| .. |
|
.editorconfig
|
…
|
|
|
activation_kernels.cu
|
Fixed mish gradient
|
5 years ago |
|
activation_layer.c
|
fix calloc realloc failure
|
6 years ago |
|
activation_layer.h
|
…
|
|
|
activations.c
|
Minor fix
|
5 years ago |
|
activations.h
|
Added GELU activation
|
5 years ago |
|
art.c
|
…
|
|
|
avgpool_layer.c
|
fix calloc realloc failure
|
6 years ago |
|
avgpool_layer.h
|
…
|
|
|
avgpool_layer_kernels.cu
|
…
|
|
|
batchnorm_layer.c
|
self-adversarial training
|
5 years ago |
|
batchnorm_layer.h
|
final compile fix
|
5 years ago |
|
blas.c
|
Added Supervised and Unsupervised Contrastive loss for Classifier [net] contrastive=1 unsupervised=1 [contrastive] classes=1000
|
5 years ago |
|
blas.h
|
Added Contrastive loss
|
5 years ago |
|
blas_kernels.cu
|
Added [conv_lstm] lstm_activation=tanh shortcut=1
|
5 years ago |
|
box.c
|
Added param to [yolo] iou_thresh_kind=giou iou_thresh=0.213
|
5 years ago |
|
box.h
|
Added param to [yolo] iou_thresh_kind=giou iou_thresh=0.213
|
5 years ago |
|
captcha.c
|
…
|
|
|
cifar.c
|
…
|
|
|
classifier.c
|
Added Supervised and Unsupervised Contrastive loss for Classifier [net] contrastive=1 unsupervised=1 [contrastive] classes=1000
|
5 years ago |
|
classifier.h
|
…
|
|
|
coco.c
|
Added flag averaging detections: ./darknet detector demo ... -avgframes 3
|
5 years ago |
|
col2im.c
|
option to build project using cpp compiler if c not working/available
|
6 years ago |
|
col2im.h
|
…
|
|
|
col2im_kernels.cu
|
…
|
|
|
compare.c
|
fix calloc realloc failure
|
6 years ago |
|
connected_layer.c
|
Added loss_scale for Mixed-Precision on Tensor-Cores
|
5 years ago |
|
connected_layer.h
|
Compile fix
|
5 years ago |
|
conv_lstm_layer.c
|
Added Contrastive loss
|
5 years ago |
|
conv_lstm_layer.h
|
Added [conv_lstm] bottleneck=1 - to make the layer cheaper
|
5 years ago |
|
convolutional_kernels.cu
|
Fixed gradient_centralization for Grouped-Conv
|
5 years ago |
|
convolutional_layer.c
|
Minor speedup of Mish/Swish (if train==0 then don't write input of activation to the memory)
|
5 years ago |
|
convolutional_layer.h
|
Added loss_scale for Mixed-Precision on Tensor-Cores
|
5 years ago |
|
cost_layer.c
|
fix calloc realloc failure
|
6 years ago |
|
cost_layer.h
|
fix prototype
|
6 years ago |
|
cpu_gemm.c
|
…
|
|
|
crnn_layer.c
|
Added loss_scale for Mixed-Precision on Tensor-Cores
|
5 years ago |
|
crnn_layer.h
|
final compile fix
|
5 years ago |
|
crop_layer.c
|
fix calloc realloc failure
|
6 years ago |
|
crop_layer.h
|
…
|
|
|
crop_layer_kernels.cu
|
…
|
|
|
dark_cuda.c
|
Minor fix
|
5 years ago |
|
dark_cuda.h
|
Minor fix
|
5 years ago |
|
darknet.c
|
Added param [yolo] objectness_smooth=1
|
5 years ago |
|
darkunistd.h
|
Fix compilation for Windows with MinGW
|
6 years ago |
|
data.c
|
Added Supervised and Unsupervised Contrastive loss for Classifier [net] contrastive=1 unsupervised=1 [contrastive] classes=1000
|
5 years ago |
|
data.h
|
Added Contrastive loss
|
5 years ago |
|
deconvolutional_kernels.cu
|
…
|
|
|
deconvolutional_layer.c
|
fix calloc realloc failure
|
6 years ago |
|
deconvolutional_layer.h
|
…
|
|
|
demo.c
|
Added flag averaging detections: ./darknet detector demo ... -avgframes 3
|
5 years ago |
|
demo.h
|
Added flag averaging detections: ./darknet detector demo ... -avgframes 3
|
5 years ago |
|
detection_layer.c
|
fix calloc realloc failure
|
6 years ago |
|
detection_layer.h
|
…
|
|
|
detector.c
|
Added Contrastive loss
|
5 years ago |
|
dice.c
|
…
|
|
|
dropout_layer.c
|
Added [net] dynamic_minibatch=1 for increasing mini_batch_size when random=1 is used
|
5 years ago |
|
dropout_layer.h
|
Fixed DropBlock [dropout] dropblock_size_abs=7 dropblock_size_rel=0.5
|
6 years ago |
|
dropout_layer_kernels.cu
|
Fixed DropBlock
|
5 years ago |
|
gaussian_yolo_layer.c
|
Fixed [yolo] objectness_smooth=1 and lowest w/h of objectes when mosaic=1
|
5 years ago |
|
gaussian_yolo_layer.h
|
Fixed [Gaussian_yolo] layer (tested for training and detection)
|
6 years ago |
|
gemm.c
|
add mingw and cuda on github actions, restored cuda 10.2 on appveyor as a test
|
5 years ago |
|
gemm.h
|
…
|
|
|
getopt.c
|
add mingw and cuda on github actions, restored cuda 10.2 on appveyor as a test
|
5 years ago |
|
getopt.h
|
add mingw and cuda on github actions, restored cuda 10.2 on appveyor as a test
|
5 years ago |
|
gettimeofday.c
|
add mingw and cuda on github actions, restored cuda 10.2 on appveyor as a test
|
5 years ago |
|
gettimeofday.h
|
add mingw and cuda on github actions, restored cuda 10.2 on appveyor as a test
|
5 years ago |
|
go.c
|
Merge branch 'master' into master
|
5 years ago |
|
gru_layer.c
|
Added loss_scale for Mixed-Precision on Tensor-Cores
|
5 years ago |
|
gru_layer.h
|
final compile fix
|
5 years ago |
|
http_stream.cpp
|
Minor fix
|
5 years ago |
|
http_stream.h
|
Minor fix
|
5 years ago |
|
httplib.h
|
Added HTTP POST sender with detected objects, use: ./darknet detector demo ... -http_post_host webhook.site/898bbd9b-0ddd-49cf-b81d-1f56be98d870
|
6 years ago |
|
im2col.c
|
…
|
|
|
im2col.h
|
Added dilation parameter for convolutional, conv_lstm and crnn layers
|
6 years ago |
|
im2col_kernels.cu
|
…
|
|
|
image.c
|
Minor fix
|
5 years ago |
|
image.h
|
Minor fix
|
5 years ago |
|
image_opencv.cpp
|
Minor fix
|
5 years ago |
|
image_opencv.h
|
adversarial attack - minor improvement
|
5 years ago |
|
layer.c
|
Added Contrastive loss
|
5 years ago |
|
layer.h
|
…
|
|
|
list.c
|
Merge branch 'master' into master
|
5 years ago |
|
list.h
|
Fixed minor bugs and memory leaks
|
6 years ago |
|
local_layer.c
|
Added loss_scale for Mixed-Precision on Tensor-Cores
|
5 years ago |
|
local_layer.h
|
another compile fix
|
5 years ago |
|
lstm_layer.c
|
Added loss_scale for Mixed-Precision on Tensor-Cores
|
5 years ago |
|
lstm_layer.h
|
final compile fix
|
5 years ago |
|
matrix.c
|
fix calloc realloc failure
|
6 years ago |
|
matrix.h
|
…
|
|
|
maxpool_layer.c
|
Merge branch 'master' into master
|
5 years ago |
|
maxpool_layer.h
|
Added [local_avgpool] layer
|
5 years ago |
|
maxpool_layer_kernels.cu
|
Fixed strech_sway=1 param, local_avgpool for large size
|
5 years ago |
|
network.c
|
Added Contrastive loss
|
5 years ago |
|
network.h
|
adversarial attack - experimental
|
5 years ago |
|
network_kernels.cu
|
Added Contrastive loss
|
5 years ago |
|
nightmare.c
|
…
|
|
|
normalization_layer.c
|
fix calloc realloc failure
|
6 years ago |
|
normalization_layer.h
|
fix prototype
|
6 years ago |
|
option_list.c
|
fix calloc realloc failure
|
6 years ago |
|
option_list.h
|
…
|
|
|
parser.c
|
Added Supervised and Unsupervised Contrastive loss for Classifier [net] contrastive=1 unsupervised=1 [contrastive] classes=1000
|
5 years ago |
|
parser.h
|
…
|
|
|
region_layer.c
|
Added [net] dynamic_minibatch=1 for increasing mini_batch_size when random=1 is used
|
5 years ago |
|
region_layer.h
|
fix prototype
|
6 years ago |
|
reorg_layer.c
|
fix calloc realloc failure
|
6 years ago |
|
reorg_layer.h
|
fix prototype
|
6 years ago |
|
reorg_old_layer.c
|
fix calloc realloc failure
|
6 years ago |
|
reorg_old_layer.h
|
fix prototype
|
6 years ago |
|
rnn.c
|
Set cur_iteration = 0; initially
|
5 years ago |
|
rnn_layer.c
|
Added loss_scale for Mixed-Precision on Tensor-Cores
|
5 years ago |
|
rnn_layer.h
|
final compile fix
|
5 years ago |
|
rnn_vid.c
|
fix calloc realloc failure
|
6 years ago |
|
route_layer.c
|
Merge branch 'master' into master
|
5 years ago |
|
route_layer.h
|
Added groups= and groupd_id= params to the [route] layer
|
6 years ago |
|
sam_layer.c
|
Merge branch 'master' into master
|
5 years ago |
|
sam_layer.h
|
Added [sam] layer
|
6 years ago |
|
scale_channels_layer.c
|
Merge branch 'master' into master
|
5 years ago |
|
scale_channels_layer.h
|
Added Fusion features for ASFF and for BiFPN(EfficientDet)
|
6 years ago |
|
shortcut_layer.c
|
Fixed typo weights_normalizion -> weights_normalization
|
5 years ago |
|
shortcut_layer.h
|
Fixed typo weights_normalizion -> weights_normalization
|
5 years ago |
|
softmax_layer.c
|
Added Supervised and Unsupervised Contrastive loss for Classifier [net] contrastive=1 unsupervised=1 [contrastive] classes=1000
|
5 years ago |
|
softmax_layer.h
|
Added Supervised and Unsupervised Contrastive loss for Classifier [net] contrastive=1 unsupervised=1 [contrastive] classes=1000
|
5 years ago |
|
super.c
|
…
|
|
|
swag.c
|
…
|
|
|
tag.c
|
Set cur_iteration = 0; initially
|
5 years ago |
|
tree.c
|
fix calloc realloc failure
|
6 years ago |
|
tree.h
|
…
|
|
|
upsample_layer.c
|
fix calloc realloc failure
|
6 years ago |
|
upsample_layer.h
|
…
|
|
|
utils.c
|
Merge branch 'master' of github.com:AlexeyAB/darknet
|
5 years ago |
|
utils.h
|
Merge branch 'master' into master
|
5 years ago |
|
version.h
|
…
|
|
|
version.h.in
|
…
|
|
|
voxel.c
|
…
|
|
|
writing.c
|
…
|
|
|
yolo.c
|
Added flag averaging detections: ./darknet detector demo ... -avgframes 3
|
5 years ago |
|
yolo_console_dll.cpp
|
Fixed compilation for ZED camera
|
5 years ago |
|
yolo_layer.c
|
[net] mosaic_bound=1 fixed bboxes going out of bounds. Count rewritten bboxes. Changed self-adversarial-training.
|
5 years ago |
|
yolo_layer.h
|
batch inference refactoring
|
6 years ago |
|
yolo_v2_class.cpp
|
Fixed time_step=1 for darknet.so/dll
|
6 years ago |
AlexeyAB