github
/
darknet
mirror of https://github.com/AlexeyAB/darknet.git
1
0
Fork 0
Code Issues Releases Wiki Activity
1498 Commits
1 Branch
7 Tags
380 MiB
Tag: Branch: Tree: 68db8efb27
Branches Tags
${ item.name }
Create tag ${ searchTerm }
Create branch ${ searchTerm }
from '68db8efb27'
${ noResults }
Commit Graph

185 Commits (68db8efb27170b411481e8765558d89b45635101)

Author SHA1 Message Date
AlexeyAB 68db8efb27 Added Fusion features for ASFF and for BiFPN(EfficientDet)
6 years ago
AlexeyAB 0d98f20be1 You can train with mini_batch=128 (batch=256 subdivisions=2) on GPU 8 GB VRAM + 128 GB CPU-RAM and more. Set in cfg-file: optimized_memory=3 and workspace_size_limit_MB=2000 or 4000
6 years ago
AlexeyAB b832c727dc Fixed random=1 for assisted_excitation_forward
6 years ago
dccho b4c0fbaec8 fix memory free bug
6 years ago
AlexeyAB 10c40551dc GIoU + Gaussian fix
6 years ago
dccho dd34fe156a enable random resize training for efficient net
6 years ago
AlexeyAB 11142d00be Fixed non-square network for Training Classifier
6 years ago
AlexeyAB 70bf88ce01 Fixed swish-activation for [shortcut_layer]
6 years ago
AlexeyAB ded620ac6b fixed memory deallocation
6 years ago
AlexeyAB d91d59a22f Optimized memory allocation for Detection (inference only), without allocation memory for training
6 years ago
AlexeyAB bf8ea4183d Added MISH activation, use activation=mish in [convolutional] layers
6 years ago
AlexeyAB e345b8793c Fixed ./darknet partial for Anti-aliasing
6 years ago
AlexeyAB b3a2495298 Added Gaussian YOLOv3 layer [Gaussian_yolo]
6 years ago
AlexeyAB e6486ab594 Use ignore_thresh only if class_id matched. Temporary changed Assisted_Excitation (reduces background activations rather than enhancing objects activations). Added antialiasiong=2 for 2x2.
6 years ago
AlexeyAB b0b1584a06 Minor fix
6 years ago
AlexeyAB e33019e669 Fixed stride_x and stride_y. Fixed AntiAliasing. Minor fixes.
6 years ago
AlexeyAB 9c02df864e Fixed assisted_excitation and added also for [shortcut] layer
6 years ago
AlexeyAB be5d0d6693 Added assisted_excitation=1 for [convolutional] layer on GPU
6 years ago
AlexeyAB 9e26472b1a Added antialiasing=1 param for [maxpool]-layer on GPU and CPU
6 years ago
AlexeyAB 80ceee4fca Added antialiasing=1 param for [convolutional]-layer on CPU (only forward inference)
6 years ago
AlexeyAB 11b8b2bddf Added antialiasing=1 param for [convolutional]-layer on GPU
6 years ago
AlexeyAB a4a06c337a Addex stride_x= and stride_y= for [convolutional] layer for MatrixNet
6 years ago
AlexeyAB 102ab710a9 Fixed fuse_conv_batchnorm() for TridentNet
6 years ago
AlexeyAB 480b3eccb1 Compile fix
6 years ago
AlexeyAB c1f184249e Added assisted_excitation_forward for training on CPU
6 years ago
AlexeyAB 7d4143332d Minor output fixes
6 years ago
AlexeyAB 88cccfcad4 Added Swish-activation
6 years ago
AlexeyAB 8c80ba6b53 Fixed weights-size for Depthwise grouped convolutional
6 years ago
AlexeyAB f0582446f2 Fixed Depth-wise (grouped) convolution for CPU and GPU (CUDNN=0/1)
6 years ago
AlexeyAB bd689f91ac Added TridentNet (state-of-art on MS COCO)
6 years ago
AlexeyAB 88ce9dcca6 Added dilation parameter for convolutional, conv_lstm and crnn layers
6 years ago
AlexeyAB 038289eb7d Some conv-lstm, sgdr and other fixes
6 years ago
AlexeyAB 4f72fcc015 Added grouped convolutional (depth-wise convolutional)
6 years ago
AlexeyAB 31dc6c8680 Added LSTM sequence detector, and blur data augmentation (for OpenCV only)
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
AlexeyAB 75f2a3e7cf Added object Detection & Tracking using conv-rnn layer on frames from video
6 years ago
AlexeyAB 9b09abe122 Fixed convolutional-layer when it is used as base for crnn-layer
6 years ago
Stefano Sinigardi 00de023601 fully separate C-API from CPP-API
6 years ago
Stefano Sinigardi b3579380dc improve compatibility with c++ compilers, prepare for CMake
6 years ago
AlexeyAB 28106c0fd8 Optimized memory allocation for XNOR on CPU
6 years ago
AlexeyAB 9e07605bc5 get_connected_workspace_size() and get_convolutional_workspace_size()
6 years ago
AlexeyAB 12b6e93893 CHECK_CUDA is used everywhere
6 years ago
AlexeyAB f09a9c3315 XNOR uses Tensor Cores on Turing GPU CC>=7.3 (not Volta)
6 years ago
AlexeyAB c7309c1fdb Fixed CRNN (RNN based on Convolution) layer
6 years ago
AlexeyAB 640bdbc063 LSTM, RNN, GRU - use connected_layer that uses cuDNN. Fixed CRNN for conv-layer with cuDNN.
6 years ago
AlexeyAB 090d934c0f Minor speedup on CPU
6 years ago
AlexeyAB 2d3220cef5 Look at wmma::bmma_sync(), bmmaBitOpXOR, bmmaAccumulateOpPOPC
6 years ago
AlexeyAB 46be08db37 Minor fix
6 years ago
AlexeyAB 3a51f4af74 Experimental repack
6 years ago