nihui/ncnn - ncnn - 开源协同云脑生态支撑系统

Commit Graph

Author	SHA1	Message	Date
nihuini	0d6cc01d55	innerproduct handle mish activation, fix naive C testing, fix #1930	6 years ago
nihui	b5e288b521	layer creator function is not necessary for built-in layers	6 years ago
Tijmen Verhulsdonck	73aa99e83c	LSTM arm/x86 + fp16 innerproduct arm (#1881 ) * added fp16 weight storage version * Small changes * Fixed fp16 weight storage layers * fix innerproduct * fix loop error * Fix windows build. Disable fp 16 conversion when detecting int8 weights. Implement requested changes. * Restyled by clang-format * Restyled by astyle * Restyled by clang-format * Restyled by astyle * Update option.cpp Set fp16 storage based on vulkan being used or not. * added ability for storing state in lstm layer * added avx lstm * added arm lstm * fix innerproduct activation location and add 4 parallel channel version * Restyled by clang-format * Restyled by astyle * Restyled by clang-format * Restyled by astyle * revert arm file * commit before switch * implement requested changes * Restyled by clang-format * Restyled by astyle * Restyled by clang-format * Restyled by astyle * More x86 optimized implementations of common layers. Added LSTM layers for arm and x86 + a ctest to verify the layer accuracy Added fp16 innerproduct for arm * fix non avx build * Add fp16 arm compiler and cpu checks. Remove statefullness from LSTM implementation. * Fix build check for fp16 arm * Bypass lstm_fp16 if not supported * Build order was incorrect * fix std::min missing in windows build * Restyled by clang-format * Restyled by astyle * Restyled by clang-format * Restyled by astyle * attempting to fix gnu build by enabling: -mfp16-format=ieee to fix the missing __fp16 type * remove double "fix" * Specify ieee fp16 format * implement requested changes * fix arm non-fp16 build * fix arm lstm * Restyled/pull 1881 (#15) * Restyled by clang-format * Restyled by astyle * Restyled by clang-format * Restyled by astyle Co-authored-by: Restyled.io <commits@restyled.io> * Check blob size on arm lstm * fix styling Co-authored-by: Restyled.io <commits@restyled.io>	6 years ago
nihui	3ef995ed1e	format code style and setup restyled.io (#1840 )	6 years ago
nihuini	bb7202fb52	add missing sigmoid activation in innerproduct arm, fix #1716	6 years ago
nihuini	85d5e5d3e4	fix innerproduct vulkan pack8 and arm neon, disable packing_layout for int8 test	6 years ago
nihui	b7c82fcc45	code clean, concat bf16s	6 years ago
nihui	7d1eec3d5d	the use_bf16_storage option	6 years ago
nihui	e14716dfef	convolution and pooling make padding helper, flatten innerproduct pooling bf16s neon	6 years ago
nihui	6f2ef1932d	int8 code refactoring wip, add int8 test	6 years ago
nihuini	bffb2af2ff	fix build on armv7 without neon	6 years ago
nihuini	336d1c1edd	remove the ncnn namespace for in source Option	6 years ago
nihuini	567e2bd501	a dirty hack for resolving int8 pack4 crash	6 years ago
nihuini	cd4be6d0fa	call vulkan create_pipeline on the vkdev condition, drop opt_cpu hacks	6 years ago
nihuini	581a06d471	since innerproduct pack4 always consumes flattened blob, which layout is same as pack1 branch, so reuse pack1 implementation to reduce memory usage	6 years ago
nihui	9e529354fb	arm neon optimization for conv1x1s2 pack4	6 years ago
nihuini	50d5896ce7	reshape pack4	6 years ago
nihui	48e3e7d49c	move neon activation into a wrapper function	6 years ago
nihui	b37ecab630	auto flatten before innerproduct pack4	6 years ago
nihui	394f6786b9	neon enable support_packing	6 years ago
nihuini	c4f23ae8ad	rename Mat packing to elempack	7 years ago
nihuini	8da6e60db1	innerproduct pack4 arm neon	7 years ago
nihui	3e003ffd98	fuse sigmoid	7 years ago
nihuini	c6e075cef7	fuse deconv/innerproduct relu arm	7 years ago
nihui	a169cec363	core int8 inference, quantize and dequantize, net using flag, caffe2ncnn reads int8 scale table	8 years ago
nihui	9706cd1447	implement ncnn blob/workspace allocator, fine-grained per-layer openmp threads control, fix #469	8 years ago
dong	6ea09ebf2c	Use aarch64 assembly to replace arm intrinsics	8 years ago
nihui	df218110be	unroll num_output for innerproduct, about 60% speed gain	8 years ago
nihui	08e261f423	innerproduct produce continous blob, fix #236	8 years ago
nihuini	a84ba8fc0f	element type storage support in Mat, move data member the first so that a pointer to Mat is a pointer to data, convenient index access for float vector	8 years ago
zengping	a54f14feca	[fix-compile-warnings] fix compiler warnings, and add werror in CMakeLists.txt (#217 ) * [fix-compile-warnings] fix compiler warnings, and add werror in CMakeLists.txt * [fix-compile-warnings] fix compiler warnings, remote ycm_extra_conf.py	8 years ago
nihuini	a3be17eb7e	special path for 1x1xc innerproduct	8 years ago
nihuini	b7db8be4f6	add ncnn source qwq	9 years ago

33 Commits (6ebb774e376db0fadba67e258955b4cdf3c507bc)