|
e9226be3ed
|
calc_anchors can be used without OpenCV, and it uses (1-IoU) instead of Euclidean distance
|
2018-12-11 15:59:23 +03:00 |
|
|
fb1ee79576
|
Minor fix
|
2018-12-10 02:22:45 +03:00 |
|
|
3256967119
|
Update Readme.md
|
2018-12-10 01:37:40 +03:00 |
|
|
a621235783
|
Switch to Tensor Cores after 2000 iterations.
|
2018-12-10 01:35:08 +03:00 |
|
|
89c6213af0
|
When is used -map flag for training - calculate mAP for each 4 Epochs.
|
2018-12-10 01:33:32 +03:00 |
|
|
ca3bf2fa92
|
Update Readme.md
|
2018-12-10 01:30:48 +03:00 |
|
|
89cb877689
|
Update Readme.md
|
2018-12-10 01:24:58 +03:00 |
|
|
7bd6b2f720
|
Compilation fix
|
2018-12-09 18:28:51 +03:00 |
|
|
dc7f8a32ae
|
mAP calculation during training, if is used flag -map
|
2018-12-09 18:18:47 +03:00 |
|
|
742bb7c7ce
|
Compile fix
|
2018-12-07 22:52:07 +03:00 |
|
|
7c2f302321
|
Fixed nan issue for training with CUDNN_HALF=1 by using Tensor Cores
|
2018-12-07 22:40:10 +03:00 |
|
|
21a4ec9390
|
Saving loss-chart for each 100 iterations automatically
|
2018-11-26 11:11:56 +03:00 |
|
|
9f7d7c58b5
|
Minor fixes. Use CUDA 10.0
|
2018-11-17 02:48:46 +03:00 |
|
|
14ed6fcb6e
|
Added -iou_thresh 0.05 flag for ./darknet detector map
|
2018-11-08 15:05:47 +03:00 |
|
|
25f65f6878
|
Added fast_binarize_weights_gpu()
|
2018-11-05 22:38:35 +03:00 |
|
|
2c5e383c04
|
Update Readme.md
|
2018-11-01 03:32:31 +03:00 |
|
|
08e8e0c8c2
|
Minor fix
|
2018-10-28 23:58:08 +03:00 |
|
|
31df5e2735
|
Update Readme.md
|
2018-10-25 21:39:27 +03:00 |
|
|
7f4b859514
|
Merge pull request #1816 from vinliao/master
Fix wrong coco.data path on darknet.py
|
2018-10-24 21:04:35 +03:00 |
|
|
f81b9209ff
|
fix wrong path
|
2018-10-24 12:47:19 +07:00 |
|
|
31ac46ba22
|
Fixed bug for 32-bit compilation without GPU.
|
2018-10-23 21:59:57 +03:00 |
|
|
d487bdf471
|
transpose 32x32 on GPU
|
2018-10-19 22:55:25 +03:00 |
|
|
9e2c894a32
|
Transpose on CPU fix
|
2018-10-19 16:31:55 +03:00 |
|
|
e7ad4e6c77
|
Compile fix
|
2018-10-19 02:44:54 +03:00 |
|
|
00282aa1b8
|
Fixed maxpool_layer by using cuDNN.
Added transpose 32 bit on CPU.
|
2018-10-19 02:40:20 +03:00 |
|
|
bdbbb2fd69
|
Minor fix
|
2018-10-18 17:30:49 +03:00 |
|
|
acf1489981
|
Fixed detector test for 1-channel images (black-white / infrared)
|
2018-10-18 16:01:29 +03:00 |
|
|
3b59178dec
|
Minor fix
|
2018-10-18 00:04:26 +03:00 |
|
|
76706dc4a6
|
Added yolov3-openimages.cfg and SELU activation
|
2018-10-15 20:16:08 +03:00 |
|
|
2bb44454a5
|
Minor fix
|
2018-10-14 00:20:23 +03:00 |
|
|
344f8c6d27
|
Fix building darknet.so on MacOS
|
2018-10-12 19:31:02 +03:00 |
|
|
7762372e5b
|
Compilation fix for OpenCV 3.4.x and Jetson Xavier.
|
2018-10-12 15:12:36 +03:00 |
|
|
7ee4135910
|
Fix OpenCV compilation. maxpool_layer on GPU uses cuDNN.
|
2018-10-10 13:51:36 +03:00 |
|
|
dbc75aec77
|
Merge pull request #1724 from eon129/master
Classifier now is working for CPU/GPU
|
2018-10-10 13:46:46 +03:00 |
|
|
38801979b2
|
Classifier now is working. The file softmax_layer.c was outdated
|
2018-10-06 14:12:02 -05:00 |
|
|
24b60456c7
|
Version 0.2.5 - fixed )
|
2018-10-03 23:47:31 +03:00 |
|
|
0cf56d1ccf
|
Version 0.2.5 )
|
2018-10-03 23:45:06 +03:00 |
|
|
6c8642b3eb
|
Added darknet53.cfg and resnext152-32x4d.cfg
|
2018-10-01 21:58:00 +03:00 |
|
|
c02216120b
|
Compile fix
|
2018-09-28 20:43:47 +03:00 |
|
|
5cc2a7e20d
|
Added mistakes checking
|
2018-09-28 20:27:30 +03:00 |
|
|
67f33ff5dc
|
Another compile fix
|
2018-09-27 23:49:03 +03:00 |
|
|
e92c635f35
|
Compile fix
|
2018-09-27 23:27:14 +03:00 |
|
|
c0e2512af2
|
Activation improvement, more robust timer.
|
2018-09-27 23:10:54 +03:00 |
|
|
e203ab9046
|
Minor fixes
|
2018-09-25 02:39:30 +03:00 |
|
|
129417ae7f
|
Update Readme.md
|
2018-09-24 02:06:35 +03:00 |
|
|
7dd97537fb
|
XNOR-net tiny-yolo_xnor.cfg ~2x faster than cuDNN on CUDA (nVidia GPU Maxwell)
|
2018-09-22 02:01:14 +03:00 |
|
|
0224ba3d0d
|
Compile fix
|
2018-09-17 23:59:29 +03:00 |
|
|
03e95320a1
|
XNOR coalesced memory access, and avoid bank conflicts
|
2018-09-17 23:39:25 +03:00 |
|
|
8e0b265067
|
Update Readme.md
|
2018-09-14 15:54:01 +03:00 |
|
|
b829fee30b
|
Update Readme.md
|
2018-09-14 15:29:52 +03:00 |
|