JavaCPP Presets 1.5.11

November 16, 2024 version 1.5.11

• Enable distributed package using Gloo in presets for PyTorch (pull #1510)
• Add presets for the CUPTI module of CUDA (pull #1531)
• Add new ClangMemoryMgmtExample in samples for LLVM (pull #1522)
• Enable opencv_python3 module for macosx-arm64 as well (pull #1517)
• Introduce macosx-arm64 builds for CPython (pull #1511), NumPy (pull #1515), SciPy (pull #1516)
• Update and fix the sample code of the presets for LLVM (pull #1501)
• Fix Vulkan GPU acceleration for FFmpeg (pull #1497)
• Build FFmpeg with zimg to enable zscale filter (pull #1481)
• Enable PulseAudio support for FFmpeg on Linux (pull #1472)
• Virtualize btCollisionWorld, btOverlapFilterCallback, btOverlapCallback from Bullet Physics SDK (pull #1475)
• Upgrade presets for OpenCV 4.10.0, FFmpeg 7.1, Spinnaker 4.0.0.116 (pull #1524), MKL 2025.0, DNNL 3.6.1, OpenBLAS 0.3.28, CMINPACK 1.3.11, GSL 2.8, CPython 3.13.0, NumPy 2.1.3, SciPy 1.14.1, LLVM 19.1.3, LibRaw 0.21.2 (pull #1520), Leptonica 1.85.0, Tesseract 5.5.0, libffi 3.4.6, CUDA 12.6.2, cuDNN 9.5.1, NCCL 2.23.4, nvCOMP 4.1.0.6, OpenCL 3.0.17, NVIDIA Video Codec SDK 12.2.72, PyTorch 2.5.1 (pull #1466), SentencePiece 0.2.0, TensorFlow Lite 2.18.0, TensorRT 10.6.0.26, Triton Inference Server 2.51.0, ONNX 1.17.0, ONNX Runtime 1.20.0, TVM 0.18.0, and their dependencies

JavaCPP Presets 1.5.10

January 29, 2024 version 1.5.10

• Introduce macosx-arm64 builds for PyTorch (pull #1463)
• Reenable linux-arm64 builds for CPython and NumPy (pull #1386)
• Enable Vulkan GPU acceleration for FFmpeg (pull #1460)
• Include timeapi.h for system API of Windows (pull #1447)
• Add Android and Windows builds to presets for DepthAI (pull #1441)
• Add presets for nvCOMP 3.0.5 (pull #1434), SentencePiece 0.1.99 (pull #1384)
• Refactor and improve presets for PyTorch (pull #1360)
• Include mkl_lapack.h header file in presets for MKL (issue #1388)
• Map new higher-level C++ API of Triton Inference Server (pull #1361)
• Upgrade presets for OpenCV 4.9.0, FFmpeg 6.1.1, HDF5 1.14.3, MKL 2024.0, DNNL 3.3.4, OpenBLAS 0.3.26, ARPACK-NG 3.9.1, CPython 3.12.1, NumPy 1.26.3, SciPy 1.12.0, LLVM 17.0.6, Leptonica 1.84.1, Tesseract 5.3.4, CUDA 12.3.2, cuDNN 8.9.7, NCCL 2.19.3, OpenCL 3.0.15, PyTorch 2.1.2 (pull #1426), TensorFlow Lite 2.15.0, Triton Inference Server 2.41.0, DepthAI 2.24.0, ONNX 1.15.0, ONNX Runtime 1.16.3, TVM 0.14.0, and their dependencies

JavaCPP Presets 1.5.9

June 6, 2023 version 1.5.9

• Virtualize nvinfer1::IGpuAllocator from TensorRT to allow customization (pull #1367)
• Add new SampleJpegEncoder code for nvJPEG module of CUDA (pull #1365)
• Map std::vector of CameraParams, ImageFeatures, and MatchesInfo from cv::detail (issue bytedeco/javacv#2027)
• Fix H.264 decoder of FFmpeg by increasing MAX_SLICES to 256 (pull #1349)
• Link FFmpeg with latest version of VA-API libraries (pull #1296)
• Build HDF5 with support for SZIP enabled (pull #1334)
• Map missing functions from mkl_trans.h in presets for MKL (issue #1331)
• Bundle the official Java API of HDF5 via the hdf5_java library (pull #1327)
• Map missing cblas_?axpby() functions in presets for MKL (issue #1326)
• Prefix with "fisheye" all functions from the cv::fisheye:: namespace to avoid collisions (pull #1324)
• Remove mapping for platform-dependent enum values in presets for libffi (pull #1318)
• Fix mapping of cv::fisheye::calibrate() function from opencv_calib3d (issue #1185)
• Add an RPATH to the tesseract program to avoid loading issues (issue #1314)
• Bundle the CUPTI module of CUDA in the presets for PyTorch (pull #1307)
• Build FFmpeg with AOMedia AV1 and SVT-AV1 codecs (pull #1303)
• Map c10::OptionalArrayRef<int64_t> from PyTorch to long... as well for convenience (issue #1300)
• Remove mapping for c10::OptionalArrayRef to simplify calls in presets for PyTorch (issue #1300)
• Virtualize btMotionState and btDefaultMotionState from Bullet Physics SDK to allow callbacks (pull #1297)
• Define STRING_BYTES_CHARSET to "UTF-8" for FFmpeg since it appears to assume that (issue bytedeco/javacv#1945)
• Map at::ITensorListRef as used by at::cat() in presets for PyTorch (issue #1293)
• Map torch::data::datasets::ChunkDataReader and related data loading classes from PyTorch (issue #1215)
• Add missing predefined AVChannelLayout in presets for FFmpeg (issue #1286)
• Map c10::impl::GenericDict as returned by c10::IValue::toGenericDict() in presets for PyTorch
• Introduce linux-armhf and linux-x86 builds to presets for TensorFlow Lite (pull #1268)
• Add presets for LibRaw 0.21.1 (pull #1211)
• Upgrade presets for OpenCV 4.7.0, FFmpeg 6.0 (issue bytedeco/javacv#1693), HDF5 1.14.1, Hyperscan 5.4.2 (issue #1308), Spinnaker 3.0.0.118 (pull #1313), librealsense2 2.53.1 (pull #1305), MKL 2023.1, DNNL 3.1, OpenBLAS 0.3.23, ARPACK-NG 3.9.0, CPython 3.11.3, NumPy 1.24.3, SciPy 1.10.1, LLVM 16.0.4, Leptonica 1.83.0, Tesseract 5.3.1, CUDA 12.1.1, cuDNN 8.9.1, NCCL 2.18.1, OpenCL 3.0.14, NVIDIA Video Codec SDK 12.1.14, PyTorch 2.0.1, TensorFlow Lite 2.12.0, TensorRT 8.6.1.6, Triton Inference Server 2.33.0, DepthAI 2.21.2, ONNX 1.14.0, ONNX Runtime 1.15.0, TVM 0.12.0, Bullet Physics SDK 3.25, and their dependencies

JavaCPP Presets 1.5.8

November 2, 2022 version 1.5.8

• Fix mapping of torch::ExpandingArrayWithOptionalElem in presets for PyTorch (issue #1250)
• Disable OpenMP for Tesseract to work around performance issues (pull #1201)
• Enable NVIDIA GPU acceleration for FFmpeg on ARM and POWER as well (issue bytedeco/javacv#1894)
• Add support for HEVC and Opus to FLV format in presets for FFmpeg (pull #1228)
• Build Leptonica with OpenJPEG for JPEG 2000 support
• Introduce linux-arm64 and macosx-arm64 builds to presets for libpostal (pull #1199 and pull #1205)
• Map missing factory functions in torch:: namespace using torch_ as prefix in presets for PyTorch (issue #1197)
• Add presets for the nvJPEG module of CUDA (issue #1193)
• Introduce Android builds for TensorFlow Lite (discussion #1180)
• Map std::vector<cv::Ptr<cv::mcc::CChecker> > for CCheckerDetector.getListColorChecker() (issue bytedeco/javacpp#571)
• Include missing opencv2/mcc/ccm.hpp header file in presets for OpenCV (discussion bytedeco/javacpp#568)
• Fix a few incorrectly mapped instances of std::unordered_map for PyTorch (issue #1164)
• Migrate builds for Leptonica and Tesseract to CMake (issue #1163)
• Introduce macosx-arm64 builds for LZ4 (pull #1243), libffi (issue #1182), Leptonica, and Tesseract (issue #814)
• Map instances of torch::OrderedDict using C++ templates from PyTorch (issue #623)
• Add presets for Bullet Physics SDK 3.24 (pull #1153)
• Add long[] pytorch.Tensor.shape() method for convenience (pull #1161)
• Enable DNNL codegen as BYOC backend in presets for TVM
• Allow passing raw pointer as deleter to from_blob(), etc functions of PyTorch (discussion #1160)
• Include cudnn_backend.h header file in presets for CUDA (issue #1158)
• Bundle zlibwapi.dll required by cuDNN in redist artifacts of presets for CUDA (issue bytedeco/javacv#1767)
• Harmonize string and buffer pointer types of function parameters from DepthAI (issue #1155)
• Bundle correctly OpenMP library for PyTorch builds on Mac as well (issue #1225)
• Remove dependency on CUDA from presets for Triton Inference Server (pull #1151)
• Disable signal handlers of DepthAI known to cause issues with the JDK (issue #1118)
• Upgrade presets for OpenCV 4.6.0, FFmpeg 5.1.2, HDF5 1.12.2, LZ4 1.9.4, MKL 2022.2, DNNL 2.7.1, OpenBLAS 0.3.21 (issue #1171), CPython 3.10.8, NumPy 1.23.4, SciPy 1.9.3, Gym 0.26.2, LLVM 15.0.3, libffi 3.4.4, Tesseract 5.2.0, CUDA 11.8.0, cuDNN 8.6.0, NCCL 2.15.5, OpenCL 3.0.12, MXNet 1.9.1, PyTorch 1.12.1, TensorFlow Lite 2.10.0, TensorRT 8.4.3.1, Triton Inference Server 2.26.0, ALE 0.8.0, DepthAI 2.18.0, ONNX 1.12.0, ONNX Runtime 1.13.1, TVM 0.10.0, Skia 2.88.3, cpu_features 0.7.0, ModSecurity 3.0.8, and their dependencies

JavaCPP Presets 1.5.7

February 11, 2022 version 1.5.7

• Build FFmpeg with WebP encoding support (pull #1133)
• Include sys/ipc.h and sys/shm.h for system APIs of Linux and Mac OS X (pull #1132)
• Map c10::ArrayRef<at::Tensor>(std::vector<at::Tensor>&) constructor from PyTorch for convenience (discussion #1128)
• Add long rs2_get_frame_data_address() to reduce garbage for real-time applications using librealsense2 (discussion bytedeco/javacpp#532)
• Add to torch.Tensor convenient create(), createBuffer(), and createIndexer() factory methods for PyTorch
• Upgrade requirements to Android 7.0 for camera support in OpenCV and FFmpeg (issue bytedeco/javacv#1692)
• Include new llvm-c/Transforms/PassBuilder.h header file in presets for LLVM (pull #1093)
• Introduce macosx-arm64 builds to presets for OpenCV, FFmpeg, OpenBLAS (issue #1069), LLVM (pull #1092)
• Add presets for LZ4 1.9.3 (pull #1094), Triton Inference Server 2.18.0 (pull #1085)
• Add presets for the NvToolsExt (NVTX) module of CUDA (issue #1068)
• Increase the amount of function pointers available for callbacks in presets for Qt (pull #1080)
• Map C++ JIT classes and functions of TorchScript in presets for PyTorch (issue #1068)
• Synchronize cachePackage() and prevent repeated package caching in all presets (pull #1071)
• Build FFmpeg with VA-API enabled and bundle its libraries to avoid loading issues (issue bytedeco/javacv#1188)
• Upgrade presets for OpenCV 4.5.5, FFmpeg 5.0 (pull #1125), librealsense2 2.50.0, Arrow 6.0.1, MKL 2022.0, DNNL 2.5.2, OpenBLAS 0.3.19, FFTW 3.3.10, CPython 3.10.2, NumPy 1.22.2, SciPy 1.8.0, Gym 0.21.0, LLVM 13.0.1, libpostal 1.1, Leptonica 1.82.0, Tesseract 5.0.1, CUDA 11.6.0, cuDNN 8.3.2, NCCL 2.11.4, MXNet 1.9.0, PyTorch 1.10.2, TensorFlow Lite 2.8.0, TensorRT 8.2.3.0, ALE 0.7.3, DepthAI 2.14.1, ONNX 1.10.2, ONNX Runtime 1.10.0, TVM 0.8.0, ModSecurity 3.0.6, and their dependencies

JavaCPP Presets 1.5.6

August 2, 2021 version 1.5.6

• Change opencv_core.Mat constructors to create column vectors out of arrays for consistency (issue #1064)
• Add presets for the new barcode and wechat_qrcode modules of OpenCV
• Work around loading issues with execution providers in presets for ONNX Runtime
• Annotate the presets for LLVM with @NoException to reduce unneeded C++ overhead (pull #1052)
• Update samples for LLVM 12 including new samples/llvm/OrcJit.java using libffi (pull #1050)
• Enable GTK support in presets for OpenCV when building on ARM as well
• Correct enum classes in presets for Spinnaker (pull #1048)
• Add Windows build for ONNX (issue #983)
• Add linux-arm64 builds to presets for DNNL, OpenCL, TensorRT (pull #1044), and ONNX Runtime
• Build FFmpeg with libxml2, enabling support for DASH demuxing (pull #1033), and libsrt for SRT protocol support (pull #1036)
• Add @MemberGetter for av_log_default_callback() in presets for FFmpeg (issue #812)
• Include cudaGL.h and cuda_gl_interop.h header files in presets for CUDA (pull #1027)
• Add presets for libffi 3.4.2 (issue #833), NVIDIA Video Codec SDK 11.1.5 (pull #1020), PyTorch 1.9.0 (issue #623), TensorFlow Lite 2.5.0, DepthAI 2.8.0, ModSecurity 3.0.5 (pull #1012)
• Map std::vector<cv::Range> to RangeVector in opencv_core.Mat for convenience (issue bytedeco/javacv#1607)
• Include genericaliasobject.h, context.h, tracemalloc.h, and datetime.h for CPython (issue #1017)
• Add samples using LLVM modules to deal with bitcode and object files (pull #1016)
• Upgrade presets for OpenCV 4.5.3, FFmpeg 4.4 (pull #1030), Spinnaker 2.4.0.143 (pull #1040), librealsense2 2.44.0 (pull #1031), Arrow 4.0.1, HDF5 1.12.1, MKL 2021.3, DNNL 2.3, OpenBLAS 0.3.17, GSL 2.7, CPython 3.9.6, NumPy 1.21.1, SciPy 1.7.0, Gym 0.18.3, LLVM 12.0.1 (pull #1065), Leptonica 1.81.1, CUDA 11.4.0, cuDNN 8.2.2, NCCL 2.10.3, TensorRT 8.0.1.6, ONNX 1.9.0, ONNX Runtime 1.8.1, Skia 2.80.3, and their dependencies

JavaCPP Presets 1.5.5

March 8, 2021 version 1.5.5

• Bundle LLD executable in presets for LLVM as required by TVM on Windows
• Prevent public static final objects from getting deallocated by PointerScope (issue bytedeco/javacv#1599)
• Fix compatibility of Leptonica with JavaFX by upgrading to libpng 1.6.37 (pull #1007)
• Introduce linux-arm64 build for CUDA, cuDNN, and NCCL (issue #735)
• Add new array constructors to opencv_core.Mat that copy data for convenience (pull #1002)
• Rebase PrimitiveScalar on PrimitiveScalarBase in presets for Arrow for easy access to data() (issue #998)
• Add NamedMetadataOperations.h implementing data retrieval operations for LLVM nodes (pull #995)
• Enable OpenMP for ONNX Runtime on Mac once again (issue #917)
• Build OpenCV without OpenBLAS when environment variable NOOPENBLAS=yes (pull #987)
• Enable OpenCL acceleration for DNNL (issue #938)
• Introduce monkey patching when loading presets for CPython to relocate home more reliably
• Add display sample for librealsense2 (pull #978)
• Fix builds for libpostal on Mac and Windows (issue #903)
• Fix builds for NumPy and SciPy on Linux when using a cross compiler or not
• Update presets for Visual Studio 2019 on Windows
• Add presets for OpenCL 3.0, TVM 0.7.0 and bundle its official Java API (TVM4J) via the jnitvm_runtime library
• Include free() in presets for FTTW as required by fftw_export_wisdom_to_string() (issue bytedeco/javacpp#429)
• Include all missing header files from the opencv_ximgproc module (issue #958)
• Disable assembly optimizations for libx264 with FFmpeg on Mac to work around crashes (issue bytedeco/javacv#1519)
• Add linux-armhf and linux-arm64 builds for librealsense and librealsense2 (pull #951)
• License default builds of FFmpeg under LGPL v3 and move GPL-enabled builds to -gpl extension (pull #950)
• Upgrade presets for OpenCV 4.5.1, FFmpeg 4.3.2, Arrow 3.0.0, Hyperscan 5.4.0, MKL 2021.1, OpenBLAS 0.3.13, ARPACK-NG 3.8.0, CMINPACK 1.3.8, FFTW 3.3.9, librealsense2 2.40.0 (pull #946), DNNL 2.1.1, CPython 3.9.2, NumPy 1.20.1, SciPy 1.6.1, Gym 0.18.0, LLVM 11.1.0 (pull #1001), OpenPose 1.7.0, CUDA 11.2.1, cuDNN 8.1.1, NCCL 2.8.4, MXNet 1.8.0, TensorFlow 1.15.5, TensorRT 7.2.3.4, ONNX 1.8.1, ONNX Runtime 1.7.0, Qt 5.15.2, Skia 2.80.2, cpu_features 0.6.0, and their dependencies

JavaCPP Presets 1.5.4

September 9, 2020 version 1.5.4

• Bundle libpostal_data program, executable via Loader.load() for convenience (issue #939)
• Enable all stable target architectures in the presets for LLVM (pull #937)
• Virtualize QObject and its subclasses from Qt to allow customization (issue bytedeco/javacpp#419)
• Bundle programs from Clang and LLVM, executable via Loader.load() for convenience (issue #833)
• Include nnvm/c_api.h header file in presets for MXNet (issue #912)
• Enable OpenMP for DNNL on Mac using same library name as MKL to prevent conflicts (issue #907)
• Fix loading issue with opencv_ximgproc (issue #911)
• Build LibTIFF after WebP to make sure they link correctly in presets for Leptonica
• Virtualize IInt8Calibrator plus subclasses from TensorRT to allow customization (issue #902)
• Replace requires with requires static in JPMS .platform modules (pull #900)
• Add presets for OpenPose 1.6.0 (pull #898)
• Add comparison against MKL in llvm/samples/polly/MatMulBenchmark.java
• Add requires org.bytedeco.javacpp.${javacpp.platform.module} to load jnijavacpp with JPMS (pull #893)
• Bundle configuration files required by AOT compilation with GraalVM (issue eclipse/deeplearning4j#7362)
• Add support for Windows to presets for Qt (issue #862)
• Fix JPMS modules for CUDA, ARPACK-NG, GSL, SciPy, Gym, MXNet (pull #880 and pull #881), OpenCV, CPython, LLVM, Tesseract, Qt (pull #928)
• Build OpenBLAS with a TARGET even for DYNAMIC_ARCH to avoid SIGILL (issue eclipse/deeplearning4j#8747)
• Upgrade presets for OpenCV 4.4.0, FFmpeg 4.3.1 (pull #891), Arrow 1.0.1, Hyperscan 5.3.0, MKL 2020.3, MKL-DNN 0.21.5, DNNL 1.6.2, OpenBLAS 0.3.10, CPython 3.7.9, NumPy 1.19.1, SciPy 1.5.2, Gym 0.17.2, LLVM 10.0.1, Leptonica 1.80.0, CUDA 11.0.3, cuDNN 8.0.3, NCCL 2.7.8, MXNet 1.7.0, TensorFlow 1.15.3, TensorRT 7.1, ONNX 1.7.0 (pull #882), ONNX Runtime 1.4.0 (pull #887), Qt 5.15.0, Skia 2.80.1, and their dependencies
• Add FullOptimization.h allowing users to fully optimize LLVM modules (pull #869)

JavaCPP Presets 1.5.3

April 14, 2020 version 1.5.3

• Add presets for the new intensity_transform and rapid modules of OpenCV
• Add support for Polly optimizer to presets for LLVM (pull #864)
• Fix loading issue with opencv_dnn_superres (issue bytedeco/javacv#1396)
• Add support for Windows to presets for TensorRT (pull #860)
• Add dependency on presets for jnijavacpp and javacpp-platform artifact to fix issues at load time (issue bytedeco/javacv#1305)
• Bundle the official Java API of ONNX Runtime via the jnionnxruntime library
• Add CUDA-enabled build for ONNX Runtime via -gpu extension
• Fix presets for LLVM 9.0 where libclang would fail to load on Windows (issue #830)
• Add Windows build for ONNX Runtime, map the C++ API, and refine support for DNNL (pull #841)
• Add convenient Py_AddPath() helper method to presets for CPython
• Include OrcBindings.h and other missing header files for LLVM (issue #833)
• Fix -platform artifacts on JPMS by commenting out requires to Android modules (issue #814 and pull #831)
• Include timecode.h, among other missing header files, in the avutil module of FFmpeg (issue #822)
• Map a few more inherited constructors missing from the presets for MKL-DNN and DNNL
• Make sure clone() actually returns new PIX, FPIX, or DPIX objects with presets for Leptonica
• Add opencv_python3 module and corresponding loader class with sample code to the presets for OpenCV
• Bundle OpenSSL in the presets for CPython for consistency across platforms (issue #796)
• Add presets for Arrow 0.16.0, SciPy 1.4.1 (issue #747), Gym 0.17.1, Hyperscan 5.2.1 (pull #849)
• Upgrade presets for OpenCV 4.3.0, FFmpeg 4.2.2, Spinnaker 1.27.0.48, HDF5 1.12.0, MKL 2020.1, MKL-DNN 0.21.4, DNNL 1.3, OpenBLAS 0.3.9, CPython 3.7.7, NumPy 1.18.2, LLVM 10.0.0, CUDA 10.2, cuDNN 7.6.5, NCCL 2.6.4, MXNet 1.6.0, TensorFlow 1.15.2, TensorRT 7.0, ALE 0.6.1, Leptonica 1.79.0, Tesseract 4.1.1, ONNX Runtime 1.2.0, Qt 5.14.2, Skia 1.68.1, and their dependencies

JavaCPP Presets 1.5.2

November 5, 2019 version 1.5.2

• Add presets for the cudacodec, cudafeatures2d, cudastereo, and cudabgsegm modules of OpenCV (issue #806)
• Fix mapping of warpAffine and warpPerspective from opencv_cudawarping (issue #806)
• Add linux-armhf and linux-arm64 builds for HDF5 (issue #794)
• Add build for Mac OS X to presets for nGraph (issue #799)
• Update presets for Visual Studio 2017 on Windows
• Bundle the opencv_annotation, opencv_interactive-calibration, opencv_version, opencv_visualisation, and tesseract programs
• Add linux-armhf, linux-arm64, linux-ppc64le, and windows-x86 builds for CPython, NumPy, and LLVM (pull #768)
• Include audio_ops.h, list_ops.h, lookup_ops.h, and manip_ops.h for TensorFlow
• Add necessary platform properties to build -gpu extensions on linux-arm64 and linux-ppc64le (issue #769)
• Add packages missing from TensorFlow (issue #773)
• Fix JPMS module names for OpenBLAS and Tesseract (issue #772)
• Include env.h, kernels.h, and ops.h to allow creating custom operations using the C API of TensorFlow
• Add profiles to parent pom.xml to detect host and use its artifacts, for example: mvn -Djavacpp.platform.custom -Djavacpp.platform.host ...
• Add -platform-gpu, -platform-python, -platform-python-gpu, and -platform-redist artifacts for convenience
• Add presets for librealsense2 2.29.0, DNNL 1.1, ONNX Runtime 0.5.0
• Upgrade presets for OpenCV 4.1.2, FFmpeg 4.2.1, librealsense 1.12.4, MKL 2019.5, MKL-DNN 0.21.2, OpenBLAS 0.3.7, GSL 2.6, CPython 3.7.5, NumPy 1.17.3, LLVM 9.0.0, CUDA 10.1 Update 2, cuDNN 7.6.2, NCCL 2.4.8, MXNet 1.5.1, TensorFlow 1.15.0, TensorRT 6.0, ONNX 1.6.0 (pull #795), nGraph 0.26.0, Qt 5.13.1, cpu_features 0.4.1, and their dependencies