yolo_darknet_sur_un_gros_centre_de_calcul
Ceci est une ancienne révision du document !
Table des matières
Yolo Darknet sur un gros centre de calcul
Yolo v3 avec une bonne carte graphique Nvidia 1060 GTX
Ressources et documentation de YOLO Darknet
Darknet
Training
Idem à Yolo Darknet sur un portable Optimus mais avec 64000 images 704×704 au lieu de 1000 images 416*416
Yolo v3
Il faut une carte graphique avec 4Go de RAM Minimum:
export PATH=/usr/local/cuda-10.0/bin${PATH:+:${PATH}}
./darknet detector train cfg/obj.data cfg/yolo-obj.cfg darknet53.conv.74 -map
Plantage après quelques heures.
Yolo v3 tiny
TODO recopier les lignes de commandes et la modif cfg
[net] # Testing #batch=1 #subdivisions=1 # Training batch=64 subdivisions=64 width=704 height=704 channels=3 momentum=0.9 decay=0.0005 angle=0 saturation = 1.5 exposure = 1.5 hue=.1 learning_rate=0.001 burn_in=1000 max_batches = 500200 policy=steps steps=400000,450000 scales=.1,.1 [convolutional] batch_normalize=1 filters=16 size=3 stride=1 pad=1 activation=leaky [maxpool] size=2 stride=2 [convolutional] batch_normalize=1 filters=32 size=3 stride=1 pad=1 activation=leaky [maxpool] size=2 stride=2 [convolutional] batch_normalize=1 filters=64 size=3 stride=1 pad=1 activation=leaky [maxpool] size=2 stride=2 [convolutional] batch_normalize=1 filters=128 size=3 stride=1 pad=1 activation=leaky [maxpool] size=2 stride=2 [convolutional] batch_normalize=1 filters=256 size=3 stride=1 pad=1 activation=leaky [maxpool] size=2 stride=2 [convolutional] batch_normalize=1 filters=512 size=3 stride=1 pad=1 activation=leaky [maxpool] size=2 stride=1 [convolutional] batch_normalize=1 filters=1024 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=96 activation=linear [yolo] mask = 3,4,5 anchors = 10,14, 23,27, 37,58, 81,82, 135,169, 344,319 classes=27 num=6 jitter=.3 ignore_thresh = .7 truth_thresh = 1 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, 8 [convolutional] batch_normalize=1 filters=256 size=3 stride=1 pad=1 activation=leaky [convolutional] size=1 stride=1 pad=1 filters=96 activation=linear [yolo] mask = 0,1,2 anchors = 10,14, 23,27, 37,58, 81,82, 135,169, 344,319 classes=27 num=6 jitter=.3 ignore_thresh = .7 truth_thresh = 1 random=1
export PATH=/usr/local/cuda-10.0/bin${PATH:+:${PATH}}
./darknet partial cfg/yolov3-tiny.cfg yolov3-tiny.weights yolov3-tiny.conv.15 15
./darknet detector train cfg/obj.data cfg/yolov3-tiny_obj_labo.cfg yolov3-tiny.conv.15
yolo_darknet_sur_un_gros_centre_de_calcul.1554558945.txt.gz · Dernière modification : 2019/04/06 13:55 de serge