_2d,
float) else iou_thr_2d
for iou_thr_2d_single in iou_thr_2d:
mean_ap, _ = eval_map(results,
annotations,
scale_ranges=None,
iou_thr=iou_thr_2d_single,
dataset=self.CLASSES,
logger=logger)
eval_results['mAP_' + str(iou_thr_2d_single)] = mean_ap
return eval_results
def evaluate(
self,
results,
metric="mAP",
logger=None,
proposal_nums=(100, 300, 1000),
iou_thr=0.5,
scale_ranges=None,
):
"""Evaluate the dataset.
Args:
results (list): Testing results of the dataset.
metric (str | list[str]): Metrics to be evaluated.
logger (logging.Logger | None | str): Logger used for printing
related information during evaluation. Default: None.
proposal_nums (Sequence[int]): Proposal number used for evaluating
recalls, such as recall@100, recall@1000.
Default: (100, 300, 1000).
iou_thr (float | list[float]): IoU threshold. It must be a float
when evaluating mAP, and can be a list when evaluating recall.
Default: 0.5.
scale_ranges (list[tuple] | None): Scale ranges for evaluating mAP.
Default: None.
"""
if not isinstance(metric, str):
assert len(metric) == 1
metric = metric[0]
allowed_metrics = ["mAP", "recall"]
if metric not in allowed_metrics:
raise KeyError("metric {} is not supported".format(metric))
annotations = [self.get_ann_info(i) for i in range(len(self))]
eval_results = {}
if metric == "mAP":
assert isinstance(iou_thr, float)
mean_ap, _ = eval_map(
results,
annotations,
scale_ranges=scale_ranges,
iou_thr=iou_thr,
dataset=self.CLASSES,
logger=logger,
)
eval_results["mAP"] = mean_ap
elif metric == "recall":
gt_bboxes = [ann["bboxes"] for ann in annotations]
if isinstance(iou_thr, float):
iou_thr = [iou_thr]
recalls = eval_recalls(
gt_bboxes, results, proposal_nums, iou_thr, logger=logger
)
for i, num in enumerate(proposal_nums):
for j, iou in enumerate(iou_thr):
eval_results["recall@{}@{}".format(num, iou)] = recalls[i, j]
if recalls.shape[1] > 1:
ar = recalls.mean(axis=1)
for i, num in enumerate(proposal_nums):
eval_results["AR@{}".format(num)] = ar[i]
return eval_results
def get_iou_recall_curve(cfg_path, pth_path, out_path):
cfg = mmcv.Config.fromfile(cfg_path)
# set cudnn_benchmark
if cfg.get('cudnn_benchmark', False):
torch.backends.cudnn.benchmark = True
cfg.model.pretrained = None
cfg.data.test.test_mode = True
# init distributed env first, since logger depends on the dist info.
distributed = False
# build the dataloader
# TODO: support multiple images per gpu (only minor changes are needed)
dataset = build_dataset(cfg.data.test)
data_loader = build_dataloader(
dataset,
imgs_per_gpu=1,
workers_per_gpu=cfg.data.workers_per_gpu,
dist=distributed,
shuffle=False)
# build the model and load checkpoint
model = init_detector(cfg_path, pth_path, device='cuda:0')
results = [None for _ in range(len(dataset))]
print(dataset)
for idx in range(0, len(dataset), 1):
data = dataset[idx]['img'][0].numpy()
data = data.transpose((1, 2, 0))
# compute output
result = inference_detector(model, data)
results[idx] = result
annotations = [dataset.get_ann_info(i) for i in range(len(dataset))]
out_list = []
for thr in range(0, 21):
mean_ap, eval_results = eval_map(
results,
annotations,
scale_ranges=None,
iou_thr=thr/20,
dataset=None,
logger=None,
nproc=1)
out_list.append([thr/20, eval_results[0]['ap'][-1], eval_results[1]['ap'][-1],mean_ap])
print(out_list)
f = open(out_path, 'w')
for ln in out_list:
f.write(str(ln[0]))
f.write(',')
f.write(str(ln[1]))
f.write(',')
f.write(str(ln[2]))
f.write('\n')
f.close()
def vid_eval(result_file, dataset, iou_thr=0.5):
det_results = mmcv.load(result_file)
gt_bboxes = []
gt_labels = []
gt_ignore = []
for i in range(len(dataset)):
ann = dataset.get_ann_info(i)
bboxes = ann['bboxes']
labels = ann['labels']
if 'bboxes_ignore' in ann:
ignore = np.concatenate([
np.zeros(bboxes.shape[0], dtype=np.bool),
np.ones(ann['bboxes_ignore'].shape[0], dtype=np.bool)
])
gt_ignore.append(ignore)
bboxes = np.vstack([bboxes, ann['bboxes_ignore']])
labels = np.concatenate([labels, ann['labels_ignore']])
gt_bboxes.append(bboxes)
gt_labels.append(labels)
if not gt_ignore:
gt_ignore = None
if hasattr(dataset, 'year') and dataset.year == 2007:
dataset_name = 'voc07'
else:
dataset_name = ('airplane', 'antelope', 'bear', 'bicycle', 'bird',
'bus', 'car', 'cattle', 'dog', 'domestic_cat',
'elephant', 'fox', 'giant_panda', 'hamster', 'horse',
'lion', 'lizard', 'monkey', 'motorcycle', 'rabbit',
'red_panda', 'sheep', 'snake', 'squirrel', 'tiger',
'train', 'turtle', 'watercraft', 'whale', 'zebra')
# dataset_name = dataset.CLASSES
eval_map(det_results,
gt_bboxes,
gt_labels,
gt_ignore=gt_ignore,
scale_ranges=None,
iou_thr=iou_thr,
dataset=dataset_name,
print_summary=True)
def evaluate(self,
results,
metric='mAP',
logger=None,
proposal_nums=(100, 300, 1000),
iou_thr=0.5,
scale_ranges=None):
if not isinstance(metric, str):
assert len(metric) == 1
metric = metric[0]
allowed_metrics = ['mAP', 'recall']
if metric not in allowed_metrics:
raise KeyError('metric {} is not supported'.format(metric))
annotations = [self.get_ann_info(i) for i in range(len(self))]
eval_results = {}
if metric == 'mAP':
assert isinstance(iou_thr, float)
if self.year == 2007:
ds_name = 'voc07'
else:
#ds_name = self.dataset.CLASSES
ds_name = (
'Benign',
'Malignant',
)
mean_ap, _ = eval_map(results,
annotations,
scale_ranges=None,
iou_thr=iou_thr,
dataset=ds_name,
logger=logger)
eval_results['mAP'] = mean_ap
elif metric == 'recall':
gt_bboxes = [ann['bboxes'] for ann in annotations]
if isinstance(iou_thr, float):
iou_thr = [iou_thr]
recalls = eval_recalls(gt_bboxes,
results,
proposal_nums,
iou_thr,
logger=logger)
for i, num in enumerate(proposal_nums):
for j, iou in enumerate(iou_thr):
eval_results['recall@{}@{}'.format(num, iou)] = recalls[i,
j]
if recalls.shape[1] > 1:
ar = recalls.mean(axis=1)
for i, num in enumerate(proposal_nums):
eval_results['AR@{}'.format(num)] = ar[i]
return eval_results
def chips_eval(result_file, dataset, iou_thr=0.5):
det_results = mmcv.load(result_file)
gt_bboxes = []
gt_labels = []
gt_ignore = []
gt_masks = []
for i in range(len(dataset)):
ann = dataset.get_ann_info(i)
bboxes = ann['bboxes']
labels = ann['labels']
masks = ann['masks']
if 'bboxes_ignore' in ann:
ignore = np.concatenate([
np.zeros(bboxes.shape[0], dtype=np.bool),
np.ones(ann['bboxes_ignore'].shape[0], dtype=np.bool)
])
gt_ignore.append(ignore)
bboxes = np.vstack([bboxes, ann['bboxes_ignore']])
# masks = np.vstack([bboxes, ann['masks_ignore']])
labels = np.concatenate([labels, ann['labels_ignore']])
gt_bboxes.append(bboxes)
gt_labels.append(labels)
gt_masks.append(masks)
if not gt_ignore:
gt_ignore = None
print(len(det_results[1][1]))
print(det_results[1][1])
dataset_name = dataset.CLASSES
eval_map(
det_results,
gt_bboxes,
# gt_masks,
gt_labels,
gt_ignore=gt_ignore,
scale_ranges=None,
iou_thr=iou_thr,
dataset=dataset_name,
print_summary=True)
def evaluate(
self,
results,
metric="mAP",
logger=None,
proposal_nums=(100, 300, 1000),
iou_thr=0.5,
scale_ranges=None,
):
if not isinstance(metric, str):
assert len(metric) == 1
metric = metric[0]
allowed_metrics = ["mAP", "recall"]
if metric not in allowed_metrics:
raise KeyError("metric {} is not supported".format(metric))
annotations = [self.get_ann_info(i) for i in range(len(self))]
eval_results = {}
if metric == "mAP":
assert isinstance(iou_thr, float)
if self.year == 2007:
ds_name = "voc07"
else:
ds_name = self.dataset.CLASSES
mean_ap, _ = eval_map(
results,
annotations,
scale_ranges=None,
iou_thr=iou_thr,
dataset=ds_name,
logger=logger,
)
eval_results["mAP"] = mean_ap
elif metric == "recall":
gt_bboxes = [ann["bboxes"] for ann in annotations]
if isinstance(iou_thr, float):
iou_thr = [iou_thr]
recalls = eval_recalls(gt_bboxes,
results,
proposal_nums,
iou_thr,
logger=logger)
for i, num in enumerate(proposal_nums):
for j, iou in enumerate(iou_thr):
eval_results["recall@{}@{}".format(num, iou)] = recalls[i,
j]
if recalls.shape[1] > 1:
ar = recalls.mean(axis=1)
for i, num in enumerate(proposal_nums):
eval_results["AR@{}".format(num)] = ar[i]
return eval_results
def txt_eval(result_file, dataset, iou_thr=0.5):
det_results = mmcv.load(result_file)
gt_bboxes = []
gt_labels = []
gt_ignore = []
for i in range(len(dataset)):
ann = dataset.get_ann_info(i)
bboxes = ann['bboxes']
labels = ann['labels']
# for k, v in ann.items():
# print(k)
if ('bboxes_ignore' in ann) and ('labels_ignore' in ann):
print('1')
ignore = np.concatenate([
np.zeros(bboxes.shape[0], dtype=np.bool),
np.ones(ann['bboxes_ignore'].shape[0], dtype=np.bool)
])
gt_ignore.append(ignore)
bboxes = np.vstack([bboxes, ann['bboxes_ignore']])
labels = np.concatenate([labels, ann['labels_ignore']])
gt_bboxes.append(bboxes)
gt_labels.append(labels)
if not gt_ignore:
#print('2')
gt_ignore = gt_ignore
dataset_name = dataset.CLASSES
print(dataset_name)
eval_map(det_results,
gt_bboxes,
gt_labels,
gt_ignore=None,
scale_ranges=None,
iou_thr=iou_thr,
dataset=dataset_name,
print_summary=True)
def evaluate(self,
results,
metric=None,
iou_thr=(0.25, 0.5),
iou_thr_2d=(0.5, ),
logger=None,
show=False,
out_dir=None,
pipeline=None):
"""Evaluate.
Evaluation in indoor protocol.
Args:
results (list[dict]): List of results.
metric (str | list[str]): Metrics to be evaluated.
iou_thr (list[float]): AP IoU thresholds.
iou_thr_2d (list[float]): AP IoU thresholds for 2d evaluation.
show (bool): Whether to visualize.
Default: False.
out_dir (str): Path to save the visualization results.
Default: None.
pipeline (list[dict], optional): raw data loading for showing.
Default: None.
Returns:
dict: Evaluation results.
"""
# evaluate 3D detection performance
if isinstance(results[0], dict):
return super().evaluate(results, metric, iou_thr, logger, show,
out_dir, pipeline)
# evaluate 2D detection performance
else:
eval_results = OrderedDict()
annotations = [self.get_ann_info(i) for i in range(len(self))]
iou_thr_2d = (iou_thr_2d) if isinstance(iou_thr_2d,
float) else iou_thr_2d
for iou_thr_2d_single in iou_thr_2d:
mean_ap, _ = eval_map(
results,
annotations,
scale_ranges=None,
iou_thr=iou_thr_2d_single,
dataset=self.CLASSES,
logger=logger)
eval_results['mAP_' + str(iou_thr_2d_single)] = mean_ap
return eval_results
def evaluate(self,
results,
metric=['mAP'],
logger=None,
proposal_nums=(4, 10, 50),
iou_thr=0.5,
scale_ranges=None):
"""Evaluate the dataset.
Args:
results (list): Testing results of the dataset.
metric (str | list[str]): Metrics to be evaluated.
logger (logging.Logger | None | str): Logger used for printing
related information during evaluation. Default: None.
proposal_nums (Sequence[int]): Proposal number used for evaluating
recalls, such as recall@100, recall@1000.
Default: (100, 300, 1000).
iou_thr (float | list[float]): IoU threshold. It must be a float
when evaluating mAP, and can be a list when evaluating recall.
Default: 0.5.
scale_ranges (list[tuple] | None): Scale ranges for evaluating mAP.
Default: None.
"""
metrics = metric if isinstance(metric, list) else [metric]
allowed_metrics = ['mAP']
for metric in metrics:
if metric not in allowed_metrics:
raise KeyError(f'metric {metric} is not supported')
annotations = [self.get_ann_info(i) for i in range(len(self))]
eval_results = {}
for metric in metrics:
if metric == 'mAP':
assert isinstance(iou_thr, float)
mean_ap, _ = eval_map(results,
annotations,
scale_ranges=scale_ranges,
iou_thr=iou_thr,
dataset=self.CLASSES,
logger=logger)
eval_results['mAP'] = mean_ap
return eval_results
def voc_eval(result_file, dataset):
det_results = mmcv.load(result_file)
gt_bboxes = []
gt_labels = []
gt_ignore = []
for i in range(len(dataset)):
ann = dataset.get_ann_info(i)
bboxes = ann['bboxes']
labels = ann['labels']
if 'bboxes_ignore' in ann:
ignore = np.concatenate([
np.zeros(bboxes.shape[0], dtype=np.bool),
np.ones(ann['bboxes_ignore'].shape[0], dtype=np.bool)
])
gt_ignore.append(ignore)
bboxes = np.vstack([bboxes, ann['bboxes_ignore']])
labels = np.concatenate([labels, ann['labels_ignore']])
gt_bboxes.append(bboxes)
gt_labels.append(labels)
if not gt_ignore:
gt_ignore = gt_ignore
if hasattr(dataset, 'year') and dataset.year == 2007:
dataset_name = 'voc07'
else:
dataset_name = dataset.CLASSES
ious = [0.1, 0.3, 0.5]
mean_aps = 0
for iou in ious:
mean_ap, _ = eval_map(det_results,
gt_bboxes,
gt_labels,
gt_ignore=gt_ignore,
scale_ranges=None,
iou_thr=iou,
dataset=dataset_name,
print_summary=True)
mean_aps += mean_ap
print("\naverage mean_ap = {}\n".format(mean_aps / 3))
def evaluate(result_file, dataset, iou_thresholds):
det_results = mmcv.load(result_file)
gt_bboxes = []
gt_labels = []
gt_ignore = []
for i in range(len(dataset)):
ann = dataset.get_ann_info(i)
bboxes = ann['bboxes']
labels = ann['labels']
if 'bboxes_ignore' in ann:
ignore = np.concatenate([
np.zeros(bboxes.shape[0], dtype=np.bool),
np.ones(ann['bboxes_ignore'].shape[0], dtype=np.bool)
])
gt_ignore.append(ignore)
bboxes = np.vstack([bboxes, ann['bboxes_ignore']])
labels = np.concatenate([labels, ann['labels_ignore']])
gt_bboxes.append(bboxes)
gt_labels.append(labels)
if not gt_ignore:
gt_ignore = None
if hasattr(dataset, 'year') and dataset.year == 2007:
dataset_name = 'voc07'
else:
dataset_name = dataset.CLASSES
mean_aps = []
for iou_thr in iou_thresholds:
mean_ap, _ = eval_map(det_results,
gt_bboxes,
gt_labels,
gt_ignore=gt_ignore,
scale_ranges=None,
iou_thr=iou_thr,
dataset=dataset_name,
print_summary=False)
mean_aps.append(mean_ap)
print(f'MAP: {np.mean(mean_aps)}')
def main():
args = parse_args()
cfg = mmcv.Config.fromfile(args.config)
# set cudnn_benchmark
if cfg.get('cudnn_benchmark', False):
torch.backends.cudnn.benchmark = True
cfg.model.pretrained = None
cfg.data.test.test_mode = True
# 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)
# build the dataloader
# TODO: support multiple images per gpu (only minor changes are needed)
dataset = build_dataset(cfg.data.test)
data_loader = build_dataloader(
dataset,
imgs_per_gpu=1,
workers_per_gpu=cfg.data.workers_per_gpu,
dist=distributed,
shuffle=False)
if args.out is None:
dataset_name = dataset.name if hasattr(dataset, 'name') else 'val'
if hasattr(cfg.data.test, 'task'):
dataset_name = dataset_name + '_' + cfg.data.test.task
model_name = os.path.basename(args.checkpoint).split('.')[0]
model_dir = os.path.dirname(args.checkpoint)
args.out = os.path.join(model_dir, 'raw_results',
dataset_name + '_' + model_name + '.pkl')
elif not args.out.endswith(('.pkl', '.pickle')):
raise ValueError('The output file must be a pkl file.')
mmcv.mkdir_or_exist(os.path.dirname(args.out))
rank, _ = get_dist_info()
eval_types = args.eval
if not os.path.isfile(args.out):
# build the model and load checkpoint
model = build_detector(
cfg.model, train_cfg=None, test_cfg=cfg.test_cfg)
fp16_cfg = cfg.get('fp16', None)
if fp16_cfg is not None:
wrap_fp16_model(model)
checkpoint = load_checkpoint(
model, args.checkpoint, map_location='cpu')
# old versions did not save class info in checkpoints, this walkaround is
# for backward compatibility
if 'CLASSES' in checkpoint['meta']:
model.CLASSES = checkpoint['meta']['CLASSES']
else:
model.CLASSES = dataset.CLASSES
if not distributed:
model = MMDataParallel(model, device_ids=[0])
outputs = single_gpu_test(model, data_loader, args.show)
else:
model = MMDistributedDataParallel(model.cuda())
outputs = multi_gpu_test(model, data_loader, args.tmpdir)
if rank == 0:
if hasattr(dataset, 'raw_annotations'):
filenames = [
dataset.raw_annotations[dataset.ids[i]]['filename']
for i in range(len(dataset))
]
else:
filenames = [
img_info['filename'] for img_info in dataset.img_infos
]
print('\nwriting results to {}'.format(args.out))
results = {
'file_names': filenames,
'outputs': outputs,
}
mmcv.dump(results, args.out, protocol=2)
elif rank == 0:
results = mmcv.load(args.out, encoding='latin1')
outputs = results['outputs']
if eval_types and rank == 0:
print('Starting evaluate {}'.format(' and '.join(eval_types)))
if not hasattr(dataset, 'coco'):
if hasattr(dataset, 'raw_annotations'):
gt_bboxes = [
dataset.raw_annotations[dataset.ids[i]]['ann']['bboxes']
for i in range(len(dataset))
]
gt_labels = [
dataset.raw_annotations[dataset.ids[i]]['ann']['classes']
for i in range(len(dataset))
]
if cfg.data.test.with_ignore:
gt_ignores = [l <= 0 for l in gt_labels]
else:
gt_ignores = [l == 0 for l in gt_labels]
gt_labels = [np.abs(l) for l in gt_labels]
if 'corners' in eval_types:
gt_corners = [
dataset.raw_annotations[dataset.ids[i]]['ann']
['corners'] for i in range(len(dataset))
]
gt_poses = [
dataset.raw_annotations[dataset.ids[i]]['ann']['poses']
for i in range(len(dataset))
]
eval_corners(
outputs,
gt_corners,
gt_poses,
gt_labels,
gt_ignores,
gt_bboxes=gt_bboxes,
display=True)
det_bboxes = corners2bboxes(outputs,
len(dataset.CLASSES) - 1)
eval_map(
det_bboxes, gt_bboxes, gt_labels, gt_ignore=gt_ignores)
else:
eval_map(
outputs, gt_bboxes, gt_labels, gt_ignore=gt_ignores)
else:
gt_bboxes = [
img_info['ann']['bboxes'] for img_info in dataset.img_infos
]
gt_labels = [
img_info['ann']['labels'] for img_info in dataset.img_infos
]
if len(outputs[0]) == 5:
outputs = corners2bboxes(outputs, len(dataset.classes) - 1)
eval_map(outputs, gt_bboxes, gt_labels, iou_thr=0.4)
else:
if eval_types == ['proposal_fast']:
result_file = args.out
coco_eval(result_file, eval_types, dataset.coco, CLASSES=dataset.CLASSES)
else:
if not isinstance(outputs[0], dict):
result_files = results2json(dataset, outputs, args.out)
coco_eval(result_files, eval_types, dataset.coco, CLASSES=dataset.CLASSES, show=True)
else:
for name in outputs[0]:
print('\nEvaluating {}'.format(name))
outputs_ = [out[name] for out in outputs]
result_file = args.out + '.{}'.format(name)
result_files = results2json(dataset, outputs_,
result_file)
coco_eval(result_files, eval_types, dataset.coco, CLASSES=dataset.CLASSES, show=True)
def evaluate(self,
results,
metric='mAP',
logger=None,
proposal_nums=(100, 300, 1000),
iou_thr=0.5,
scale_ranges=None):
"""Evaluate in VOC protocol.
Args:
results (list[list | tuple]): Testing results of the dataset.
metric (str | list[str]): Metrics to be evaluated. Options are
'mAP', 'recall'.
logger (logging.Logger | str, optional): Logger used for printing
related information during evaluation. Default: None.
proposal_nums (Sequence[int]): Proposal number used for evaluating
recalls, such as recall@100, recall@1000.
Default: (100, 300, 1000).
iou_thr (float | list[float]): IoU threshold. It must be a float
when evaluating mAP, and can be a list when evaluating recall.
Default: 0.5.
scale_ranges (list[tuple], optional): Scale ranges for evaluating
mAP. If not specified, all bounding boxes would be included in
evaluation. Default: None.
Returns:
dict[str, float]: AP/recall metrics.
"""
if not isinstance(metric, str): # isinstance判断metric是不是str类别
assert len(metric) == 1
metric = metric[0]
allowed_metrics = ['mAP', 'recall']
if metric not in allowed_metrics:
raise KeyError(f'metric {metric} is not supported')
annotations = [self.get_ann_info(i) for i in range(len(self))]
eval_results = OrderedDict()
if metric == 'mAP':
assert isinstance(iou_thr, float)
if self.year == 2007:
ds_name = 'voc07'
else:
ds_name = self.CLASSES
mean_ap, _ = eval_map(results,
annotations,
scale_ranges=None,
iou_thr=iou_thr,
dataset=ds_name,
logger=logger)
eval_results['mAP'] = mean_ap
elif metric == 'recall':
gt_bboxes = [ann['bboxes'] for ann in annotations]
if isinstance(iou_thr, float):
iou_thr = [iou_thr]
recalls = eval_recalls(gt_bboxes,
results,
proposal_nums,
iou_thr,
logger=logger)
for i, num in enumerate(proposal_nums):
for j, iou in enumerate(iou_thr):
eval_results[f'recall@{num}@{iou}'] = recalls[i, j]
if recalls.shape[1] > 1:
ar = recalls.mean(axis=1)
for i, num in enumerate(proposal_nums):
eval_results[f'AR@{num}'] = ar[i]
return eval_results
# merge
results = []
iter = 1
for img_id in tqdm(img_list):
iter += 1
img_name = img_id + '.jpg'
det_nms = []
if img_name in detecions:
det = np.array(detecions[img_name])
det = det[det[:, -2] > args.score_thr]
for i in range(args.nclass):
det_nms.append(nms(det[det[:, -1] == i, :5], iou_thr=0.5)[0])
else:
det_nms = [np.array([]).reshape(0, 5) for i in range(args.nclass)]
results.append(det_nms)
# ground truth
xml_file = osp.join(source_anno, img_id + '.xml')
bboxes, labels = getGtFromXml(xml_file)
annotations.append({"bboxes": bboxes, "labels": labels})
# show
img_file = osp.join(source_img, img_name)
# model.show_result(img_file, det_nms, out_file='source_result.jpg')
# voc metric
eval_results = eval_map(results, annotations, iou_thr=0.5,
logger="print")[0]
pass
def evaluate(self,
results,
metric='track',
logger=None,
resfile_path=None,
bbox_iou_thr=0.5,
track_iou_thr=0.5):
eval_results = dict()
if isinstance(metric, list):
metrics = metric
elif isinstance(metric, str):
metrics = [metric]
else:
raise TypeError('metric must be a list or a str.')
allowed_metrics = ['bbox', 'track']
for metric in metrics:
if metric not in allowed_metrics:
raise KeyError(f'metric {metric} is not supported.')
if 'track' in metrics:
resfiles, names, tmp_dir = self.format_results(
results, resfile_path, metrics, logger)
print_log('Evaluate CLEAR MOT results.', logger=logger)
distth = 1 - track_iou_thr
accs = []
for name in names:
if 'half-train' in self.ann_file:
gt_file = osp.join(self.img_prefix,
f'{name}/gt/gt_half-train.txt')
elif 'half-val' in self.ann_file:
gt_file = osp.join(self.img_prefix,
f'{name}/gt/gt_half-val.txt')
else:
gt_file = osp.join(self.img_prefix, f'{name}/gt/gt.txt')
res_file = osp.join(resfiles['track'], f'{name}.txt')
gt = mm.io.loadtxt(gt_file)
res = mm.io.loadtxt(res_file)
ini_file = osp.join(self.img_prefix, f'{name}/seqinfo.ini')
if osp.exists(ini_file):
acc, ana = mm.utils.CLEAR_MOT_M(
gt, res, ini_file, distth=distth)
else:
acc = mm.utils.compare_to_groundtruth(
gt, res, distth=distth)
accs.append(acc)
mh = mm.metrics.create()
summary = mh.compute_many(
accs,
names=names,
metrics=mm.metrics.motchallenge_metrics,
generate_overall=True)
str_summary = mm.io.render_summary(
summary,
formatters=mh.formatters,
namemap=mm.io.motchallenge_metric_names)
print(str_summary)
eval_results.update({
mm.io.motchallenge_metric_names[k]: v['OVERALL']
for k, v in summary.to_dict().items()
})
if tmp_dir is not None:
tmp_dir.cleanup()
if 'bbox' in metrics:
if isinstance(results, dict):
bbox_results = results['bbox_results']
elif isinstance(results, list):
bbox_results = results
else:
raise TypeError('results must be a dict or a list.')
annotations = [self.get_ann_info(info) for info in self.data_infos]
mean_ap, _ = eval_map(
bbox_results,
annotations,
iou_thr=bbox_iou_thr,
dataset=self.CLASSES,
logger=logger)
eval_results['mAP'] = mean_ap
for k, v in eval_results.items():
if isinstance(v, float):
eval_results[k] = float(f'{(v):.3f}')
return eval_results
def main():
args = parse_args()
# assert args.show or args.json_out, \
# ('Please specify at least one operation (save or show the results) '
# 'with the argument "--out" or "--show" or "--json_out"')
if args.json_out is not None and args.json_out.endswith('.json'):
args.json_out = args.json_out[:-5]
cfg = mmcv.Config.fromfile(args.config)
checkpoint_file = args.checkpoint
if not checkpoint_file:
def _epoch_num(name):
return int(
re.findall('epoch_[0-9]*.pth',
name)[0].replace('epoch_', '').replace('.pth', ''))
pths = sorted(glob.glob(os.path.join(cfg.work_dir, 'epoch_*.pth')),
key=_epoch_num)
if len(pths) > 0:
print("Found {}, use it as checkpoint by default.".format(
pths[-1]))
checkpoint_file = pths[-1]
if not checkpoint_file:
raise ValueError("Checkpoints not found, check work_dir non empty.")
# set cudnn_benchmark
if cfg.get('cudnn_benchmark', False):
torch.backends.cudnn.benchmark = True
cfg.model.pretrained = None
cfg.data.test.test_mode = True
# 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)
# build the dataloader
# TODO: support multiple images per gpu (only minor changes are needed)
dataset = build_dataset(cfg.data.test)
data_loader = build_dataloader(
dataset,
imgs_per_gpu=1,
workers_per_gpu=cfg.data.workers_per_gpu,
dist=distributed,
shuffle=args.shuffle) # TODO: hack shuffle True
# build the model and load checkpoint
model = build_detector(cfg.model, train_cfg=None, test_cfg=cfg.test_cfg)
fp16_cfg = cfg.get('fp16', None)
if fp16_cfg is not None:
wrap_fp16_model(model)
checkpoint = load_checkpoint(model, checkpoint_file, map_location='cpu')
# old versions did not save class info in checkpoints, this walkaround is
# for backward compatibility
if 'CLASSES' in checkpoint['meta']:
model.CLASSES = checkpoint['meta']['CLASSES']
else:
model.CLASSES = dataset.CLASSES
num_evals = args.num_evals
if num_evals < 0:
num_evals = len(data_loader)
if not distributed:
model = MMDataParallel(model, device_ids=[0])
outputs = single_gpu_test(model, data_loader, num_evals, args.show)
else:
model = MMDistributedDataParallel(model.cuda())
outputs = multi_gpu_test(model, data_loader, num_evals, args.tmpdir)
rank, _ = get_dist_info()
if rank == 0:
gt_bboxes, gt_labels, gt_ignore, dataset_name = get_seq_gts(dataset)
print('\nStarting evaluate {}'.format(dataset_name))
eval_map(outputs,
gt_bboxes,
gt_labels,
gt_ignore,
scale_ranges=None,
iou_thr=0.5,
dataset=dataset_name,
print_summary=True)
# Save predictions in the COCO json format
if args.json_out and rank == 0:
if not isinstance(outputs[0], dict):
results2json(dataset, outputs, args.json_out)
else:
for name in outputs[0]:
outputs_ = [out[name] for out in outputs]
result_file = args.json_out + '.{}'.format(name)
results2json(dataset, outputs_, result_file)
def evaluate(self,
results,
metric='track',
logger=None,
resfile_path=None,
bbox_iou_thr=0.5,
track_iou_thr=0.5):
"""Evaluation in MOT Challenge.
Args:
results (list[list | tuple]): Testing results of the dataset.
metric (str | list[str]): Metrics to be evaluated. Options are
'bbox', 'track'. Defaults to 'track'.
logger (logging.Logger | str | None): Logger used for printing
related information during evaluation. Default: None.
resfile_path (str, optional): Path to save the formatted results.
Defaults to None.
bbox_iou_thr (float, optional): IoU threshold for detection
evaluation. Defaults to 0.5.
track_iou_thr (float, optional): IoU threshold for tracking
evaluation.. Defaults to 0.5.
Returns:
dict[str, float]: MOTChallenge style evaluation metric.
"""
eval_results = dict()
if isinstance(metric, list):
metrics = metric
elif isinstance(metric, str):
metrics = [metric]
else:
raise TypeError('metric must be a list or a str.')
allowed_metrics = ['bbox', 'track']
for metric in metrics:
if metric not in allowed_metrics:
raise KeyError(f'metric {metric} is not supported.')
if 'track' in metrics:
resfiles, names, tmp_dir = self.format_results(
results, resfile_path, metrics)
print_log('Evaluate CLEAR MOT results.', logger=logger)
distth = 1 - track_iou_thr
accs = []
for name in names:
if 'half-train' in self.ann_file:
gt_file = osp.join(self.img_prefix,
f'{name}/gt/gt_half-train.txt')
elif 'half-val' in self.ann_file:
gt_file = osp.join(self.img_prefix,
f'{name}/gt/gt_half-val.txt')
else:
gt_file = osp.join(self.img_prefix, f'{name}/gt/gt.txt')
res_file = osp.join(resfiles['track'], f'{name}.txt')
gt = mm.io.loadtxt(gt_file)
res = mm.io.loadtxt(res_file)
ini_file = osp.join(self.img_prefix, f'{name}/seqinfo.ini')
if osp.exists(ini_file) and 'MOT15' not in self.img_prefix:
acc, ana = mm.utils.CLEAR_MOT_M(gt,
res,
ini_file,
distth=distth)
else:
acc = mm.utils.compare_to_groundtruth(gt,
res,
distth=distth)
accs.append(acc)
mh = mm.metrics.create()
summary = mh.compute_many(accs,
names=names,
metrics=mm.metrics.motchallenge_metrics,
generate_overall=True)
str_summary = mm.io.render_summary(
summary,
formatters=mh.formatters,
namemap=mm.io.motchallenge_metric_names)
print(str_summary)
eval_results.update({
mm.io.motchallenge_metric_names[k]: v['OVERALL']
for k, v in summary.to_dict().items()
})
if tmp_dir is not None:
tmp_dir.cleanup()
if 'bbox' in metrics:
if isinstance(results, dict):
bbox_results = results['bbox_results']
elif isinstance(results, list):
bbox_results = results
else:
raise TypeError('results must be a dict or a list.')
annotations = [self.get_ann_info(info) for info in self.data_infos]
mean_ap, _ = eval_map(bbox_results,
annotations,
iou_thr=bbox_iou_thr,
dataset=self.CLASSES,
logger=logger)
eval_results['mAP'] = mean_ap
for k, v in eval_results.items():
if isinstance(v, float):
eval_results[k] = float(f'{(v):.3f}')
return eval_results