Seioch
commented
3 years ago Looking at it a bit closer, there are only two output bins for yolov4-tiny, but for some reason tk::dnn::darknetParser returns a tk::dnn::Network with 3 outputs. No idea why this is happening.
I recently trained a yolov4-tiny model, exported the weights using the darknet utility, and prepared to convert the model to TensorRT, but I keep getting this error:
Any ideas?
CFG:
` [net] batch=128 subdivisions=32 width=416 height=416 channels=3 momentum=0.9 decay=0.0005 angle=0 saturation = 1.5 exposure = 1.5 hue=.1
learning_rate=0.00261 burn_in=1000 max_batches = 6000 policy=steps steps=4000,5000 scales=.1,.1
[convolutional] batch_normalize=1 filters=32 size=3 stride=2 pad=1 activation=leaky
[convolutional] batch_normalize=1 filters=64 size=3 stride=2 pad=1 activation=leaky
[convolutional] batch_normalize=1 filters=64 size=3 stride=1 pad=1 activation=leaky
[route] layers=-1 groups=2 group_id=1
[convolutional] batch_normalize=1 filters=32 size=3 stride=1 pad=1 activation=leaky
[convolutional] batch_normalize=1 filters=32 size=3 stride=1 pad=1 activation=leaky
[route] layers = -1,-2
[convolutional] batch_normalize=1 filters=64 size=1 stride=1 pad=1 activation=leaky
[route] layers = -6,-1
[maxpool] size=2 stride=2
[convolutional] batch_normalize=1 filters=128 size=3 stride=1 pad=1 activation=leaky
[route] layers=-1 groups=2 group_id=1
[convolutional] batch_normalize=1 filters=64 size=3 stride=1 pad=1 activation=leaky
[convolutional] batch_normalize=1 filters=64 size=3 stride=1 pad=1 activation=leaky
[route] layers = -1,-2
[convolutional] batch_normalize=1 filters=128 size=1 stride=1 pad=1 activation=leaky
[route] layers = -6,-1
[maxpool] size=2 stride=2
[convolutional] batch_normalize=1 filters=256 size=3 stride=1 pad=1 activation=leaky
[route] layers=-1 groups=2 group_id=1
[convolutional] batch_normalize=1 filters=128 size=3 stride=1 pad=1 activation=leaky
[convolutional] batch_normalize=1 filters=128 size=3 stride=1 pad=1 activation=leaky
[route] layers = -1,-2
[convolutional] batch_normalize=1 filters=256 size=1 stride=1 pad=1 activation=leaky
[route] layers = -6,-1
[maxpool] size=2 stride=2
[convolutional] batch_normalize=1 filters=512 size=3 stride=1 pad=1 activation=leaky
##################################
[convolutional] batch_normalize=1 filters=256 size=1 stride=1 pad=1 activation=leaky
[convolutional] batch_normalize=1 filters=512 size=3 stride=1 pad=1 activation=leaky
[convolutional] size=1 stride=1 pad=1 filters=129 activation=linear
[yolo] mask = 3,4,5 anchors=11, 19, 21, 16, 17, 24, 54, 18, 100, 10, 25, 47, 79, 25, 185, 37, 146,159 classes=38 num=9 jitter=.3 scale_x_y = 1.05 cls_normalizer=1.0 iou_normalizer=0.07 iou_loss=ciou ignore_thresh = .7 truth_thresh = 1 random=0 resize=1.5 nms_kind=greedynms beta_nms=0.6
[route] layers = -4
[convolutional] batch_normalize=1 filters=128 size=1 stride=1 pad=1 activation=leaky
[upsample] stride=2
[route] layers = -1, 23
[convolutional] batch_normalize=1 filters=256 size=3 stride=1 pad=1 activation=leaky
[convolutional] size=1 stride=1 pad=1 filters=129 activation=linear
[yolo] mask = 0,1,2 anchors=11, 19, 21, 16, 17, 24, 54, 18, 100, 10, 25, 47, 79, 25, 185, 37, 146,159 classes=38 num=9 jitter=.3 scale_x_y = 1.05 cls_normalizer=1.0 iou_normalizer=0.07 iou_loss=ciou ignore_thresh = .7 truth_thresh = 1 random=0 resize=1.5 nms_kind=greedynms beta_nms=0.6`