Releases: Tencent/ncnn
android ios macos linux windows webassembly watchos tvos visionos 预编译库 20241226 5285895
编译版本，默认配置，android-ndk-r27c，xcode 15.2，ubuntu-20.04，ubuntu-22.04，ubuntu-24.04，vs2015，vs2017，vs2019，vs2022，emscripten-3.1.28
| file | content | arch |
|---|---|---|
| ncnn-full-source.zip | 包含全部 submodule 代码的完整源码 | |
| ncnn-android.zip | android 静态库/动态库 | armeabi-v7a + arm64-v8a + x86 + x86_64 + riscv64 |
| ncnn-android-vulkan.zip | android 静态库/动态库，支持 GPU | armeabi-v7a + arm64-v8a + x86 + x86_64 + riscv64 |
| ncnn-apple.zip | apple xcframework，ios + ios-simulator + macos + mac-catalyst + watchos + watchos-simulator + tvos + tvos-simulator + visionos + visionos-simulator | arm64 + arm64e + x86_64 |
| ncnn-apple-vulkan.zip | apple xcframework，ios + ios-simulator + macos + mac-catalyst + watchos + watchos-simulator + tvos + tvos-simulator + visionos + visionos-simulator，支持 GPU | arm64 + arm64e + x86_64 |
| ncnn-ios.zip | ios 静态库 | arm64 |
| ncnn-ios-vulkan.zip | ios 静态库，支持 GPU | arm64 |
| ncnn-ios-simulator.zip | ios simulator 静态库 | x86_64 + arm64 |
| ncnn-ios-simulator-vulkan.zip | ios simulator 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-macos.zip | macos 静态库 | x86_64 + arm64 |
| ncnn-macos-vulkan.zip | macos 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-mac-catalyst.zip | mac catalyst 静态库 | x86_64 + arm64 |
| ncnn-mac-catalyst-vulkan.zip | mac catalyst 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-watchos.zip | watchos 静态库 | armv7k + arm64_32 |
| ncnn-watchos-simulator.zip | watchos simulator 静态库 | x86_64 + arm64 |
| ncnn-tvos.zip | tvos 静态库 | x86_64 + arm64 |
| ncnn-tvos-vulkan.zip | tvos 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-tvos-simulator.zip | tvos simulator 静态库 | x86_64 + arm64 |
| ncnn-tvos-simulator-vulkan.zip | tvos simulator 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-visionos.zip | visionos 静态库 | arm64 |
| ncnn-visionos-vulkan.zip | visionos 静态库，支持 GPU | arm64 |
| ncnn-visionos-simulator.zip | visionos simulator 静态库 | x86_64 + arm64 |
| ncnn-visionos-simulator-vulkan.zip | visionos simulator 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-ubuntu.zip | ubuntu linux 静态库/动态库，支持 GPU，模型转换工具 | x86_64 |
| ncnn-windows.zip | windows 静态库/动态库，支持 GPU，模型转换工具 | x86 + x64 + arm + arm64 |
| ncnn-webassembly.zip | webassembly 静态库 | wasm32 + simd + threads + simd-threads |
embed 支持int8量化
gemm 支持int8量化
multiheadattention 支持int8量化
新增spectrogram和inverse spectrogram实现
arm rmsnorm neon优化
arm layernorm neon fp32/bf16s/fp16s优化
x86 rmsnorm sse2/avx/avx512优化
x86 layernorm sse2/avx/avx512优化
x86 gemm int8 sse2/xop/avx/avx512/vnni/vnniint8优化
更新riscv vector标准到1.0，重写全部ncnn riscv优化代码，自动探测rvv/zfh/zvfh/xtheadvector并分发
riscv gemm rvv优化支持128bit/256bit vlen
禁用x86倒数优化避免可能的精度损失
改善harmonyos cpu拓扑结构abi兼容性
暂时禁用mesa驱动的vulkan矩阵扩展支持
兼容ndk-21编译asimdfhm目标的错误导致的问题
兼容clang-18编译avx512bf16时编译器崩溃的问题
禁用msvc对windows arm平台exp/tanh的svml优化以解决计算错误
探测avxvnniint8/avxvnniint16/avxneconvert指令集
runtime cpu开启时仅使用ncnn cmake内置的编译参数
删除windows arm32支持(@Shironana817)
android默认启用16kb pagesize编译，android-api升级到21
vkCreateDevice失败时不直接崩溃(@Upliner)
为powerpc架构跳过0.5附近数值的unaryop round测试用例
pnnx更新到torch-2.5
pnnx支持从traced inputs自动设定inputshape
pnnx编译不再输出来自torch头文件的警告
pnnx重排pass level2内的全部顺序，并复用pattern
pnnx不再保存debug中间模型(@LJoson)
pnnx输出python脚本的onnx导出代码更新到export(@whyb)
pnnx合并t5-layernorm为rmsnorm
pnnx不再折叠具有动态shape的tensor
pnnx在输出的python脚本中使用隐含的int转换避免trace时常数化
pnnx转换Tensor.select为ncnn crop+squeeze
pnnx转换onnx constantofshape为torch.zeros/ones
pnnx修正onnx clip在可选min/max缺失时的转换问题
ci更新riscv64工具链
ci添加c908/spacemit-x60
ci webassembly兼容node>20
ci android添加riscv64目标并打包
添加vim3 vulkan跑分数据(@GIBEREZ)
New Contributors
- @ankushgoel27 made their first contribution in #5709
- @Shironana817 made their first contribution in #5811
- @GIBEREZ made their first contribution in #5821
Full Changelog: 2024082...2024122
Contributors
Assets 42
- 13.3 MB
2024-12-26T03:28:00Z - 19.6 MB
2024-12-26T03:28:00Z - 27.6 MB
2024-12-26T03:28:00Z - 15.5 MB
2024-12-26T03:28:00Z - 81.5 MB
2024-12-26T03:28:00Z - 55.1 MB
2024-12-26T03:28:00Z - 21.3 MB
2024-12-26T03:27:59Z - 10.7 MB
2024-12-26T03:27:59Z - 7.01 MB
2024-12-26T03:27:59Z - 4.29 MB
2024-12-26T03:27:59Z -
2024-12-26T02:56:43Z -
2024-12-26T02:56:43Z - Loading
android ios macos linux windows webassembly 预编译库 20240820 a6d3ef5
编译版本，默认配置，android-ndk-r27，xcode 15.2，ubuntu-20.04，ubuntu-22.04，ubuntu-24.04，vs2015，vs2017，vs2019，vs2022，emscripten-3.1.28
| file | content | arch |
|---|---|---|
| ncnn-full-source.zip | 包含全部 submodule 代码的完整源码 | |
| ncnn-android.zip | android 静态库/动态库 | armeabi-v7a + arm64-v8a + x86 + x86_64 |
| ncnn-android-vulkan.zip | android 静态库/动态库，支持 GPU | armeabi-v7a + arm64-v8a + x86 + x86_64 |
| ncnn-apple.zip | apple xcframework，ios + ios-simulator + macos + mac-catalyst + watchos + watchos-simulator + tvos + tvos-simulator + visionos + visionos-simulator | arm64 + arm64e + x86_64 |
| ncnn-apple-vulkan.zip | apple xcframework，ios + ios-simulator + macos + mac-catalyst + watchos + watchos-simulator + tvos + tvos-simulator + visionos + visionos-simulator，支持 GPU | arm64 + arm64e + x86_64 |
| ncnn-ios.zip | ios 静态库 | arm64 |
| ncnn-ios-vulkan.zip | ios 静态库，支持 GPU | arm64 |
| ncnn-ios-simulator.zip | ios simulator 静态库 | x86_64 + arm64 |
| ncnn-ios-simulator-vulkan.zip | ios simulator 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-macos.zip | macos 静态库 | x86_64 + arm64 |
| ncnn-macos-vulkan.zip | macos 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-mac-catalyst.zip | mac catalyst 静态库 | x86_64 + arm64 |
| ncnn-mac-catalyst-vulkan.zip | mac catalyst 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-watchos.zip | watchos 静态库 | armv7k + arm64_32 |
| ncnn-watchos-simulator.zip | watchos simulator 静态库 | x86_64 + arm64 |
| ncnn-tvos.zip | tvos 静态库 | x86_64 + arm64 |
| ncnn-tvos-vulkan.zip | tvos 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-tvos-simulator.zip | tvos simulator 静态库 | x86_64 + arm64 |
| ncnn-tvos-simulator-vulkan.zip | tvos simulator 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-visionos.zip | visionos 静态库 | arm64 |
| ncnn-visionos-vulkan.zip | visionos 静态库，支持 GPU | arm64 |
| ncnn-visionos-simulator.zip | visionos simulator 静态库 | x86_64 + arm64 |
| ncnn-visionos-simulator-vulkan.zip | visionos simulator 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-ubuntu.zip | ubuntu linux 静态库/动态库，支持 GPU，模型转换工具 | x86_64 |
| ncnn-windows.zip | windows 静态库/动态库，支持 GPU，模型转换工具 | x86 + x64 + arm + arm64 |
| ncnn-webassembly.zip | webassembly 静态库 | wasm32 + simd + threads + simd-threads |
新增RMSNorm层和对应的pnnx转换，单元测试
x86 convolution tiled gemm优化
量化工具支持 rnn/lstm/gru 动态量化
x86 lstm int8 sse2/xop/avx2/avx512/avx512vnni/avxvnni优化
arm rnn/lstm/gru int8 neon/asimdhp/asimddp优化
multiheadattention支持qdim参数与embed_dim不同
multiheadattention支持scale参数
更新pybind11到2.12支持numpy2
添加wasi支持(@quink-black)
添加x86/arm convolution/slice/concat oom单元测试
onnx2ncnn工具添加警告和推荐使用pnnx的信息输出(@lll143653)
修复x86 avx512 vnni指令派发失效的问题
增强x86/arm计算内核在内存不足时的错误返回
仅在windows arm平台使用ruapu指令集探测
windows mingw编译时支持大小核和SMT探测
修复powerpc vsx计算abs可能的错误
修复arm vfpv4条件下可能的fp16s/bf16s同时启用的冲突
修复aarch64架构l2-cache很小时因gemm K分块可能的越界读错误
修复riscv v tanh计算错误(@zhangyang2057)
arm/convolution_3x3_pack1to8_fp16s使用ldr/str替代ld1/st1优化(@quink-black)
修复c_api无参数函数声明(@quink-black)
c_api添加set_vulkan_device接口(@Baiyuetribe)
pyncnn添加从python bytes内存加载模型的接口(@joeyballentine)
为VkAndroidHardwareBufferImageAllocator添加NCNN_PLATFORM_API宏(@Xyzhao1999)
修复mingw64编译时avx崩溃和termux编译错误(@TianZerL)
修复在关闭NCNN_BF16时arm riscv编译错误
修复x86-wsl编译时的无用变量警告(@Tabbleman)
create_gpu_instance()中不进行destroy_gpu_instance()(@Asd-g)
更新ruapu.h(@lazyparser)
修复ndk-r27在cmake阶段的编译错误(@Galasnow)
添加yolov8示例代码(@whyb)
pnnx支持转换dynamo导出的onnx
pnnx默认编译onnx2pnnx支持，支持转换conv/convtranspose/pad/linear/softmax/relu/resize/upsample/avgpool/maxpool/batchnorm/lrn/layernorm/instancenorm/groupnorm/rnn/lstm/gru/prelu/gelu/elu/leakyrelu/relu6/celu/hardshrink/hardsigmoid/hardswish/clip/multiheadattention/reducemin/reducemax/reducemean/reducesum/reduceprod/logsoftmax/logsigmoid/mish/selu/sigmoid/silu/softmin/softplus/softshrink/softsign/tanh/tanhshrink/expand/permute/repeat/reshape/select/slice/cat/ceil/chunk/flatten/floor/maximum/minimum/split/squeeze/stack/transpose/unbind/unsqueeze
pnnx支持转换onnx指定inputshape
pnnx转换onnx遇到动态shape时尝试折叠非动态轴相关的常量
pnnx转换onnx合并简单的shape运算pattern
pnnx清除onnx中无用的cast
pnnx接受bf16的模型转换和输入输出类型
pnnx转换torch.tile/torch.where/torch.logaddexp
pnnx转换无dilation参数的F.maxpool到ncnn
pnnx转换1到2个轴参数的torch.roll到ncnn
pnnx转换有dim参数的torch.max/torch.min时返回tuple并自动删除没有用到的indice输出
pnnx合并onnx sdpa和qdim mha
pnnx识别sdpa的batch轴
pnnx支持torch-2.3和torch-2.4
pnnx不再折叠有就地操作的别名tensor为常量
pnnx转换到的ncnn模型py自动替换long为int
ci添加windows clang
ci添加harmonyos
ci添加mingw(@TianZerL)
ci添加esp32和esp32编译文档(@luxincn)
重构release ci脚本
发布ubuntu 24.04预编译包
发布visionos/visionos-simulator vulkan预编译包
pypi发布python 3.13预编译包
更新pytorch/onnx模型转换文档(@whyb)
添加riscv-gnu-toolchain编译文档(@Tabbleman)
添加harmonyos vulkan编译文档(@cugxchen)
修正vulkan-notes文档的错误(@roachsinai)
更新qcom855plus跑分数据
添加RaspberryPi 5 GPU超频跑分数据(@CharlieYu4994)
添加EPYC7742和V100跑分数据(@sakria9)
添加Snapdragon 888跑分数据(@chainsx)
添加RaspberryPi 5 CPU超频跑分数据(@chainsx)
添加OrangePi 5Plus跑分数据(@inspireMeNow)
添加Snapdragon 765G跑分数据(@inspireMeNow)
添加CVITEK SG2000跑分数据(@inspireMeNow)
添加OrangePi CM4跑分数据(@py1066)
添加Axera AX630C跑分数据(@UOPiceman)
添加Kunpeng 920 7260跑分数据(@violet73)
New Contributors
- @quink-black made their first contribution in #5436
- @Tabbleman made their first contribution in #5444
- @roachsinai made their first contribution in #5472
- @Asd-g made their first contribution in #5437
- @lazyparser made their first contribution in #5499
- @CharlieYu4994 made their first contribution in #5518
- @Xyzhao1999 made their first contribution in #5521
- @sakria9 made their first contribution in #5528
- @inspireMeNow made their first contribution in #5550
- @py1066 made their first contribution in #5551
- @UOPiceman made their first contribution in #5559
- @luxincn made their first contribution in #5567
- @zhangyang2057 made their first contribution in #5584
- @violet73 made their first contribution in #5606
Full Changelog: 2024041...2024082
Contributors
Assets 42
android ios macos linux windows webassembly 预编译库 20240410 56775de
编译版本，默认配置，android-ndk-r26c，xcode 15.2，ubuntu-20.04，ubuntu-22.04，vs2015，vs2017，vs2019，vs2022，emscripten-3.1.28
| file | content | arch |
|---|---|---|
| ncnn-full-source.zip | 包含全部 submodule 代码的完整源码 | |
| ncnn-android.zip | android 静态库/动态库 | armeabi-v7a + arm64-v8a + x86 + x86_64 |
| ncnn-android-vulkan.zip | android 静态库/动态库，支持 GPU | armeabi-v7a + arm64-v8a + x86 + x86_64 |
| ncnn-apple.zip | apple xcframework，ios + ios-simulator + macos + mac-catalyst + watchos + watchos-simulator + tvos + tvos-simulator + visionos + visionos-simulator | arm64 + arm64e + x86_64 |
| ncnn-apple-vulkan.zip | apple xcframework，ios + ios-simulator + macos + mac-catalyst + watchos + watchos-simulator + tvos + tvos-simulator + visionos + visionos-simulator，支持 GPU | arm64 + arm64e + x86_64 |
| ncnn-ios.zip | ios 静态库 | arm64 |
| ncnn-ios-vulkan.zip | ios 静态库，支持 GPU | arm64 |
| ncnn-ios-simulator.zip | ios simulator 静态库 | x86_64 + arm64 |
| ncnn-ios-simulator-vulkan.zip | ios simulator 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-macos.zip | macos 静态库 | x86_64 + arm64 |
| ncnn-macos-vulkan.zip | macos 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-mac-catalyst.zip | mac catalyst 静态库 | x86_64 + arm64 |
| ncnn-mac-catalyst-vulkan.zip | mac catalyst 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-watchos.zip | watchos 静态库 | armv7k + arm64_32 |
| ncnn-watchos-simulator.zip | watchos simulator 静态库 | x86_64 + arm64 |
| ncnn-tvos.zip | tvos 静态库 | x86_64 + arm64 |
| ncnn-tvos-vulkan.zip | tvos 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-tvos-simulator.zip | tvos simulator 静态库 | x86_64 + arm64 |
| ncnn-tvos-simulator-vulkan.zip | tvos simulator 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-visionos.zip | visionos 静态库 | arm64 |
| ncnn-visionos-simulator.zip | visionos simulator 静态库 | x86_64 + arm64 |
| ncnn-ubuntu.zip | ubuntu linux 静态库/动态库，支持 GPU，模型转换工具 | x86_64 |
| ncnn-windows.zip | windows 静态库/动态库，支持 GPU，模型转换工具 | x86 + x64 + arm + arm64 |
| ncnn-webassembly.zip | webassembly 静态库 | wasm32 + simd + threads + simd-threads |
解耦合layer cpu和vulkan，不再使用virtual public继承
支持编译动态库时编译单元测试
单层特性掩码支持禁用多线程
extractor set_num_threads和set_vulkan_compute现在是无操作
gpu shader增加uniform类型改善adreno上fp16兼容性
检测vulkan矩阵扩展8x8x16配置，fp16a条件下默认使用fp16累加
更新stb_image rvv/neon优化
x86 mish avx512优化(@wnqn1597)
riscv gemm fp32 rvv优化(@Xinyu302)
加载模型上传权重时不保留无用的临时数据
c-api新增draw rectangle/text/circle/line接口(@Deepdive543443)
修复armv7平台加载fp16模型sigbus错误
修复reduction L2norm denormal产生inf的问题
修复arm平台pixel_resize rounding导致的数值误差
修复softmax arm fp16计算错误
修复risc-v rvv输出fp16没有自动转换的问题
修复destroy_gpu_instance在驱动加载不完整时crash的问题(@shatyuka)
destroy_gpu_instance等待全部设备idle(@whyb)
修复low-level api没有load_param直接create_pipeline可能的崩溃
修复ncnnoptimize在shape推断的崩溃
ncnnoptimize支持更多新算子，修复gemm权重丢失问题
被调试时候禁用signal指令集检测
windows-arm平台使用ruapu cpu指令集检测
arm vfpv4支持时启用自动转换fp16
在arm64架构中总是报告支持neon和vfpv4
simplevk寻找更多已知的vulkan驱动路径
修复旧cpp标准下risc-v rvv编译错误
修复某些老编译器在debug模式下编译错误
修复uwp平台编译
修复test_reduction运行时的警告
修复NCNN_PIXEL_DRAWING禁用时候编译错误(@shatyuka)
支持MSVC使用LLVM openmp运行时的配合编译(@shatyuka)
修复yolov8 python示例返回空发生错误(@dsplvd)
pnnx解耦torchscript加载，清理cxxabi hack，修复whole-archive链接
pnnx加载dynamo onnx，默认不启用编译
pnnx改善函数化，支持更多slice+inplace复合操作
pnnx转换torch.masked_select/torch.slice_scatter
pnnx支持超过4G的模型
pnnx macos编译universal wheel
pnnx添加entrypoint脚本
pnnx支持动态slice下标
pnnx转换softmin logsoftmax dtype参数
pnnx处理index_put传入空indices和标量数值
pnnx转换一些cudnn conv2d变种
pnnx合并完整slices为tensor_split
pnnx合并静态embedding
pnnx不消除会导致shape变化的数学操作
pnnx改善torch-2.1 mha attn_mask探测
pnnx修复无bias tensor的nn.Conv2d转换
pnnx转换torch.stack负数dim
pnnx添加torch.arange单元测试
pnnx修复图匹配失败时可能的越界访问问题
pnnx识别embedding输入的batch轴为0
pnnx python添加控制fp16参数(@MollySophia)
pnnx添加torch-2.2 ci
github ci使用4并行编译
更新cmake ios工具链，添加visionos ci，watchos支持arm64_32架构
添加apple a17和m3 cpu名称
不再编译apple平台32bit支持，不再编译ios arm64e架构，提升最低部署版本到ios-13
统一android python macos ci
不再打包和发布apple bitcode和32bit预编译包，新增visionos预编译包，新增tvos-gpu预编译包，更新openmp到18.1.2
改善a53/a55双发射文档(@luqiang-guo)
添加windows上protobuf>=22.0编译文档(@Galasnow)
更新macos编译文档(@lll143653)
清理无用的代码警告(@hokamilkv)
修正FAQ的拼写错误(@eltociear)
修正拼写错误(@hugo-syn)
修正拼写错误(@afredooo)
修正convolution_x86注释错误(@strongtz)
添加markdown文档代码辅助标志(@hugo-syn)
添加OneCloud跑分数据(@mizu-bai)
添加AWS c5.4xlarge跑分数据(@mizu-bai)
添加Xeon Phi 3120A跑分数据(@mizu-bai)
添加orangepi zero2跑分数据(@wonderfullook)
添加Dimensity 9300 MT6989跑分数据(@MollySophia)
添加PhytiumPi跑分数据(@HalfSweet)
添加remipi跑分数据(@dreamcmi)
添加radxa zero 3w跑分数据(@Qengineering)
New Contributors
- @wonderfullook made their first contribution in #5277
- @hugo-syn made their first contribution in #5301
- @FartSimps0n made their first contribution in #5304
- @HalfSweet made their first contribution in #5312
- @strongtz made their first contribution in #5310
- @afredooo made their first contribution in #5339
- @shatyuka made their first contribution in #5346
- @dsplvd made their first contribution in #5345
- @Galasnow made their first contribution in #5359
- @hokamilkv made their first contribution in #5365
Full Changelog: 2024010...2024041
Contributors
Assets 38
android ios macos linux windows webassembly 预编译库 20240102 1e88fb8
编译版本，默认配置，android-ndk-r26b，xcode 13.4.1，ubuntu-20.04，ubuntu-22.04，vs2015，vs2017，vs2019，vs2022，emscripten-3.1.28
| file | content | arch |
|---|---|---|
| ncnn-full-source.zip | 包含全部 submodule 代码的完整源码 | |
| ncnn-android.zip | android 静态库/动态库 | armeabi-v7a + arm64-v8a + x86 + x86_64 |
| ncnn-android-vulkan.zip | android 静态库/动态库，支持 GPU | armeabi-v7a + arm64-v8a + x86 + x86_64 |
| ncnn-apple.zip | apple xcframework，ios + ios-simulator + macos + mac-catalyst，with and w/o bitcode | armv7 + arm64 + arm64e + i386 + x86_64 |
| ncnn-apple-vulkan.zip | apple xcframework，ios + ios-simulator + macos + mac-catalyst，支持 GPU，with and w/o bitcode | arm64 + arm64e + x86_64 |
| ncnn-ios.zip | ios 静态库，with and w/o bitcode | armv7 + arm64 + arm64e |
| ncnn-ios-vulkan.zip | ios 静态库，支持 GPU，with and w/o bitcode | arm64 + arm64e |
| ncnn-ios-simulator.zip | ios simulator 静态库，with and w/o bitcode | i386 + x86_64 + arm64 |
| ncnn-ios-simulator-vulkan.zip | ios simulator 静态库，支持 GPU，with and w/o bitcode | x86_64 + arm64 |
| ncnn-macos.zip | macos 静态库 | x86_64 + arm64 |
| ncnn-macos-vulkan.zip | macos 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-mac-catalyst.zip | mac catalyst 静态库，with and w/o bitcode | x86_64 + arm64 |
| ncnn-mac-catalyst-vulkan.zip | mac catalyst 静态库，支持 GPU，with and w/o bitcode | x86_64 + arm64 |
| ncnn-ubuntu.zip | ubuntu linux 静态库/动态库，支持 GPU，模型转换工具 | x86_64 |
| ncnn-windows.zip | windows 静态库/动态库，支持 GPU，模型转换工具 | x86 + x64 + arm + arm64 |
| ncnn-webassembly.zip | webassembly 静态库 | wasm32 + simd + threads + simd-threads |
内建vulkan驱动加载功能，不依赖vulkan-sdk编译gpu功能，可直接加载显卡驱动文件
msvc编译启用arm neon指令加速，启用arm64 asimdhp编译
实现python pnnx pypi包和python调用接口/文档(@Hideousmon)
arm convolution int8 直接卷积重构支持任意elempack
优化 vulkan global pooling性能
优化resize bilinear性能
压缩字体数据减小二进制体积
deconvolution支持动态权重和对应pnnx转换
新增跑分数据rank card(@Qengineering)
支持big-endian架构平台，powerpc32位
添加woa linux ci
添加msvc禁用exceptions/rtti的编译开关
在macos上使用信号探测avx512指令集支持情况
支持寻找32位显卡驱动文件(@whyb)
启用benchmark编译打印4维shape(@Deepdive543443)
修复riscv-int8 sigmoid激活的测试失败问题(@MollySophia)
修复deconvolution x86 bias非对齐访问的问题
修复prelu x86 sse指令非对齐访问的问题(@aioa)
修复windows上openmp设置线程数为0的警告
修复在支持16bit/8bit的gpu上有关fp16sa shader使用fp16 shared变量的警告
修复nvidia vulkan驱动在程序退出的crash
修复vkimagemat from_android_hardware_buffer缺失的elemsize参数错误
修复simpleocv Mat模板ptr的偏移错误
添加更过的gpu相关python绑定接口(@joeyballentine)
android vulkan包的api版本降低到14/21
pnnx支持转换recompute_scale_factor=True的nn.Upsample
新增nn.Identity测试
修复pnnx路径切分的问题
修复pnnx生成ncnn py空格对齐(@cmdbug)
pnnx生成的py可以直接执行推理
python pnnx返回优化后的torch模型
删除无用的代码(@ningjiang233)
改善cmake toolchain文件(@zchrissirhcz)
新增watchos和tvos ci
修复linux sde ci的运行错误
更新POWER clang版本信息的文档(@JeremyRand)
更新有关vulkan/libomp-dev依赖的文档(@JeremyRand)
更新有关编译python模块CMAKE_TOOLCHAIN_FILE环境变量的文档(@JeremyRand)
修复Rasberry拼写错误(@JeremyRand)
FAQ新增有关pyncnn数据连续性的文档(@lll143653)
更新readme下载页表格
添加Nintendo 3DS编译信息(@Deepdive543443)
添加oncloud amlogic s805跑分数据(@mizu-bai)
添加树莓派5 gpu跑分数据(@FantasyGmm)
添加Jetson TX2跑分数据(@FantasyGmm)
添加8gen2跑分数据(@mahirumahiru)
添加2K2000跑分数据(@RevySR)
更新Jetson Orin Nano/树莓派5跑分数据(@Qengineering)
添加visionfive2跑分数据(@wzyforgit)
New Contributors
- @Deepdive543443 made their first contribution in #5116
- @ningjiang233 made their first contribution in #5139
- @FantasyGmm made their first contribution in #5152
- @mahirumahiru made their first contribution in #5180
- @Qengineering made their first contribution in #5216
- @lll143653 made their first contribution in #5220
- @joeyballentine made their first contribution in #5165
Full Changelog: 2023102...2024010
Contributors
Assets 42
android ios macos linux windows webassembly 预编译库 20231027 3116e02
编译版本，默认配置，android-ndk-r25c，xcode 13.4.1，ubuntu-20.04，ubuntu-22.04，vs2015，vs2017，vs2019，vs2022，emscripten-3.1.28
| file | content | arch |
|---|---|---|
| ncnn-full-source.zip | 包含全部 submodule 代码的完整源码 | |
| ncnn-android.zip | android 静态库/动态库 | armeabi-v7a + arm64-v8a + x86 + x86_64 |
| ncnn-android-vulkan.zip | android 静态库/动态库，支持 GPU | armeabi-v7a + arm64-v8a + x86 + x86_64 |
| ncnn-apple.zip | apple xcframework，ios + ios-simulator + macos + mac-catalyst，with and w/o bitcode | armv7 + arm64 + arm64e + i386 + x86_64 |
| ncnn-apple-vulkan.zip | apple xcframework，ios + ios-simulator + macos + mac-catalyst，支持 GPU，with and w/o bitcode | arm64 + arm64e + x86_64 |
| ncnn-ios.zip | ios 静态库，with and w/o bitcode | armv7 + arm64 + arm64e |
| ncnn-ios-vulkan.zip | ios 静态库，支持 GPU，with and w/o bitcode | arm64 + arm64e |
| ncnn-ios-simulator.zip | ios simulator 静态库，with and w/o bitcode | i386 + x86_64 + arm64 |
| ncnn-ios-simulator-vulkan.zip | ios simulator 静态库，支持 GPU，with and w/o bitcode | x86_64 + arm64 |
| ncnn-macos.zip | macos 静态库 | x86_64 + arm64 |
| ncnn-macos-vulkan.zip | macos 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-mac-catalyst.zip | mac catalyst 静态库，with and w/o bitcode | x86_64 + arm64 |
| ncnn-mac-catalyst-vulkan.zip | mac catalyst 静态库，支持 GPU，with and w/o bitcode | x86_64 + arm64 |
| ncnn-ubuntu.zip | ubuntu linux 静态库/动态库，支持 GPU，模型转换工具 | x86_64 |
| ncnn-windows.zip | windows 静态库/动态库，支持 GPU，模型转换工具 | x86 + x64 + arm + arm64 |
| ncnn-webassembly.zip | webassembly 静态库 | wasm32 + simd + threads + simd-threads |
x86 convolution int8 gemm重构支持任意elempack
x86 convolution int8 winograd重构支持任意elempack
arm convolution int8 gemm重构支持任意elempack
arm convolution int8 winograd重构支持任意elempack
gelu vulkan优化(@FhqTreap)
convolution1d vulkan优化(@FhqTreap)
gridsample x86优化(@Yoh-Z)
riscv gemm fp32优化(@Xinyu302)
新增erf/shrink和onnx转换(@brightening-eyes)
新增diag和pnnx转换(@wnqn1597)
新增celu和pnnx转换(@wnqn1597)
新增simplemath，允许不依赖libm编译使用数学函数(@HonestDeng)
pooling adaptive支持动态的输出尺寸和pnnx转换
elu selu支持4维输入输出
slice支持indices参数
memorydata支持tag参数和fp16存储
x86 selu shufflechannel优化(@wnqn1597)
修复convolution vulkan在固定shape时的结果错误
修复权重tag潜在的溢出(@lrw04)
按层加载模型减少内存占用(@daquexian)
修复老版本gcc编译avx2 gather的错误(@chainsx)
修复老版本gcc编译_mm256_set_m128的错误(@whyb)
修复新版本protobuf编译问题
修复老版本glibc round编译问题
修复c906工具链编译错误
pyncnn启用vulkan支持(@Hideousmon)
pyncnn添加load_param_mem接口(@JeremyRand @theflyingzamboni)
pnnx支持torch-2.1
pnnx消除moduleop的输出unpack
pnnx moduleop将权重shape作为参数写入param，内部权重顺序为使用顺序
pnnx改善reflect replicated pad匹配
pnnx合并conv3d-bn和deconv3d-bn
pnnx转换torch.narrow(@zyt1024)
pnnx转换torch.lgamma(@shudorcl)
pnnx转换torch.positive(@nicochen1118)
pnnx转换torch.cumprod(@Jiang-Weibo)
pnnx转换torch.mv/nn.ReplicationPad3d(@ShuRaymond)
pnnx转换F.pairwise_distance(@Marsyule)
pnnx转换torch.view_as_real/torch.view_as_complex(@Baiyuetribe)
修复pnnx与新版本protobuf编译问题(@HuPengsheet)
修复pnnx改变目录下划线的错误
onnx2ncnn支持celu转换(@brightening-eyes)
自动为pull request添加label
修复ohos工具链编译错误
改进codeformat脚本使用函数(@xiezheng-XD)
添加rk3566 rk3588s跑分数据(@chainsx)
添加Allwinner T527跑分数据(@YuzukiTsuru)
添加树莓派5b跑分数据(@Pillar1989)
添加RTX A3000跑分数据(@chainsx)
添加多款pc跑分数据(@whyb)
New Contributors
- @chainsx made their first contribution in #4959
- @xiezheng-XD made their first contribution in #4993
- @zyt1024 made their first contribution in #4918
- @FhqTreap made their first contribution in #5001
- @shudorcl made their first contribution in #4976
- @nicochen1118 made their first contribution in #4999
- @Jiang-Weibo made their first contribution in #5002
- @brightening-eyes made their first contribution in #5012
- @wnqn1597 made their first contribution in #4935
- @ShuRaymond made their first contribution in #4974
- @HuPengsheet made their first contribution in #5034
- @Marsyule made their first contribution in #4942
- @Pillar1989 made their first contribution in #5058
- @Xinyu302 made their first contribution in #4903
- @Hideousmon made their first contribution in #5020
- @HonestDeng made their first contribution in #4905
Full Changelog: 2023081...2023102
Contributors
Assets 38
android ios macos linux windows webassembly 预编译库 20230816 39721ee
编译版本，默认配置，android-ndk-r25c，xcode 13.4.1，ubuntu-20.04，ubuntu-22.04，vs2015，vs2017，vs2019，vs2022，emscripten-3.1.28
| file | content | arch |
|---|---|---|
| ncnn-full-source.zip | 包含全部 submodule 代码的完整源码 | |
| ncnn-android.zip | android 静态库/动态库 | armeabi-v7a + arm64-v8a + x86 + x86_64 |
| ncnn-android-vulkan.zip | android 静态库/动态库，支持 GPU | armeabi-v7a + arm64-v8a + x86 + x86_64 |
| ncnn-apple.zip | apple xcframework，ios + ios-simulator + macos + mac-catalyst，with and w/o bitcode | armv7 + arm64 + arm64e + i386 + x86_64 |
| ncnn-apple-vulkan.zip | apple xcframework，ios + ios-simulator + macos + mac-catalyst，支持 GPU，with and w/o bitcode | arm64 + arm64e + x86_64 |
| ncnn-ios.zip | ios 静态库，with and w/o bitcode | armv7 + arm64 + arm64e |
| ncnn-ios-vulkan.zip | ios 静态库，支持 GPU，with and w/o bitcode | arm64 + arm64e |
| ncnn-ios-simulator.zip | ios simulator 静态库，with and w/o bitcode | i386 + x86_64 + arm64 |
| ncnn-ios-simulator-vulkan.zip | ios simulator 静态库，支持 GPU，with and w/o bitcode | x86_64 + arm64 |
| ncnn-macos.zip | macos 静态库 | x86_64 + arm64 |
| ncnn-macos-vulkan.zip | macos 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-mac-catalyst.zip | mac catalyst 静态库，with and w/o bitcode | x86_64 + arm64 |
| ncnn-mac-catalyst-vulkan.zip | mac catalyst 静态库，支持 GPU，with and w/o bitcode | x86_64 + arm64 |
| ncnn-ubuntu.zip | ubuntu linux 静态库/动态库，支持 GPU，模型转换工具 | x86_64 |
| ncnn-windows.zip | windows 静态库/动态库，支持 GPU，模型转换工具 | x86 + x64 + arm + arm64 |
| ncnn-webassembly.zip | webassembly 静态库 | wasm32 + simd + threads + simd-threads |
实现全部的binaryop explicit广播规则类型
x86直接卷积权重变换的avx2/avx512优化
x86 int8直接卷积支持任意elempack和sse2/xop/avx2/avx512/vnni优化
ppc64 power8/power9 vsx工具链支持，编译器检查和intrinsic翻译优化(@JeremyRand)
更新glslang并启用VK_KHR_cooperative_matrix扩展和优化
修复pyncnn自定义layer模型权重加载
c_api新增Mat border/layer_to_index api(@Mek101)
VkCompute::submit_and_wait现在能返回错误值(@Upliner)
修复老版本clang编译时too many microtasks问题
修复clang-cl cpuid函数兼容性(@charlescao460)
修复新版本protobuf c++17编译问题
修复老版本编辑器sleep递归调用错误(@whyb)
编译时检查loongarch lasx扩展支持并自动启用
清理multiheadattention arm优化代码
binaryop支持一维outer axis广播规则，保持旧的兼容行为
benchncnn支持从命令行参数中指定自定义模型和输入(@tpoisonooo)
macos平台静态编译链接需要的系统库(@Baiyuetribe)
更改amd集显上的显存分配策略为仅设备优先，修复在bios设置大显存时分配失败问题
onnx2ncnn遇到不支持transpose类型输出错误信息(@huoshuai-dot)
pnnx支持多算子到多算子的图变换
pnnx新增转换torch.round/trunc/fill/index_put/to/type_as/topk/fmod/cross/t/maximum/minimum
pnnx合并chinese-clip/sam-iamge-encoder attention结构
pnnx合并F.scaled_dot_product_attention
pnnx消除无用的expand/expand_as/type_as
pnnx修正fp16模型在优化时的权重变换错误
pnnx修正负数shape索引越界问题(@Justin62628)
pnnx修复转换后py文件执行时权限错误(@zhenjiaguo)
pnnx转换ncnn global pooling后自动添加reshape
pnnx转换非zero padding模式的卷积到ncnn
pnnx转换2维nn.Linear为ncnn gemm
pnnx转换torch.stack为ncnn concat+reshape
pnnx转换torch.t到ncnn permute(@XiaBing992)
pnnx转换logsigmoid/log_softmax为ncnn sigmoid/softmax+log(@lrw04)
pnnx修复slice_copy输出的类型信息
pnnx修复表达式中int64转换溢出问题
pnnx修复reshape表达式消除后的ghost结点
pnnx合并表达式时折叠shape为1的类似标量的权重
pnnx合并表达式支持max/min
pnnx改善图中有inplace操作时的输出结点连接探测，带来更多的常量折叠
添加ncnn glsl扩展文档以及中文版(@whyb)
修正faq文档错误(@KYShek)
改进cmake寻找vulkan提示用语(@zchrissirhcz)
更新vs2017编译步骤的细节(@brcarry)
新增Intel oneAPI编译步骤(@mizu-bai)
更新loongarch ci工具链，添加loongarch lsx覆盖率
更新python ci版本，新增python-3.12包
更新rpi3b+/rpi4b测试数据
更新huawei kunpeng 920测试数据(@MobtgZhang)
新增3A6000和TH1520 gpu测试数据(@Rabenda)
新增RDK X3 Module测试数据(@LJoson)
修复ios模拟器gpu badge(@732857315)
New Contributors
- @charlescao460 made their first contribution in #4738
- @Justin62628 made their first contribution in #4765
- @zhenjiaguo made their first contribution in #4801
- @732857315 made their first contribution in #4836
- @KYShek made their first contribution in #4837
- @JeremyRand made their first contribution in #4807
- @Upliner made their first contribution in #4828
- @Mek101 made their first contribution in #4855
- @brcarry made their first contribution in #4872
- @Rabenda made their first contribution in #4894
- @Baiyuetribe made their first contribution in #4859
- @XiaBing992 made their first contribution in #4940
- @lrw04 made their first contribution in #4925
Full Changelog: 2023051...2023081
Contributors
Assets 38
android ios macos linux windows webassembly 预编译库 20230517 903ec7c
编译版本，默认配置，android-ndk-r25c，xcode 13.4.1，ubuntu-20.04，ubuntu-22.04，vs2015，vs2017，vs2019，vs2022，emscripten-3.1.28
| file | content | arch |
|---|---|---|
| ncnn-full-source.zip | 包含全部 submodule 代码的完整源码 | |
| ncnn-android.zip | android 静态库/动态库 | armeabi-v7a + arm64-v8a + x86 + x86_64 |
| ncnn-android-vulkan.zip | android 静态库/动态库，支持 GPU | armeabi-v7a + arm64-v8a + x86 + x86_64 |
| ncnn-apple.zip | apple xcframework，ios + ios-simulator + macos + mac-catalyst，with and w/o bitcode | armv7 + arm64 + arm64e + i386 + x86_64 |
| ncnn-apple-vulkan.zip | apple xcframework，ios + ios-simulator + macos + mac-catalyst，支持 GPU，with and w/o bitcode | arm64 + arm64e + x86_64 |
| ncnn-ios.zip | ios 静态库，with and w/o bitcode | armv7 + arm64 + arm64e |
| ncnn-ios-vulkan.zip | ios 静态库，支持 GPU，with and w/o bitcode | arm64 + arm64e |
| ncnn-ios-simulator.zip | ios simulator 静态库，with and w/o bitcode | i386 + x86_64 + arm64 |
| ncnn-ios-simulator-vulkan.zip | ios simulator 静态库，支持 GPU，with and w/o bitcode | x86_64 + arm64 |
| ncnn-macos.zip | macos 静态库 | x86_64 + arm64 |
| ncnn-macos-vulkan.zip | macos 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-mac-catalyst.zip | mac catalyst 静态库，with and w/o bitcode | x86_64 + arm64 |
| ncnn-mac-catalyst-vulkan.zip | mac catalyst 静态库，支持 GPU，with and w/o bitcode | x86_64 + arm64 |
| ncnn-ubuntu.zip | ubuntu linux 静态库/动态库，支持 GPU，模型转换工具 | x86_64 |
| ncnn-windows.zip | windows 静态库/动态库，支持 GPU，模型转换工具 | x86 + x64 + arm + arm64 |
| ncnn-webassembly.zip | webassembly 静态库 | wasm32 + simd + threads + simd-threads |
arm convolution winograd重构支持任意elempack
arm convolution sgemm重构支持任意elempack
arm convolution直接卷积重构支持任意elempack
arm deconvolution/matmul 调用 gemm 完成计算
arm softmax支持任意elempack和bf16/fp16优化
arm multiheadattention fp16sa softmax优化
arm/x86 convolution1d直接卷积重构支持任意elempack和优化
粗糙的vulkan gemm和multiheadattention优化
multiheadattention支持输入attention mask
sigmoid/swish/clip/gelu/mish/tanh支持4d输入
减少double类型的使用(@zhiliu6)
arm a53/a55架构检测和流水线优化
允许注册自定义层替代内置实现
x86 asin/acos/atan/atan2 sse2/avx/avx512优化(@MouriNaruto)
sse_mathfunc迁移floor/ceil(@Yoh-Z)
x86 mathfun迁移abs(@Yoh-Z)
simpleocv新增cv::imdecode内存加载图片(@AlOa)
新增配合vulkan vma使用的三种扩展支持(@whyb)
新增获取vkinstance的接口(@whyb)
新增通用的sleep接口(@whyb)
innerproduct允许2维高度1的输入输出
修复multiheadattention分配内存存在的多线程竞争问题
修复在获取不到cache信息时的除0错误
修复scale avx512计算错误
修复exynos9810非法指令错误
老旧adreno驱动中禁用fp16a以解决计算错误
绕过n卡padding shader编译错误
移除platform.h中无用的aarch64判断(@dreamcmi)
修正modelwriter squeeze层参数id错误(@irexyc)
修复gcc-13编译错误(@hillwoodroc)
修复gcc-5.2 aarch64编译错误
修复aosp编译错误(@caofx0418)
修复n卡上benchmark退出时的crash(@triple-Mu)
修复获取cpu cache信息潜在的fd泄漏
优化lightmode循环条件(@MambaWong)
绕过新版moltenvk的兼容性问题
绕过n卡在multiheadattention softmax结果偶发nan的兼容性问题
调用cpu.h接口时强制初始化全局cpu信息
pnnx支持torch-2.0
pnnx支持complex数据类型
pnnx转换torch.baddbmm/torch.mm/torch.stft家族/torch.std/F.scaled_dot_product_attention
pnnx支持fp16权重的torchscript
pnnx支持非forward的其他函数入口
pnnx当只有一个动态维时候折叠reshape的shape表达式
pnnx识别常数常量和表达式中的折叠
pnnx自动删除maxpool无indices输出项
pnnx总是删除convtransposed output_size参数
pnnx合并gelu表达式
pnnx合并vit/clip/diffusers attention
修正pnnx的RNN/GRU省略输出项的python代码生成
修正pnnx转换ir时潜在的负INT_MAX下溢问题
修正pnnx fprintf类型不匹配(@kernelbin)
修复pnnx windows编译错误(@Yoh-Z)
pyncnn model zoo添加yolov7-tiny(@kennybradley)
pyncnn model zoo添加yolov8s(@triple-Mu)
macos pypi包使用完整版本号
改善wasm ci编译效率
更新ci swiftshader版本
更新cmake ios toolchain，新增ios-simulator arm64和mac catalyst ci
添加qnx toolchain和编译步骤(@zchrissirhcz)
删除ubuntu-18.04的ci
更新3A5000 benchmark数据(@wzyforgit)
新增2K1000LA benchmark数据(@lrzlin)
新增icpc icc benchmark数据(@mizu-bai)
新增Hyper-V Linux Guest benchmark数据(@MouriNaruto)
新增和更新op4lts/op5/VF2/FT2000/3A4000 benchmark数据(@MobtgZhang)
更新centos编译文档(@inisis)
更新windows msvc编译文档(@kernelbin)
faq新增关于cmake版本升级的内容(@inisis)
faq新增关于显卡节能模式的内容(@whyb)
修正citation和benchmark文档中的拼写错误(@zchrissirhcz)
修正pnnx代码和readme中的拼写错误(@jsyzdej @zchrissirhcz)
New Contributors
- @whyb made their first contribution in #4614
- @AlOa made their first contribution in #4557
- @caofx0418 made their first contribution in #4622
- @dreamcmi made their first contribution in #4628
- @kernelbin made their first contribution in #4647
- @lrzlin made their first contribution in #4656
- @irexyc made their first contribution in #4658
- @jsyzdej made their first contribution in #4665
- @hillwoodroc made their first contribution in #4684
- @MobtgZhang made their first contribution in #4687
- @kennybradley made their first contribution in #4693
Full Changelog: 2023022...2023051
Contributors
Assets 38
android ios macos linux windows webassembly 预编译库 20230223 ff80ac2
编译版本，默认配置，android-ndk-r25c，xcode 14.0.1，ubuntu-18.04，ubuntu-20.04，ubuntu-22.04，vs2015，vs2017，vs2019，vs2022，emscripten-3.1.28
| file | content | arch |
|---|---|---|
| ncnn-full-source.zip | 包含全部 submodule 代码的完整源码 | |
| ncnn-android.zip | android 静态库/动态库 | armeabi-v7a + arm64-v8a + x86 + x86_64 |
| ncnn-android-vulkan.zip | android 静态库/动态库，支持 GPU | armeabi-v7a + arm64-v8a + x86 + x86_64 |
| ncnn-ios.zip | ios 静态库，with and w/o bitcode | armv7 + arm64 + arm64e + i386 + x86_64 |
| ncnn-ios-vulkan.zip | ios 静态库，支持 GPU，with and w/o bitcode | arm64 + arm64e + x86_64 |
| ncnn-macos.zip | macos 静态库 | x86_64 + arm64 |
| ncnn-macos-vulkan.zip | macos 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-ubuntu.zip | ubuntu linux 静态库/动态库，支持 GPU，模型转换工具 | x86_64 |
| ncnn-windows.zip | windows 静态库/动态库，支持 GPU，模型转换工具 | x86 + x64 + arm + arm64 |
| ncnn-webassembly.zip | webassembly 静态库 | wasm32 + simd + threads + simd-threads |
扩充binaryop broadcast规则
新增copyto算子，对应于torch inplace slice copy操作
x86 gemm优化，新增 transpose_output 参数
x86 multiheadattention优化
x86 groupnorm优化(@EdVince)
arm gemm优化，包括fp16s/fp16sa
arm gelu优化(@EdVince)
arm multiheadattention优化(@EdVince)
新增获取cpu l2/l3 cache大小接口，通过sysconf/win32-api和linux sysfs
x86 gemm 依据l2 cache分块的优化
x86 convolution/deconvolution/deformableconv2d/matmul 调用 gemm 完成计算
x86 convolution winograd重构支持任意elempack
x86 convolution直接卷积重构支持任意elempack
x86公共的bfloat转换函数
slice/eltwise/concat支持4d输入
c api新增获取output indexes names接口
改善vulkan winograd f43 fp16计算数值稳定性
修复gpu信息bug bliz初始化问题(@weirdseed)
修正arm bfloat2float和float2bfloat命名相反的问题
更新riscv winograd f32系数，修复一些警告
更好的riscv rvv tanh实现
为ncnnoptimize/ncnn2int8添加新加的算子和参数
修复musl libc编译问题
更新stb image和image write，启用arm neon优化
更新emsdk版本到3.1.28，开启SIMPLEOCV(@ncnnnnn)
pnnx新增torch.cumsum转换(@csukuangfj)
pnnx新增torch.atan2/log10转换
pnnx自动替换pow(x,2)为square(x)
修正pnnx windows slice end参数问题(@Yoh-Z)
pnnx自动删除无用的Tensor.clone(@Yoh-Z)
pnnx自动展开模型输入tuple和list类型
pnnx转ncnn时分析binaryop broadcast规则并插入适当的reshape
pnnx折叠常数常量，修复常数转换MemoryData兼容性问题
pnnx合并pixel unshuffle(@Yoh-Z)
去除pnnx readme多余空行(@inisis)
去除pnnx无用的include(@XiangYyang)
修正pyncnn output_indexes接口错误(@wyushun)
修复最新macos vulkan sdk兼容性问题(@w1ndseeker)
删除python代码无用的import(@dianjiaogit)
修复macos ci的xcode版本和vulkan sdk安装问题
更新ci中已废弃的create release步骤
添加CITATION.cff(@tpoisonooo)
更新cpu benchmark数据(@wzyforgit)
修复README编译状态badge(@tpoisonooo)
修复README编译链接(@tuduweb)
修正拼写错误(@hwdef @hiteshhedwig)
添加ncnn-fortran例子(@mizu-bai)
添加sherpa-ncnn实时语音识别例子(@csukuangfj)
New Contributors
- @mizu-bai made their first contribution in #4423
- @inisis made their first contribution in #4428
- @dianjiaogit made their first contribution in #4378
- @wyushun made their first contribution in #4453
- @w1ndseeker made their first contribution in #4472
- @hiteshhedwig made their first contribution in #4486
- @weirdseed made their first contribution in #4493
- @XiangYyang made their first contribution in #4497
- @tuduweb made their first contribution in #4530
Full Changelog: 2022112...2023022
Contributors
Assets 28
android ios macos linux windows webassembly 预编译库 20221128 03550ba
编译版本，默认配置，android-ndk-r25b，xcode 12.4，ubuntu-18.04，ubuntu-20.04，ubuntu-22.04，vs2015，vs2017，vs2019，vs2022，emscripten-2.0.8
| file | content | arch |
|---|---|---|
| ncnn-full-source.zip | 包含全部 submodule 代码的完整源码 | |
| ncnn-android.zip | android 静态库/动态库 | armeabi-v7a + arm64-v8a + x86 + x86_64 |
| ncnn-android-vulkan.zip | android 静态库/动态库，支持 GPU | armeabi-v7a + arm64-v8a + x86 + x86_64 |
| ncnn-ios.zip | ios 静态库，with and w/o bitcode | armv7 + arm64 + arm64e + i386 + x86_64 |
| ncnn-ios-vulkan.zip | ios 静态库，支持 GPU，with and w/o bitcode | arm64 + arm64e + x86_64 |
| ncnn-macos.zip | macos 静态库 | x86_64 + arm64 |
| ncnn-macos-vulkan.zip | macos 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-ubuntu.zip | ubuntu linux 静态库/动态库，支持 GPU，模型转换工具 | x86_64 |
| ncnn-windows.zip | windows 静态库/动态库，支持 GPU，模型转换工具 | x86 + x64 + arm + arm64 |
| ncnn-webassembly.zip | webassembly 静态库 | wasm32 + simd + threads + simd-threads |
新增loongarch64 lsx向量指令集优化，包括absval/batchnorm/bias/binaryop/cast/clip/concat/convolution1d/convolutiondepthwise/convolution/crop/deconvolutiondepthwise/deconvolution/dequantize/dropout/eltwise/flatten/hardsigmoid/hardswish/innerproduct/interp/mish/packing/padding/pooling/prelu/quantize/relu/requantize/sigmoid/slice/softmax/swish/tanh/unaryop算子(@junchao-loongson)
layernorm x86优化(@LinHeLurking @LRY89757)
batchnorm/elu/prelu/gelu x86优化(@LRY89757)
softmax arm neon优化(@luqiang-guo)
batchnorm/instancenorm riscv vector优化(@thelastlin)
deformableconv2d x86优化(@miemie2013)
elu vulkan优化(@Yoh-Z)
convolution int8 x86 sse2/avx2优化
更新riscv vector segment load/store(@thelastlin)
改善内存池回收机制(@LinHeLurking)
新增获取cpu物理核心数量api，默认线程数设为物理大核心数量
实现控制单层运算特性是否启用的参数
更通用的macos/ios cpu特性探测过程，a15/a16/m2启用bf16和i8mm指令集
统一innerproduct x86 fp32/fp16s内核代码
修复在android省电模式cpu离线导致openmp崩溃的问题
实现glu算子与对应的pnnx转换(@csukuangfj)
新增fold和unfold算子
新增gridsample算子与对应的pnnx转换(@LRY89757)
lstm支持proj_size参数
groupnorm支持1d/2d/4d输入计算
squeeze/expanddims支持4d输入输出
multiheadattention支持kdim vdim参数
修复convolutiondepthwise allocator的错误设置(@w8501)
修正windows arm环境中convolution权重为空的问题
修复onnx2ncnn blob名字超出255长度的问题(@ZhangGe6)
修正expanddims axes参数id错误的问题(@LiuYi-Up)
修正c api allocator无法工作的问题(@qiqikit)
更严格的编译器armv7 fp16功能检查和兼容
修复老版本gcc编译avx512代码的编译错误(@bestpower)
修复windows-arm64编译(@zchrissirhcz)
修复在老版本ndk引用ncnn链接atomic内置函数失败的问题
修复新版本pybind11编译错误(@tpoisonooo)
python模块支持mat.numpy()(@csukuangfj)
更新pybind11和glslang子模块
pyncnn发布python 3.11包和windows arm版本
pnnx支持pytorch 1.13
pnnx现已支持在cpu上加载gpu导出的torchscript
pnnx保存onnx-zero模型文件
pnnx转换时将常量存储在临时文件减少内存占用
pnnx新增命令行参数fp16=0/1控制是否用fp16保存onnx-zero/ncnn模型
pnnx支持大部分数学函数转换，新增nn.Softmax2d/nn.Fold/nn.Unfold/F.fold/F.unfold/bitwise_left_shift/bitwise_right_shift转换
pnnx改善和匹配inplace slice copy操作
融合更多静态的F.convND/F.linear为nn module
合并临接的reshape
合并pad到conv中
改善pnnx F.softmax转换对dtype兼容性(@EdVince)
修正pnnx softmax/normalize/slice负数axis转换错误的问题
修正pnnx slice end下标错误问题
修正pnnx转ncnn保存fp16权重没考虑对齐的问题
pnnx遇到动态size时不再折叠为常量
pnnx自动折叠new_full/full_like
yolov5示例支持yolov5 6.2(@shaoshengsong)
修复编译警告(@tpoisonooo @veahow)
删除无用空行(@MollySophia @Menci)
修正空格对齐(@tonori)
修正拼写错误(@LRY89757 @Zepan @eltociear)
忽略.xmake目录，CMakeSettings.json，Visual Studio CMake文件(@zchrissirhcz)
重构README(@septs)
改善README布局(@magicse)
添加一些示例项目链接(@magicse @shaoshengsong)
faq新增有关禁用fp16设置的内容(@MisakaBit)
更新riscv rvv ci
新增c906 ci
新增loongarch64 lsx ci
迁移部分github action ci到腾讯ci
新增TH1520 cmake toolchain(@luyanaa)
切分大型单元测试加快多进程测试速度
新增Intel Celeron M 420跑分(@MouriNaruto)
新增T-Head TH1520跑分(@YuzukiTsuru)
新增rock5b rk3588跑分(@hwdef)
New Contributors
- @LinHeLurking made their first contribution in #4065
- @septs made their first contribution in #4114
- @w8501 made their first contribution in #4173
- @MollySophia made their first contribution in #4187
- @Menci made their first contribution in #4188
- @magicse made their first contribution in #4193
- @tonori made their first contribution in #4217
- @YuzukiTsuru made their first contribution in #4240
- @ZhangGe6 made their first contribution in #4236
- @MisakaBit made their first contribution in #4248
- @LiuYi-Up made their first contribution in #4259
- @veahow made their first contribution in #4274
- @csukuangfj made their first contribution in #4283
- @Zepan made their first contribution in #4287
- @bestpower made their first contribution in #4294
- @shaoshengsong made their first contribution in #4328
- @junchao-loongson made their first contribution in #4242
- @eltociear made their first contribution in #4358
Full Changelog: 2022072...2022112
Contributors
Assets 28
android ios macos linux windows webassembly 预编译库 20220729 b4ba207
编译版本，默认配置，android-ndk-r24，xcode 12.4，ubuntu-18.04，ubuntu-20.04，ubuntu-22.04，vs2015，vs2017，vs2019，vs2022，emscripten-2.0.8
| file | content | arch |
|---|---|---|
| ncnn-full-source.zip | 包含全部 submodule 代码的完整源码 | |
| ncnn-android.zip | android 静态库/动态库 | armeabi-v7a + arm64-v8a + x86 + x86_64 |
| ncnn-android-vulkan.zip | android 静态库/动态库，支持 GPU | armeabi-v7a + arm64-v8a + x86 + x86_64 |
| ncnn-ios.zip | ios 静态库，with and w/o bitcode | armv7 + arm64 + arm64e + i386 + x86_64 |
| ncnn-ios-vulkan.zip | ios 静态库，支持 GPU，with and w/o bitcode | arm64 + arm64e + x86_64 |
| ncnn-macos.zip | macos 静态库 | x86_64 + arm64 |
| ncnn-macos-vulkan.zip | macos 静态库，支持 GPU | x86_64 + arm64 |
| ncnn-ubuntu.zip | ubuntu linux 静态库/动态库，支持 GPU，模型转换工具 | x86_64 |
| ncnn-windows.zip | windows 静态库/动态库，支持 GPU，模型转换工具 | x86 + x64 + arm + arm64 |
| ncnn-webassembly.zip | webassembly 静态库 | wasm32 + simd + threads + simd-threads |
batchnorm avx512 优化(@LRY89757)
新增DeformableConv2d层和单元测试(@miemie2013)
修复conv3x3 winograd tensorcore权重数据错乱导致结果出错的问题
修复memorydata 4维数据转换的问题
pnnx转换torchvision.ops.DeformConv2d到ncnn
pnnx自动删除无用的 mul + torch.ones 和 add + torch.zeros
pnnx修复动态shape时删除无用pad可能的崩溃问题
pnnx修复动态shape时错误删除upsample的问题
添加sse优化文档(@DC-Zhou)
更加严格的编译器riscv vector支持检查，删除rvv-0.7.1编译支持
更新ci中android ndk路径，使用android-ndk-r24打包
New Contributors
- @DC-Zhou made their first contribution in #4053
- @miemie2013 made their first contribution in #4070
Full Changelog: 2022072...2022072
