Stefano Sinigardi
|
2c4c96f50c
|
fix regression
|
2019-07-12 17:35:38 +02:00 |
Stefano Sinigardi
|
15a19c4e83
|
avoid 'command line is too long' error on windows
|
2019-07-11 18:55:24 +02:00 |
Stefano Sinigardi
|
e299925c5c
|
re-enable cuda choice
|
2019-07-09 10:39:48 +02:00 |
AlexeyAB
|
5a6afe96d3
|
Fixed scale_channels_layer resize
|
2019-07-03 13:54:13 +03:00 |
AlexeyAB
|
54e2d0b0e8
|
Added ability to use letter_box resizing in darknet.py sample
|
2019-07-01 19:07:13 +03:00 |
AlexeyAB
|
d4402d29c2
|
Added maxpool_depth= and out_channels= params to [maxpool]
|
2019-06-30 13:53:11 +03:00 |
AlexeyAB
|
7d4143332d
|
Minor output fixes
|
2019-06-29 16:52:49 +03:00 |
AlexeyAB
|
88cccfcad4
|
Added Swish-activation
|
2019-06-26 15:48:22 +03:00 |
AlexeyAB
|
5ec35922d5
|
Fixed yolo_cpp_dll.vcxproj
|
2019-06-22 19:17:41 +03:00 |
AlexeyAB
|
c9129c2078
|
Added letter_box=1 param in [net] section (cfg-file) for keeping aspect ratio during training
|
2019-06-22 17:04:00 +03:00 |
AlexeyAB
|
1ed71f4b29
|
Fixed CUDA-implementation of back-propagation for [scale_channels]-layer
|
2019-06-22 17:03:06 +03:00 |
AlexeyAB
|
aeb9da6918
|
Fixed darknet_no_gpu.vcxproj and yolo_cpp_dll_no_gpu.vcxproj (added scale_channels_layer)
|
2019-06-22 01:37:33 +03:00 |
AlexeyAB
|
9e9b2c4936
|
Update Readme.md
|
2019-06-20 20:52:26 +03:00 |
AlexeyAB
|
cc41339805
|
Added [scale_channels] layer for squeeze-and-excitation blocks
|
2019-06-19 23:36:02 +03:00 |
AlexeyAB
|
8c80ba6b53
|
Fixed weights-size for Depthwise grouped convolutional
|
2019-06-15 16:41:22 +03:00 |
AlexeyAB
|
f0582446f2
|
Fixed Depth-wise (grouped) convolution for CPU and GPU (CUDNN=0/1)
|
2019-06-15 02:32:57 +03:00 |
Alexey
|
a1abd07e23
|
Merge pull request #3371 from cenit/dev/cenit/fixes
fixes for clang/mac
|
2019-06-13 14:45:16 +03:00 |
AlexeyAB
|
bd689f91ac
|
Added TridentNet (state-of-art on MS COCO)
|
2019-06-13 14:43:23 +03:00 |
Stefano Sinigardi
|
c0a9c741c3
|
clean up of README about deprecated option, invert logic of vcpkg flag in build.ps1 for improved clarity
|
2019-06-13 06:54:54 +02:00 |
Stefano Sinigardi
|
b78db932cd
|
fix missing pthread link for downstream users
|
2019-06-12 17:17:40 +02:00 |
Stefano Sinigardi
|
1b7a0dca09
|
revert #3389
|
2019-06-12 16:45:48 +02:00 |
Stefano Sinigardi
|
7701a8cf54
|
Merge branch 'dev/alexey/master' into dev/cenit/fixes
|
2019-06-12 16:44:30 +02:00 |
Alexey
|
e4f6b276e5
|
Merge pull request #3389 from adujardin/master
Fix Ubuntu cmake compilation
|
2019-06-12 12:09:30 +03:00 |
Aymeric DUJARDIN
|
f1f45b295c
|
Fix Ubuntu cmake compilation (nvcc fatal fopenmp + pthread link uselib)
|
2019-06-12 09:23:57 +02:00 |
Stefano Sinigardi
|
f14c6e2fd9
|
fully restore opencv option which got lost somehow
|
2019-06-10 10:56:23 +02:00 |
Stefano Sinigardi
|
f1b6a6e008
|
fixes for clang/mac
|
2019-06-10 10:33:20 +02:00 |
AlexeyAB
|
378d49e1c3
|
Bug fix for MS COCO json output
|
2019-06-09 18:46:07 +03:00 |
AlexeyAB
|
94c806ffad
|
final fix
|
2019-06-09 03:07:04 +03:00 |
AlexeyAB
|
85b0722af7
|
Another compile fix
|
2019-06-09 01:15:24 +03:00 |
AlexeyAB
|
d1be8ddb33
|
compile fix
|
2019-06-09 00:19:44 +03:00 |
AlexeyAB
|
88ce9dcca6
|
Added dilation parameter for convolutional, conv_lstm and crnn layers
|
2019-06-08 19:41:44 +03:00 |
AlexeyAB
|
12db38ccbf
|
Fixed input_shortcut_gpu() for layers with different numbers of channels
|
2019-06-08 17:42:53 +03:00 |
AlexeyAB
|
3ff27b1cdd
|
Mixup bug fix
|
2019-06-07 14:17:43 +03:00 |
AlexeyAB
|
57baf1a487
|
Added Mixup data augmentation if is used mixup=1
|
2019-06-06 02:55:11 +03:00 |
Alexey
|
2347913ef1
|
Merge pull request #2952 from cenit/dev/cenit/include
[headers] remove many installed files which should remain private, add more fixes for downstream projects
|
2019-06-05 17:37:28 +03:00 |
AlexeyAB
|
ff0dbcbb7d
|
Fix network resizing for resize_conv_lstm_layer()
|
2019-06-05 15:51:03 +03:00 |
AlexeyAB
|
abba310982
|
Minor fix in cost_layer
|
2019-06-05 01:04:16 +03:00 |
AlexeyAB
|
1f6f2e1162
|
Compile fix
|
2019-06-04 21:46:11 +03:00 |
AlexeyAB
|
eac26226a7
|
Added scale_x_y param to [yolo]-layer (for sigmoid).
In previous commit is used GIoU(generalized-iou) from https://github.com/generalized-iou/g-darknet
|
2019-06-04 21:20:59 +03:00 |
AlexeyAB
|
6e13527f06
|
Added GIoU
|
2019-06-01 01:27:18 +03:00 |
AlexeyAB
|
55dcd1bcb8
|
Compile fix
|
2019-05-27 22:10:07 +03:00 |
AlexeyAB
|
24f7d94ab4
|
Fixed resizing REORG_OLD layer
|
2019-05-27 20:55:42 +03:00 |
Alexey
|
2b7fea761f
|
Merge pull request #2923 from willbattel/save-best-map
Save weights with best mAP during training
|
2019-05-27 19:02:49 +03:00 |
Stefano Sinigardi
|
a3d5cc4664
|
Merge branch 'master' into dev/cenit/include
|
2019-05-25 20:27:19 +02:00 |
AlexeyAB
|
3aca0b7166
|
conv-LSTM training speedup
|
2019-05-24 18:48:11 +03:00 |
AlexeyAB
|
9c9344a1ff
|
Added flag -letter_box for Detection
|
2019-05-21 23:47:47 +03:00 |
AlexeyAB
|
034243d1ff
|
Compile fix
|
2019-05-21 18:06:43 +03:00 |
AlexeyAB
|
0109a8dda9
|
Some conv-lstm improvements (inference speedup)
|
2019-05-21 18:01:08 +03:00 |
AlexeyAB
|
b9ea49af25
|
Fixed reorg layer is [reorg3d], old reorg layer is [reorg]
|
2019-05-20 23:58:55 +03:00 |
AlexeyAB
|
ae3ce1b181
|
Minor fixes
|
2019-05-20 22:37:42 +03:00 |