def __init__(self, num_anchors, num_classes):
super(YoloBody, self).__init__()
# backbone
self.backbone = darknet53(None)
self.conv1 = make_three_conv([512, 1024], 1024)
self.SPP = SpatialPyramidPooling()
self.conv2 = make_three_conv([512, 1024], 2048)
self.upsample1 = Upsample(512, 256)
self.conv_for_P4 = conv2d(512, 256, 1)
self.make_five_conv1 = make_five_conv([256, 512], 512)
self.upsample2 = Upsample(256, 128)
self.conv_for_P3 = conv2d(256, 128, 1)
self.make_five_conv2 = make_five_conv([128, 256], 256)
# 3*(5+num_classes)=3*(5+20)=3*(4+1+20)=75
# 4+1+num_classes
final_out_filter2 = num_anchors * (5 + num_classes)
self.yolo_head3 = yolo_head([256, final_out_filter2], 128)
self.down_sample1 = conv2d(128, 256, 3, stride=2)
self.make_five_conv3 = make_five_conv([256, 512], 512)
# 3*(5+num_classes)=3*(5+20)=3*(4+1+20)=75
final_out_filter1 = num_anchors * (5 + num_classes)
self.yolo_head2 = yolo_head([512, final_out_filter1], 256)
self.down_sample2 = conv2d(256, 512, 3, stride=2)
self.make_five_conv4 = make_five_conv([512, 1024], 1024)
# 3*(5+num_classes)=3*(5+20)=3*(4+1+20)=75
final_out_filter0 = num_anchors * (5 + num_classes)
self.yolo_head1 = yolo_head([1024, final_out_filter0], 512)
def __init__(self, num_anchors, num_classes):
super(YoloBody, self).__init__()
# ---------------------------------------------------#
# 生成CSPdarknet53的主干模型
# 获得三个有效特征层，他们的shape分别是：
# 52,52,256
# 26,26,512
# 13,13,1024
# ---------------------------------------------------#
self.draw_features = False
self.backbone = darknet53(None)
self.conv1 = make_three_conv([512, 1024], 1024)
self.SPP = SpatialPyramidPooling()
self.conv2 = make_three_conv([512, 1024], 2048)
self.upsample1 = Upsample(512, 256)
self.conv_for_P4 = conv2d(512, 256, 1)
self.make_five_conv1 = make_five_conv([256, 512], 512)
self.upsample2 = Upsample(256, 128)
self.conv_for_P3 = conv2d(256, 128, 1)
self.make_five_conv2 = make_five_conv([128, 256], 256)
# 3*(5+num_classes) = 3*(5+20) = 3*(4+1+20)=75
final_out_filter2 = num_anchors * (5 + num_classes)
self.yolo_head3 = yolo_head([256, final_out_filter2], 128)
self.down_sample1 = conv2d(128, 256, 3, stride=2)
self.make_five_conv3 = make_five_conv([256, 512], 512)
# 3*(5+num_classes) = 3*(5+20) = 3*(4+1+20)=75
final_out_filter1 = num_anchors * (5 + num_classes)
self.yolo_head2 = yolo_head([512, final_out_filter1], 256)
self.down_sample2 = conv2d(256, 512, 3, stride=2)
self.make_five_conv4 = make_five_conv([512, 1024], 1024)
# 3*(5+num_classes)=3*(5+20)=3*(4+1+20)=75
final_out_filter0 = num_anchors * (5 + num_classes)
self.yolo_head1 = yolo_head([1024, final_out_filter0], 512)
def __init__(self, num_anchors, num_classes):
super(YoloBody, self).__init__()
# backbone
self.backbone = darknet53(None)
self.conv1 = make_three_conv([512, 1024], 1024)
self.SPP = SpatialPyramidPooling()
self.conv2 = make_three_conv([512, 1024], 2048)
self.upsample1 = Upsample(512, 256)
self.conv_for_P4 = conv2d(512, 256, 1)
self.make_five_conv1 = make_five_conv([256, 512], 512)
self.upsample2 = Upsample(256, 128)
self.conv_for_P3 = conv2d(256, 128, 1)
self.make_five_conv2 = make_five_conv([128, 256], 256)
""" YOLO Head
1. 在特征利用部分，YoloV4提取多特征层进行目标检测，一共提取三个特征层，分别位于中间层，中下层，底层，
三个特征层的 shape 分别为(76,76,256)、(38,38,512)、(19,19,1024)
2. 输出层的 shape 分别为(19,19,75)，(38,38,75)，(76,76,75)，最后一个维度为 75 是因为该图是基于voc数据集的，它的类为20种，
YoloV4 只有针对每一个特征层存在 3 个先验框，所以最后维度为 3x25；
如果使用的是coco训练集，类则为80种，最后的维度应该为255 = 3x85，三个特征层的shape为(19,19,255)，(38,38,255)，(76,76,255)
"""
# 3*(5+num_classes)=3*(5+20)=3*(4+1+20)=75
# 4+1+num_classes
final_out_filter2 = num_anchors * (5 + num_classes)
self.yolo_head3 = yolo_head([256, final_out_filter2], 128)
self.down_sample1 = conv2d(128, 256, 3, stride=2)
self.make_five_conv3 = make_five_conv([256, 512], 512)
# 3*(5+num_classes)=3*(5+20)=3*(4+1+20)=75
# 5+num_classes = 4+1+num_classes
final_out_filter1 = num_anchors * (5 + num_classes)
self.yolo_head2 = yolo_head([512, final_out_filter1], 256)
self.down_sample2 = conv2d(256, 512, 3, stride=2)
self.make_five_conv4 = make_five_conv([512, 1024], 1024)
# 3*(5+num_classes)=3*(5+20)=3*(4+1+20)=75
final_out_filter0 = num_anchors * (5 + num_classes)
self.yolo_head1 = yolo_head([1024, final_out_filter0], 512)
def __init__(self, anchors_mask, num_classes):
super(YoloBody, self).__init__()
#---------------------------------------------------#
# 生成CSPdarknet53的主干模型
# 获得三个有效特征层，他们的shape分别是：
# 52,52,256
# 26,26,512
# 13,13,1024
#---------------------------------------------------#
self.backbone = darknet53(None)
self.conv1 = make_three_conv([512,1024],1024)
self.SPP = SpatialPyramidPooling()
self.conv2 = make_three_conv([512,1024],2048)
self.upsample1 = Upsample(512,256)
self.conv_for_P4 = conv2d(512,256,1)
self.make_five_conv1 = make_five_conv([256, 512],512)
self.upsample2 = Upsample(256,128)
self.conv_for_P3 = conv2d(256,128,1)
self.make_five_conv2 = make_five_conv([128, 256],256)
# 3*(5+num_classes) = 3*(5+20) = 3*(4+1+20)=75
self.yolo_head3 = yolo_head([256, len(anchors_mask[0]) * (5 + num_classes)],128)
self.down_sample1 = conv2d(128,256,3,stride=2)
self.make_five_conv3 = make_five_conv([256, 512],512)
# 3*(5+num_classes) = 3*(5+20) = 3*(4+1+20)=75
self.yolo_head2 = yolo_head([512, len(anchors_mask[1]) * (5 + num_classes)],256)
self.down_sample2 = conv2d(256,512,3,stride=2)
self.make_five_conv4 = make_five_conv([512, 1024],1024)
# 3*(5+num_classes)=3*(5+20)=3*(4+1+20)=75
self.yolo_head1 = yolo_head([1024, len(anchors_mask[2]) * (5 + num_classes)],512)