nihui
/
ncnn
Not watched
1
0
Fork 0
Code
Releases 46 Wiki evaluate Activity
Issues 0 Pull Requests 0 Datasets Model Cloudbrain HPC
You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.
652 Commits
5 Branches
354 MB
1 Download
 
 
 
 
 
 
Tree: 5646b7d2c2
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from '5646b7d2c2'
${ noResults }
ncnn/src/layer
nihuini 5646b7d2c2 flatten image 7 years ago
..
arm fix the bug with convdw7x7 op working on int8 mode (#818) 7 years ago
shader flatten image 7 years ago
x86 fix the bug with convdw7x7 op working on int8 mode (#818) 7 years ago
absval.cpp add absval sigmoid tanh shader 7 years ago
absval.h add absval sigmoid tanh shader 7 years ago
argmax.cpp implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
argmax.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
batchnorm.cpp do not upload unnecessary pack1 weight, reduce gpu memory usage 7 years ago
batchnorm.h reduce memory usage of weight packing 7 years ago
bias.cpp implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
bias.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
binaryop.cpp fix binaryop shader binding, use shared buffer state, fix blob copy in non-light mode, fix #817 7 years ago
binaryop.h massive simd optimize in compute shader (#772) 7 years ago
bnll.cpp implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
bnll.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
clip.cpp clip shader 7 years ago
clip.h clip shader 7 years ago
concat.cpp bind the same pipeline may cause driver incorrectly optimize into one, use two pipelines to always change the current one 7 years ago
concat.h bind the same pipeline may cause driver incorrectly optimize into one, use two pipelines to always change the current one 7 years ago
convolution.cpp fix the bug with convdw7x7 op working on int8 mode (#818) 7 years ago
convolution.h new int8 implement,better accuracy (#749) 7 years ago
convolutiondepthwise.cpp fix the bug with convdw7x7 op working on int8 mode (#818) 7 years ago
convolutiondepthwise.h new int8 implement,better accuracy (#749) 7 years ago
crop.cpp respect pad param in deconv vulkan 7 years ago
crop.h crop shader 7 years ago
deconvolution.cpp respect pad param in deconv vulkan 7 years ago
deconvolution.h respect pad param in deconv vulkan 7 years ago
deconvolutiondepthwise.cpp respect pad param in deconv vulkan 7 years ago
deconvolutiondepthwise.h respect pad param in deconv vulkan 7 years ago
dequantize.cpp layer holds bottom blob scale, depthwise convolution read group scales 7 years ago
dequantize.h core int8 inference, quantize and dequantize, net using flag, caffe2ncnn reads int8 scale table 7 years ago
detectionoutput.cpp zero detected is not error 7 years ago
detectionoutput.h mxnet-ssd wip ... 7 years ago
dropout.cpp massive simd optimize in compute shader (#772) 7 years ago
dropout.h massive simd optimize in compute shader (#772) 7 years ago
eltwise.cpp add absval sigmoid tanh shader 7 years ago
eltwise.h bind the same pipeline may cause driver incorrectly optimize into one, use two pipelines to always change the current one 7 years ago
elu.cpp implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
elu.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
embed.cpp implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
embed.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
exp.cpp implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
exp.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
expanddims.cpp implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
expanddims.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
flatten.cpp flatten image 7 years ago
flatten.h flatten image 7 years ago
innerproduct.cpp new int8 implement,better accuracy (#749) 7 years ago
innerproduct.h new int8 implement,better accuracy (#749) 7 years ago
input.cpp [WIP] vulkan compute (#618) 7 years ago
input.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
instancenorm.cpp implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
instancenorm.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
interp.cpp interp vulkan 7 years ago
interp.h interp vulkan 7 years ago
log.cpp implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
log.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
lrn.cpp lrn norm across channel pack4, rename member name with pipeline prefix 7 years ago
lrn.h lrn norm across channel pack4, rename member name with pipeline prefix 7 years ago
lstm.cpp fixed cell initialization in LSTM layer 7 years ago
lstm.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
memorydata.cpp implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
memorydata.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
mvn.cpp implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
mvn.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
normalize.cpp implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
normalize.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
packing.cpp flatten image 7 years ago
packing.h packing param for identity packing when padding disabled, auto packing conversion between cpu and gpu blob 7 years ago
padding.cpp fix binaryop shader binding, use shared buffer state, fix blob copy in non-light mode, fix #817 7 years ago
padding.h massive simd optimize in compute shader (#772) 7 years ago
permute.cpp fallback to cpu forward if layer not support vulkan, automatically! 7 years ago
permute.h permute shader 7 years ago
pooling.cpp fix memleak 7 years ago
pooling.h fix memleak 7 years ago
power.cpp implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
power.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
prelu.cpp prelu shader 7 years ago
prelu.h prelu shader 7 years ago
priorbox.cpp priorbox shader, fix permute order 1 on image, fix potential staging memory leak 7 years ago
priorbox.h priorbox shader, fix permute order 1 on image, fix potential staging memory leak 7 years ago
proposal.cpp implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
proposal.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
psroipooling.cpp implement psroipooling layer 7 years ago
psroipooling.h implement psroipooling layer 7 years ago
quantize.cpp core int8 inference, quantize and dequantize, net using flag, caffe2ncnn reads int8 scale table 7 years ago
quantize.h core int8 inference, quantize and dequantize, net using flag, caffe2ncnn reads int8 scale table 7 years ago
reduction.cpp fix reduction dim 2, fix #593 7 years ago
reduction.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
relu.cpp fix binaryop shader binding, use shared buffer state, fix blob copy in non-light mode, fix #817 7 years ago
relu.h new int8 implement,better accuracy (#749) 7 years ago
reorg.cpp reorg shader 7 years ago
reorg.h reorg shader 7 years ago
requantize.cpp new int8 implement,better accuracy (#749) 7 years ago
requantize.h new int8 implement,better accuracy (#749) 7 years ago
reshape.cpp reshape shader family 7 years ago
reshape.h reshape shader family 7 years ago
rnn.cpp implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
rnn.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
roialign.cpp trival fix for the last element 7 years ago
roialign.h implement roialign layer 7 years ago
roipooling.cpp implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
roipooling.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
scale.cpp do not upload unnecessary pack1 weight, reduce gpu memory usage 7 years ago
scale.h reduce memory usage of weight packing 7 years ago
shufflechannel.cpp shufflechannel shader 7 years ago
shufflechannel.h shufflechannel shader 7 years ago
sigmoid.cpp add absval sigmoid tanh shader 7 years ago
sigmoid.h add absval sigmoid tanh shader 7 years ago
slice.cpp implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
slice.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
softmax.cpp warn users about the old wrong softmax behavior on axis not zero 7 years ago
softmax.h lrn norm across channel pack4, rename member name with pipeline prefix 7 years ago
split.cpp [WIP] vulkan compute (#618) 7 years ago
split.h [WIP] vulkan compute (#618) 7 years ago
spp.cpp implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
spp.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
squeeze.cpp implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
squeeze.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
tanh.cpp add absval sigmoid tanh shader 7 years ago
tanh.h add absval sigmoid tanh shader 7 years ago
threshold.cpp implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
threshold.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
tile.cpp implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
tile.h implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469 8 years ago
unaryop.cpp massive simd optimize in compute shader (#772) 7 years ago
unaryop.h massive simd optimize in compute shader (#772) 7 years ago
yolodetectionoutput.cpp zero detected is not error 7 years ago
yolodetectionoutput.h implement multiscale yolov2, update example model comment 7 years ago
yolov3detectionoutput.cpp zero detected is not error 7 years ago
yolov3detectionoutput.h Support yolov3 and mobilenet yolov3 7 years ago
Delete task

Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.

cancel
confirm

No Description

C++ C Python Text Protocol Buffer other