Tag: Branch: Tree: 2a873f3448

master

Yolo_v2

Yolo_v2_tested

Yolo_v3

darknet_yolo_v3

darknet_yolo_v3_optimal

darknet_yolo_v4_pre

yolov4

Branches Tags

${ item.name }

Create tag ${ searchTerm }

Create branch ${ searchTerm }

from '2a873f3448'

${ noResults }

Commit Graph

66 Commits (2a873f34485c75d44a346f92ba7dcf2e2aa57a15)

Author	SHA1	Message	Date
AlexeyAB	2a873f3448	Added label_smooth_eps=0.1 for [net] layer for Label Smoothing for Classifier	6 years ago
AlexeyAB	ebf1d7a73b	Minor fixes	6 years ago
AlexeyAB	2207acd9c4	Minor fix	6 years ago
AlexeyAB	142fcdeb1e	Added blur=1 for Classifier training if OPENCV=1 (removes textures like in stylized-ImageNet)	6 years ago
AlexeyAB	a95bc3f5b6	Fixed memory leak for ./darknet classifier predict ... < list.txt	6 years ago
AlexeyAB	7ae664f96d	improved accuracy of inference-time and FPS measurement for Classifier	6 years ago
AlexeyAB	4fcb1b8e42	Addex cutmix=1 and fixed mixup=1 for Classifier	6 years ago
AlexeyAB	9d96277a37	Added mixup=1 for Classifier training	6 years ago
AlexeyAB	3555beb914	Fixed some memory leaks in secondary functions	6 years ago
AlexeyAB	11142d00be	Fixed non-square network for Training Classifier	6 years ago
AlexeyAB	c516b6cb0a	Take TopK from obj.data file for Classifier	6 years ago
vinjn	ebb9a3cf53	Make "darknet.exe classifier predict" behavior same as "darknet.exe classifer validate"	6 years ago
AlexeyAB	cc41339805	Added [scale_channels] layer for squeeze-and-excitation blocks	6 years ago
AlexeyAB	97679224ba	All OpenCV C-functions replaced by CPP-functions	6 years ago
AlexeyAB	79c669707c	Some C OpenCV functions replaced by CPP functions	6 years ago
AlexeyAB	f19d1440cb	Moved all OpenCV functions to the image_opencv.cpp/h	6 years ago
AlexeyAB	cad99fd75d	Fixed memory leak for Training with -map flag and random=1	6 years ago
AlexeyAB	b3254ed523	Fixed many warnings	6 years ago
AlexeyAB	b6e15f1656	ZED 3D Camera support added to ./uselib (yolo_console_cpp.exe) example	6 years ago
Steve Zaretti	17fed7c4e2	Segfault if height > width ... ./darknet classifier train data/sort.data cfg/alexnet.cfg -dont_show alexnet 1 compute_capability = 750, cudnn_half = 1 layer filters size input output 0 conv 96 11 x11 / 4 128 x 256 x 3 -> 30 x 62 x 96 0.130 BF 1 max 3 x 3 / 2 30 x 62 x 96 -> 14 x 30 x 96 0.000 BF 2 conv 256 5 x 5 / 1 14 x 30 x 96 -> 14 x 30 x 256 0.516 BF 3 max 3 x 3 / 2 14 x 30 x 256 -> 6 x 14 x 256 0.000 BF 4 conv 384 3 x 3 / 1 6 x 14 x 256 -> 6 x 14 x 384 0.149 BF 5 conv 384 3 x 3 / 1 6 x 14 x 384 -> 6 x 14 x 384 0.223 BF 6 conv 256 3 x 3 / 1 6 x 14 x 384 -> 6 x 14 x 256 0.149 BF 7 max 3 x 3 / 2 6 x 14 x 256 -> 2 x 6 x 256 0.000 BF 8 connected 3072 -> 4096 9 dropout p = 0.50 4096 -> 4096 10 connected 4096 -> 4096 11 dropout p = 0.50 4096 -> 4096 12 connected 4096 -> 1261 13 softmax 1261 14 cost 1261 Total BFLOPS 1.167 Allocate additional workspace_size = 1940.23 MB Learning Rate: 0.01, Momentum: 0.9, Decay: 0.0005 893253 Loaded: 0.205056 seconds Segmentation fault (core dumped) This PR fix this issue: ./darknet classifier train data/sort.data cfg/alexnet.cfg -dont_show alexnet 1 compute_capability = 750, cudnn_half = 1 layer filters size input output 0 conv 96 11 x11 / 4 128 x 256 x 3 -> 30 x 62 x 96 0.130 BF 1 max 3 x 3 / 2 30 x 62 x 96 -> 14 x 30 x 96 0.000 BF 2 conv 256 5 x 5 / 1 14 x 30 x 96 -> 14 x 30 x 256 0.516 BF 3 max 3 x 3 / 2 14 x 30 x 256 -> 6 x 14 x 256 0.000 BF 4 conv 384 3 x 3 / 1 6 x 14 x 256 -> 6 x 14 x 384 0.149 BF 5 conv 384 3 x 3 / 1 6 x 14 x 384 -> 6 x 14 x 384 0.223 BF 6 conv 256 3 x 3 / 1 6 x 14 x 384 -> 6 x 14 x 256 0.149 BF 7 max 3 x 3 / 2 6 x 14 x 256 -> 2 x 6 x 256 0.000 BF 8 connected 3072 -> 4096 9 dropout p = 0.50 4096 -> 4096 10 connected 4096 -> 4096 11 dropout p = 0.50 4096 -> 4096 12 connected 4096 -> 1261 13 softmax 1261 14 cost 1261 Total BFLOPS 1.167 Allocate additional workspace_size = 1940.23 MB Learning Rate: 0.01, Momentum: 0.9, Decay: 0.0005 893253 Loaded: 0.509795 seconds 1, 0.000: 4.070450, 4.070450 avg, 0.010000 rate, 0.310202 seconds, 128 images Loaded: 0.416345 seconds 2, 0.000: 4.069521, 4.070357 avg, 0.010000 rate, 0.055622 seconds, 256 images Loaded: 0.451087 seconds I'll tell you later if training is converging.	6 years ago
AlexeyAB	5e850c2489	Minor fixes	6 years ago
Stefano Sinigardi	6e7c7976d7	move as many defs as possible away from darknet.h	6 years ago
Stefano Sinigardi	b3579380dc	improve compatibility with c++ compilers, prepare for CMake	6 years ago
David Smith	ba1f1363c0	fix assorted compile warnings	6 years ago
AlexeyAB	98103552fb	Minor fix	6 years ago
AlexeyAB	6c28da5def	Draw top5 accuracy on the Loss-chart for training Classifier	6 years ago
AlexeyAB	9bb7455a0e	Minor fix	6 years ago
AlexeyAB	7587d47c46	Partial fixed	6 years ago
AlexeyAB	0543278a5b	Partial fixed	6 years ago
AlexeyAB	110b5240a4	Fixed LSTM-layer	6 years ago
AlexeyAB	70260cda24	Added -mjpeg_port 8090 for Training. Loss-chart saved to chart.png even when used -dont_show. ... mAP is calculated during training even without OpenCV.	6 years ago
AlexeyAB	f0bed8ef73	Fixed drawing mAP	7 years ago
AlexeyAB	dc7f8a32ae	mAP calculation during training, if is used flag -map	7 years ago
vinjn	b36512ea2e	classifier.c - add the awesome training chart and make sure "top" is not bigger than "classes" in datacfg file.	7 years ago
vinjn	6b6f0f0ca3	classifier.c - set batch=1 in "predict", "demo" and "try" mode.	7 years ago
AlexeyAB	f52040f0bf	Classifier uses C++ OpenCV cv::VideoCapture that hasn't bugs.	7 years ago
iovodov	9207607a59	Fix issue #674: parameter "flip" is added to [net] parameters for classifier and detector ... flip=0 switches flip augmentation off. Default is flip=1 which is consistent with a previous behaviour.	7 years ago
AlexeyAB	35a66b73a2	Changed class to class_id	7 years ago
AlexeyAB	2baa7bde54	Fixed classifier for AlexNet and Resnet50	8 years ago
AlexeyAB	815e7a127b	Supported OpenCV 3.0 and 2.4.13. Supported Windows and Linux.	8 years ago
AlexeyAB	5b6dd3a07c	Sources fixed - Yolo v2 can be used on Windows (MSVS 2015 + CUDA 8.0)	9 years ago
Joseph Redmon	9a01e6ccb7	🔥 crush. crush. admit. 🔥	9 years ago
Joseph Redmon	c7a700dc22	new font strategy	9 years ago
Joseph Redmon	352ae7e65b	ADAM	9 years ago
Joseph Redmon	d8adaf8ea6	tree stuff	9 years ago
Joseph Redmon	481b57a96a	So I have this new programming paradigm.......	9 years ago
Joseph Redmon	73f7aacf35	better multigpu	9 years ago
Joseph Redmon	5c067dc447	good chance I didn't break anything	9 years ago
Joseph Redmon	6b38dcdce0	adding new tiny-yolo	9 years ago
Joseph Redmon	8f1b4e0962	updates and things	9 years ago