AlexeyAB / darknet

YOLOv4 / Scaled-YOLOv4 / YOLO - Neural Networks for Object Detection (Windows and Linux version of Darknet )

http://pjreddie.com/darknet/

Other

21.6k stars 7.95k forks source link

The Loss value does not decrease #4481

Open Look4-you opened 4 years ago

Look4-you commented 4 years ago

@AlexeyAB Hi. When I use Gaussian_yolov3_BDD.cfg to train my own dataset. When the Loss value reached about 780, then it won't decrease but float around 780 for a long time as I observed.

Here is the chart.png chart

Here is my Gaussian_yolov3_BDD.cfg: [net]

Testing

batch=1

subdivisions=1

Training

batch=64 subdivisions=32 width=640 height=480 channels=3 momentum=0.9 decay=0.0005 angle=0 saturation = 1.5 exposure = 1.5 hue=.1

learning_rate=0.0001 burn_in=1000 max_batches = 12000 policy=steps steps=8000,10000 scales=.1,.1 max_epochs = 300

[convolutional] batch_normalize=1 filters=32 size=3 stride=1 pad=1 activation=leaky

Downsample

[convolutional] batch_normalize=1 filters=64 size=3 stride=2 pad=1 activation=leaky

[convolutional] batch_normalize=1 filters=32 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 filters=64 size=3 stride=1 pad=1 activation=leaky

[shortcut] from=-3 activation=linear

Downsample

[convolutional] batch_normalize=1 filters=128 size=3 stride=2 pad=1 activation=leaky

[convolutional] batch_normalize=1 filters=64 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 filters=128 size=3 stride=1 pad=1 activation=leaky

[shortcut] from=-3 activation=linear

[convolutional] batch_normalize=1 filters=64 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 filters=128 size=3 stride=1 pad=1 activation=leaky

[shortcut] from=-3 activation=linear

Downsample

[convolutional] batch_normalize=1 filters=256 size=3 stride=2 pad=1 activation=leaky

[convolutional] batch_normalize=1 filters=128 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 filters=256 size=3 stride=1 pad=1 activation=leaky

[shortcut] from=-3 activation=linear

[convolutional] batch_normalize=1 filters=128 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 filters=256 size=3 stride=1 pad=1 activation=leaky

[shortcut] from=-3 activation=linear

[convolutional] batch_normalize=1 filters=128 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 filters=256 size=3 stride=1 pad=1 activation=leaky

[shortcut] from=-3 activation=linear

[convolutional] batch_normalize=1 filters=128 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 filters=256 size=3 stride=1 pad=1 activation=leaky

[shortcut] from=-3 activation=linear

[convolutional] batch_normalize=1 filters=128 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 filters=256 size=3 stride=1 pad=1 activation=leaky

[shortcut] from=-3 activation=linear

[convolutional] batch_normalize=1 filters=128 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 filters=256 size=3 stride=1 pad=1 activation=leaky

[shortcut] from=-3 activation=linear

[convolutional] batch_normalize=1 filters=128 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 filters=256 size=3 stride=1 pad=1 activation=leaky

[shortcut] from=-3 activation=linear

[convolutional] batch_normalize=1 filters=128 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 filters=256 size=3 stride=1 pad=1 activation=leaky

[shortcut] from=-3 activation=linear

Downsample

[convolutional] batch_normalize=1 filters=512 size=3 stride=2 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

[shortcut] from=-3 activation=linear

[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

[shortcut] from=-3 activation=linear

[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

[shortcut] from=-3 activation=linear

[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

[shortcut] from=-3 activation=linear

[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

[shortcut] from=-3 activation=linear

[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

[shortcut] from=-3 activation=linear

[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

[shortcut] from=-3 activation=linear

[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

[shortcut] from=-3 activation=linear

Downsample

[convolutional] batch_normalize=1 filters=1024 size=3 stride=2 pad=1 activation=leaky

[convolutional] batch_normalize=1 filters=512 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 filters=1024 size=3 stride=1 pad=1 activation=leaky

[shortcut] from=-3 activation=linear

[convolutional] batch_normalize=1 filters=512 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 filters=1024 size=3 stride=1 pad=1 activation=leaky

[shortcut] from=-3 activation=linear

[convolutional] batch_normalize=1 filters=512 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 filters=1024 size=3 stride=1 pad=1 activation=leaky

[shortcut] from=-3 activation=linear

[convolutional] batch_normalize=1 filters=512 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 filters=1024 size=3 stride=1 pad=1 activation=leaky

[shortcut] from=-3 activation=linear

######################

[convolutional] batch_normalize=1 filters=512 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 size=3 stride=1 pad=1 filters=1024 activation=leaky

[convolutional] batch_normalize=1 filters=512 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 size=3 stride=1 pad=1 filters=1024 activation=leaky

[convolutional] batch_normalize=1 filters=512 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 size=3 stride=1 pad=1 filters=1024 activation=leaky

[convolutional] size=1 stride=1 pad=1 filters=30 activation=linear

[Gaussian_yolo] mask = 6,7,8 anchors = 7,10, 14,24, 27,43, 32,97, 57,64, 92,109, 73,175, 141,178, 144,291 classes=1 num=9 jitter=.3 ignore_thresh = .5 truth_thresh = 1 iou_thresh=0.213 uc_normalizer=1.0 cls_normalizer=1.0 iou_normalizer=0.5 iou_loss=giou scale_x_y=1.0 random=1

[route] layers = -4

[convolutional] batch_normalize=1 filters=256 size=1 stride=1 pad=1 activation=leaky

[upsample] stride=2

[route] layers = -1, 61

[convolutional] batch_normalize=1 filters=256 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 size=3 stride=1 pad=1 filters=512 activation=leaky

[convolutional] batch_normalize=1 filters=256 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 size=3 stride=1 pad=1 filters=512 activation=leaky

[convolutional] batch_normalize=1 filters=256 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 size=3 stride=1 pad=1 filters=512 activation=leaky

[convolutional] size=1 stride=1 pad=1 filters=30 activation=linear

[Gaussian_yolo] mask = 3,4,5 anchors = 7,10, 14,24, 27,43, 32,97, 57,64, 92,109, 73,175, 141,178, 144,291 classes=1 num=9 jitter=.3 ignore_thresh = .5 truth_thresh = 1 iou_thresh=0.213 uc_normalizer=1.0 cls_normalizer=1.0 iou_normalizer=0.5 iou_loss=giou scale_x_y=1.0 random=1

[route] layers = -4

[convolutional] batch_normalize=1 filters=128 size=1 stride=1 pad=1 activation=leaky

[upsample] stride=2

[route] layers = -1, 36

[convolutional] batch_normalize=1 filters=128 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 size=3 stride=1 pad=1 filters=256 activation=leaky

[convolutional] batch_normalize=1 filters=128 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 size=3 stride=1 pad=1 filters=256 activation=leaky

[convolutional] batch_normalize=1 filters=128 size=1 stride=1 pad=1 activation=leaky

[convolutional] batch_normalize=1 size=3 stride=1 pad=1 filters=256 activation=leaky

[convolutional] size=1 stride=1 pad=1 filters=30 activation=linear

[Gaussian_yolo] mask = 0,1,2 anchors = 7,10, 14,24, 27,43, 32,97, 57,64, 92,109, 73,175, 141,178, 144,291 classes=1 num=9 jitter=.3 ignore_thresh = .5 truth_thresh = 1 iou_thresh=0.213 uc_normalizer=1.0 cls_normalizer=1.0 iou_normalizer=0.5 iou_loss=giou scale_x_y=1.0 random=1

Do you have any suggestions for this problem? Thanks in advance. By the way , I just try to use Gaussian+yolov3+Giou on my own dataset.

AlexeyAB commented 4 years ago

There is no problem. mAP increases, training goes well.

Look4-you commented 4 years ago

There is no problem. mAP increases, training goes well.

Maybe I should wait for more iterations? Should I need to change the parameter: iou_normalizer=0.07, uc_normalizer=0.07 according to #4430 ??

AlexeyAB commented 4 years ago

Maybe I should wait for more iterations?

Loss will still high, this is normal for Gaussian_yolo

Should I need to change the parameter: iou_normalizer=0.07, uc_normalizer=0.07 according to #4430 ??

You can try, but this is just an experimental suggestion.