def __init__(self, num_class: int, strides: List[int],
anchors: List[List[int]], **kwargs):
super(YOLOv3TargetGenerator, self).__init__(**kwargs)
self._num_classes = num_class
self._strides = strides
self._anchors = anchors
self._item_len = 4 + 1 + num_class
self.bbox2center = BBoxCornerToCenter(axis=-1, split=True)
self.bbox2corner = BBoxCenterToCorner(axis=-1, split=False)
def __init__(self,
features,
scales,
ratios,
classes,
roi_size,
train_patterns,
stride=32,
rpn_channel=1024,
num_sample=128,
pos_iou_thresh=0.5,
neg_iou_thresh_high=0.5,
neg_iou_thresh_low=0.0,
pos_ratio=0.25,
nms_thresh=0.3,
nms_topk=400,
post_nms=100):
#super from rcnn
super(LHRCNN, self).__init__()
self.stride = stride
self._max_batch = 1 # currently only support batch size = 1
self._max_roi = 100000 # maximum allowed ROIs
self.num_class = len(classes)
self._target_generator = set([RCNNTargetGenerator(self.num_class)])
self.k, _ = roi_size
self.CT = 10
#--------RCNN parms--------
self.train_patterns = train_patterns
self.nms_thresh = nms_thresh
self.nms_topk = nms_topk
self.post_nms = post_nms
#-------nn init------------
with self.name_scope():
#-------ligth head rcnn setting-------
self.rpn = RPN(rpn_channel, stride, scales=scales, ratios=ratios)
self.sampler = RCNNTargetSampler(num_sample, pos_iou_thresh,
neg_iou_thresh_high,
neg_iou_thresh_low, pos_ratio)
self.group_conv = Group_Conv(10 * self.k * self.k)
#-------rcnn setting---------------
self.box_to_center = BBoxCornerToCenter()
self.box_decoder = NormalizedBoxCenterDecoder()
self.cls_decoder = MultiPerClassDecoder(num_class=self.num_class +
1)
self.features = features
self.share = nn.Dense(1024,
activation='relu',
weight_initializer=mx.init.Normal(0.01))
self.clf = nn.Dense(self.num_class + 1,
weight_initializer=mx.init.Normal(0.01))
self.reg = nn.Dense(self.num_class * 4,
weight_initializer=mx.init.Normal(0.01))
def __init__(self, features, top_features, classes, short, max_size, train_patterns=None,
nms_thresh=0.3, nms_topk=400, post_nms=100, roi_mode='align', roi_size=(14, 14), stride=16, clip=None,
rpn_channel=1024, base_size=16, scales=(0.5, 1, 2), ratios=(8, 16, 32), alloc_size=(128, 128), rpn_nms_thresh=0.5,
rpn_train_pre_nms=12000, rpn_train_post_nms=2000, rpn_test_pre_nms=6000, rpn_test_post_nms=300, rpn_min_size=16,
num_sample=128, pos_iou_thresh=0.5, pos_ratio=0.25):
super(FasterRCNN, self).__init__()
self.classes = classes
self.num_classes = len(classes)
self.short = short
self.max_size = max_size
self.train_patterns = train_patterns
self.nms_thresh = nms_thresh
self.nms_topk = nms_topk
self.post_nms = post_nms
self._max_batch = 1
self._num_sample = num_sample
self._rpn_test_post_nms = rpn_test_post_nms
# return cls_target, box_target, box_mask
self._target_generater = {RCNNTargetGenerator(self.num_classes)}
self._roi_mode = roi_mode.lower()
self._roi_size = roi_size
self._stride = stride
with self.name_scope():
self.features = features
self.top_features = top_features
self.global_avg_pool = nn.GlobalAvgPool2D()
self.class_predictor = nn.Dense(
self.num_classes+1, weight_initializer=mx.init.Normal(0.01))
self.box_predictor = nn.Dense(
self.num_classes*4, weight_initializer=mx.init.Normal(0.01))
# reconstruct valid labels
self.cls_decoder = MultiPerClassDecoder(num_class=self.num_classes+1)
# (xmin, ymin, xmax, ymax) -> (x, y, h, w)
self.box_to_center = BBoxCornerToCenter()
# reconstructed bounding boxes
self.box_decoder = NormalizedBoxCenterDecoder(clip=clip)
#
self.rpn = RPN(
channels=rpn_channel, stride=stride, base_size=base_size,
scales=scales, ratios=ratios, alloc_size=alloc_size,
clip=clip, nms_thresh=rpn_nms_thresh, train_pre_nms=rpn_train_pre_nms,
train_post_nms=rpn_train_post_nms, test_pre_nms=rpn_test_pre_nms, test_post_nms=rpn_test_post_nms, min_size=rpn_min_size)
self.sampler = RCNNTargetSampler(
num_image=self._max_batch, num_proposal=rpn_train_post_nms,
num_sample=num_sample, pos_iou_thresh=pos_iou_thresh, pos_ratio=pos_ratio)
def __init__(self,
num_class: int,
strides: List[int],
anchors: List[List[int]],
**kwargs):
super(YOLOv3TargetGenerator, self).__init__(**kwargs)
self._num_classes = num_class
self._strides = strides
self._anchors = anchors
self._item_len = 4 + 1 + num_class
self.bbox2center = BBoxCornerToCenter(axis=-1, split=True)
self.bbox2corner = BBoxCenterToCorner(axis=-1, split=False)
alloc_size = (128, 128)
grid_x = np.arange(alloc_size[1])
grid_y = np.arange(alloc_size[0])
grid_x, grid_y = np.meshgrid(grid_x, grid_y)
offsets = np.concatenate((grid_x[:, :, np.newaxis], grid_y[:, :, np.newaxis]), axis=-1)
offsets = np.expand_dims(np.expand_dims(offsets, axis=0), axis=0)
self._offsets = nd.array(offsets)
def __init__(self,
backbone_cfg,
neck_cfg,
head_cfg,
num_class: int,
strides: List[int],
anchors: List[List[List[int]]],
**kwargs):
if 'prefix' not in kwargs:
kwargs['prefix'] = self.__class__.__name__.lower() + '_'
super().__init__(**kwargs)
backbone = build_backbone(backbone_cfg)
neck = build_neck(neck_cfg)
head = build_head(head_cfg)
with self.name_scope():
self.backbone = backbone
self.neck = neck
self.head = head
self._num_classes = num_class
self._strides = strides
self._anchors = anchors
self._item_len = 4 + 1 + num_class
self.bbox2center = BBoxCornerToCenter(axis=-1, split=True)
self.bbox2corner = BBoxCenterToCorner(axis=-1, split=False)
def __init__(self, num_class, **kwargs):
super(YOLOV3PrefetchTargetGenerator, self).__init__(**kwargs)
self._num_class = num_class
self.bbox2center = BBoxCornerToCenter(axis=-1, split=True)
self.bbox2corner = BBoxCenterToCorner(axis=-1, split=False)